Skip to main content

DESINFORMACION Y AGRESIVIDAD EN SOCIAL MEDIA: AGREGANDO INFORMACION Y ANALIZANDO EL LENGUAJE

PGC2018-096212-B-C31

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal de Generación de Conocimiento y Fortalecimiento Científico y Tecnológico del Sistema de I+D+i
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D (Excelencia / Generación de Conocimiento)
Call year 2018
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POLITÈCNICA DE VALÈNCIA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 51
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

MISMIS: Misinformation and Miscommunication in social media: aggregating information and analysing language, MISMIS: Desinformación y agresividad en los medios de comunicación social: agregando información y analizando el lenguaje

  • Rosso, Paolo|||0000-0002-8922-1242
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
  • Gonzalo, Julio
  • Plaza, Laura
  • Carrillo, J.
  • Amigó, E.
  • Verdejo, M. Felisa
  • Taulé, Mariona
  • Salamó, María
  • Martí, M. Antònia
[EN] The general objectives of the project are to address and monitor misinformation (biased and fake news) and miscommunication (aggressive language and hate speech) in social media, as well as to establish a high quality methodological standard for the whole research community (i) by developing rich annotated datasets, a data repository and online evaluation services; (ii) by proposing suitable evaluation metrics; and (iii) by organizing evaluation campaigns to foster research on the above issues., [ES] Los objetivos generales del proyecto son abordar y monitorizar la desinformación (noticias sesgadas y falsas) y la mala comunicación (lenguaje agresivo y mensajes de odio) en los medios de comunicación social, así como establecer un estándar metodológico de calidad para toda la comunidad investigadora mediante: i) el desarrollo de datasets anotados, un repositorio de datos y servicios de evaluación online; ii) la propuesta de métricas de evaluación adecuadas; y iii) la organización de campañas de evaluación para fomentar la investigación sobre las cuestiones mencionadas., The MISMIS project (PGC2018-096212-B) is funded by the Spanish Ministry of Science, Innovation and Universities.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Offensive Language Recognition in Social Media

  • Shushkevich, Elena
  • Cardiff, John
  • Akhtyamova, Liliya
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] This article proposes an approach to solving the problem of multiclassification within the framework of aggressive language recognition in Twitter. At the stage of preprocessing external data is added to the existing dataset, which is based on information in the links in dataset. This made it possible to expand the training dataset and thereby to improve the quality of the classification. The model created is an ensemble of classical machine learning models included Logistic Regression, Support Vector Machines, Naive Bayes models and a combination of Logistic Regression and Naive Bayes. The obtained value of macro F1-score for one of the experiments achieved 0.61, which exceeds the state-of-art published value by 1 percentage point. This indicates the potential value of the proposed approach in the field of hate speech recognition in social media., The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project MISMISFAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Tesis doctoral (DoctoralThesis)

Some Contributions to Interactive Machine Translation and to the Applications of Machine Translation for Historical Documents

  • Domingo-Ballester, Miguel|||0000-0002-7910-4536
[ES] Los documentos históricos son una parte importante de nuestra herencia cultural. Sin embargo, debido a la barrera idiomática inherente en el lenguaje humano y a las propiedades lingüísticas de estos documentos, su accesibilidad está principalmente restringida a los académicos. Por un lado, el lenguaje humano evoluciona con el paso del tiempo. Por otro lado, las convenciones ortográficas no se crearon hasta hace poco y, por tanto, la ortografía cambia según el período temporal y el autor. Por estas razones, el trabajo de los académicos es necesario para que los no expertos puedan obtener una comprensión básica de un documento determinado. En esta tesis abordamos dos tareas relacionadas con el procesamiento de documentos históricos. La primera tarea es la modernización del lenguaje que, a fin de hacer que los documentos históricos estén más accesibles para los no expertos, tiene como objetivo reescribir un documento utilizando la versión moderna del idioma original del documento. La segunda tarea es la normalización ortográfica. Las propiedades lingüísticas de los documentos históricos mencionadas con anterioridad suponen un desafío adicional para la aplicación efectiva del procesado del lenguaje natural en estos documentos. Por lo tanto, esta tarea tiene como objetivo adaptar la ortografía de un documento a los estándares modernos a fin de lograr una consistencia ortográfica. Ambas tareas las afrontamos desde una perspectiva de traducción automática, considerando el idioma original de un documento como el idioma fuente, y su homólogo moderno/normalizado como el idioma objetivo. Proponemos varios enfoques basados en la traducción automática estadística y neuronal, y llevamos a cabo una amplia experimentación que ratifica el potencial de nuestras contribuciones -en donde los enfoques estadísticos arrojan resultados iguales o mejores que los enfoques neuronales para la mayoría de los casos-. En el caso de la tarea de modernización del lenguaje, esta experimentación incluye una evaluación humana realizada con la ayuda de académicos y un estudio con usuarios que verifica que nuestras propuestas pueden ayudar a los no expertos a obtener una comprensión básica de un documento histórico sin la intervención de un académico. Como ocurre con cualquier problema de traducción automática, nuestras aplicaciones no están libres de errores. Por lo tanto, para obtener modernizaciones/normalizaciones perfectas, un académico debe supervisar y corregir los errores. Este es un procedimiento común en la industria de la traducción. La metodología de traducción automática interactiva tiene como objetivo reducir el esfuerzo necesario para obtener traducciones de alta calidad uniendo al agente humano y al sistema de traducción en un proceso de corrección cooperativo. Sin embargo,la mayoría de los protocolos interactivos siguen una estrategia de izquierda a derecha. En esta tesis desarrollamos un nuevo protocolo interactivo que rompe con esta barrera de izquierda a derecha. Hemos evaluado este nuevo protocolo en un entorno de traducción automática, obteniendo grandes reducciones del esfuerzo humano. Finalmente, dado que este marco interactivo es de aplicación general a cualquier problema de traducción, lo hemos aplicado -nuestro nuevo protocolo junto con uno de los protocolos clásicos de izquierda a derecha- a la modernización del lenguaje y a la normalización ortográfica. Al igual que en traducción automática, el marco interactivo logra disminuir el esfuerzo requerido para corregir los resultados de un sistema automático., [CA] Els documents històrics són una part important de la nostra herència cultural. No obstant això, degut a la barrera idiomàtica inherent en el llenguatge humà i a les propietats lingüístiques d'aquests documents, la seua accessibilitat està principalment restringida als acadèmics. D'una banda, el llenguatge humà evoluciona amb el pas del temps. D'altra banda, les convencions ortogràfiques no es van crear fins fa poc i, per tant, l'ortografia canvia segons el període temporal i l'autor. Per aquestes raons, el treball dels acadèmics és necessari perquè els no experts puguen obtindre una comprensió bàsica d'un document determinat. En aquesta tesi abordem dues tasques relacionades amb el processament de documents històrics. La primera tasca és la modernització del llenguatge que, a fi de fer que els documents històrics estiguen més accessibles per als no experts, té per objectiu reescriure un document utilitzant la versió moderna de l'idioma original del document. La segona tasca és la normalització ortogràfica. Les propietats lingüístiques dels documents històrics mencionades amb anterioritat suposen un desafiament addicional per a l'aplicació efectiva del processat del llenguatge natural en aquests documents. Per tant, aquesta tasca té per objectiu adaptar l'ortografia d'un document als estàndards moderns a fi d'aconseguir una consistència ortogràfica. Dues tasques les afrontem des d'una perspectiva de traducció automàtica, considerant l'idioma original d'un document com a l'idioma font, i el seu homòleg modern/normalitzat com a l'idioma objectiu. Proposem diversos enfocaments basats en la traducció automàtica estadística i neuronal, i portem a terme una àmplia experimentació que ratifica el potencial de les nostres contribucions -on els enfocaments estadístics obtenen resultats iguals o millors que els enfocaments neuronals per a la majoria dels casos-. En el cas de la tasca de modernització del llenguatge, aquesta experimentació inclou una avaluació humana realitzada amb l'ajuda d'acadèmics i un estudi amb usuaris que verifica que les nostres propostes poden ajudar als no experts a obtindre una comprensió bàsica d'un document històric sense la intervenció d'un acadèmic. Com ocurreix amb qualsevol problema de traducció automàtica, les nostres aplicacions no estan lliures d'errades. Per tant, per obtindre modernitzacions/normalitzacions perfectes, un acadèmic ha de supervisar i corregir les errades. Aquest és un procediment comú en la indústria de la traducció. La metodologia de traducció automàtica interactiva té per objectiu reduir l'esforç necessari per obtindre traduccions d'alta qualitat unint a l'agent humà i al sistema de traducció en un procés de correcció cooperatiu. Tot i això, la majoria dels protocols interactius segueixen una estratègia d'esquerra a dreta. En aquesta tesi desenvolupem un nou protocol interactiu que trenca amb aquesta barrera d'esquerra a dreta. Hem avaluat aquest nou protocol en un entorn de traducció automàtica, obtenint grans reduccions de l'esforç humà. Finalment, atès que aquest marc interactiu és d'aplicació general a qualsevol problema de traducció, l'hem aplicat -el nostre nou protocol junt amb un dels protocols clàssics d'esquerra a dreta- a la modernització del llenguatge i a la normalitzaciò ortogràfica. De la mateixa manera que en traducció automàtica, el marc interactiu aconsegueix disminuir l'esforç requerit per corregir els resultats d'un sistema automàtic., [EN] Historical documents are an important part of our cultural heritage. However,due to the language barrier inherent in human language and the linguistic properties of these documents, their accessibility is mostly limited to scholars. On the one hand, human language evolves with the passage of time. On the other hand, spelling conventions were not created until recently and, thus, orthography changes depending on the time period and author. For these reasons, the work of scholars is needed for non-experts to gain a basic understanding of a given document. In this thesis, we tackle two tasks related with the processing of historical documents. The first task is language modernization which, in order to make historical documents more accessible to non-experts, aims to rewrite a document using the modern version of the document's original language. The second task is spelling normalization. The aforementioned linguistic properties of historical documents suppose an additional challenge for the effective natural language processing of these documents. Thus, this task aims to adapt a document's spelling to modern standards in order to achieve an orthography consistency. We affront both task from a machine translation perspective, considering a document's original language as the source language, and its modern/normalized counterpart as the target language. We propose several approaches based on statistical and neural machine translation, and carry out a wide experimentation that shows the potential of our contributions¿with the statistical approaches yielding equal or better results than the neural approaches in most of the cases. For the language modernization task, this experimentation includes a human evaluation conducted with the help of scholars and a user study that verifies that our proposals are able to help non-experts to gain a basic understanding of a historical document without the intervention of a scholar. As with any machine translation problem, our applications are not error-free. Thus, to obtain perfect modernizations/normalizations, a scholar needs to supervise and correct the errors. This is a common procedure in the translation industry. The interactive machine translation framework aims to reduce the effort needed for obtaining high quality translations by embedding the human agent and the translation system into a cooperative correction process. However, most interactive protocols follow a left-to-right strategy. In this thesis, we developed a new interactive protocol that breaks this left-to-right barrier. We evaluated this new protocol in a machine translation environment, obtaining large reductions of the human effort. Finally, since this interactive framework is of general application to any translation problem, we applied it¿our new protocol together with one of the classic left-to-right protocols¿to language modernization and spelling normalization. As with machine translation, the interactive framework diminished the effort required for correcting the outputs of an automatic system., The research leading to this thesis has been partially funded by Ministerio de Economía y Competitividad (MINECO) under projects SmartWays (grant agreement RTC-2014-1466-4), CoMUN-HaT (grant agreement TIN2015-70924-C2-1-R) and MISMISFAKEnHATE (grant agreement PGC2018-096212-B-C31); Generalitat Valenciana under projects ALMAMATER (grant agreement PROMETEOII/2014/030) and DeepPattern (grant agreement PROMETEO/2019/121); the European Union through Programa Operativo del Fondo Europeo de Desarrollo Regional (FEDER) from Comunitat Valenciana (2014–2020) under project Sistemas de frabricación inteligentes para la indústria 4.0 (grant agreement ID-IFEDER/2018/025); and the PRHLT research center under the research line Machine Learning Applications.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

New Avenues in Computational Irony Detection in Social Media, Nuevos Enfoques para la Detección Computacional de la Ironía en Redes Sociales

  • Ortega-Bueno, Reynier
[ES] Esta tesis aborda la detección de ironía como un desafío lingüístico, computacional y social multifacético, considerando la variación multilingüe, la multimodalidad y el sesgo en los corpus de entrenamiento. La tesis propone una arquitectura LSTM con mecanismos de atención que integra rasgos lingüísticos y representaciones profundas para detectar la ironía y la sátira en Español. Introduce un nuevo modelo que combina transformers textuales y visuales para el análisis multimodal. Además, analiza el sesgo en los corpus de ironía, mostrando su efecto negativo en la generalización y mostrando mejoras tras su identificación y mitigacion., This work was mainly supported by Fairness and Transparency for equitable NLP applications in social media, funded by MCIN/AEI/10.13039/501100011033 and MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31) .
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Overview of PAN 2021: Authorship Verification, Profiling Hate Speech Spreaders on Twitter, and Style Change Detection.

  • Bevendorff, Janek
  • Chulvi-Ferriols, María Alberta
  • Peña-Sarracén, Gretel Liz de la
  • Kestemont, Mike
  • Manjavacas, Enrique
  • Markov, Ilia
  • Mayerl, Maximilian
  • Potthast, Martin
  • Rangel, Francisco
  • Stamatatos, Efstathios
  • Stein, Benno
  • Wiegmann, Matti
  • Wolska, Magdalena
  • Zangerle, Eva
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] The paper gives a brief overview of the three shared tasks to be organized at the PAN 2021 lab on digital text forensics and stylometry hosted at the CLEF conference. The tasks include authorship verification across domains, author profiling for hate speech spreaders, and style change detection for multi-author documents. In part the tasks are new and in part they continue and advance past shared tasks, with the overall goal of advancing the state of the art, providing for an objective evaluation on newly developed benchmark datasets., The work of the researchers from Universitat Politecnica de Valencia was partially funded by the Spanish MICINN under the project MISMISFAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31), and by the Generalitat Valenciana under the project DeepPattern (PROMETEO/2019/121).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Scalable and Language-Independent Embedding-based Approach for Plagiarism Detection Considering Obfuscation Type: No Training Phase

  • Gharavi, Erfaneh
  • Veisi, Hadi
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] The efficiency and scalability of plagiarism detection systems have become a major challenge due to the vast amount of available textual data in several languages over the Internet. Plagiarism occurs in different levels of obfuscation, ranging from the exact copy of original materials to text summarization. Consequently, designed algorithms to detect plagiarism should be robust to the diverse languages and different type of obfuscation in plagiarism cases. In this paper, we employ text embedding vectors to compare similarity among documents to detect plagiarism. Word vectors are combined by a simple aggregation function to represent a text document. This representation comprises semantic and syntactic information of the text and leads to efficient text alignment among suspicious and original documents. By comparing representations of sentences in source and suspicious documents, pair sentences with the highest similarity are considered as the candidates or seeds of plagiarism cases. To filter and merge these seeds, a set of parameters, including Jaccard similarity and merging threshold, are tuned by two different approaches: offline tuning and online tuning. The offline method, which is used as the benchmark, regulates a unique set of parameters for all types of plagiarism by several trials on the training corpus. Experiments show improvements in performance by considering obfuscation type during threshold tuning. In this regard, our proposed online approach uses two statistical methods to filter outlier candidates automatically by their scale of obfuscation. By employing the online tuning approach, no distinct training dataset is required to train the system. We applied our proposed method on available datasets in English, Persian and Arabic languages on the text alignment task to evaluate the robustness of the proposed methods from the language perspective as well. As our experimental results confirm, our efficient approach can achieve considerable performance on the different datasets in various languages. Our online threshold tuning approach without any training datasets works as well as, or even in some cases better than, the training-base method., The work of Paolo Rosso was partially funded by the Spanish MICINN under the research Project MISMIS-FAKEn-HATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

A Twitter Political Corpus of the 2019 10N Spanish Election

  • Sánchez-Junquera, Javier
  • Ponzetto, Simone Paolo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] We present a corpus of Spanish tweets of 15 Twitter accounts of politicians of the main five parties (PSOE, PP, Cs, UP and VOX) covering the campaign of the Spanish election of 10th November 2019 (10N Spanish Election). We perform a semi-automatic annotation of domainspecific topics using a mixture of keyword-based and supervised techniques. In this preliminary study we extracted the tweets of few politicians of each party with the aim to analyse their official communication strategy. Moreover, we analyse sentiments and emotions employed in the tweets. Although the limited size of the Twitter corpus due to the very short time span, we hope to provide with some first insights on the communication dynamics of social network accounts of these five Spanish political parties., The work of the authors from the Universitat Politecnica de Valencia was funded by the Spanish MICINN under the research project MISMISFAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

FacTweet: Profiling Fake News Twitter Accounts

  • Ghanem, Bilal Hisham Hasan
  • Ponzetto, Simone Paolo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] We present an approach to detect fake news in Twitter at the account level using a neural recurrent model and a variety of different semantic and stylistic features. Our method extracts a set of features from the timelines of news Twitter accounts by reading their posts as chunks, rather than dealing with each tweet independently. We show the experimental benefits of modeling latent stylistic signatures of mixed fake and real news with a sequential model over a wide range of strong baselines, The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Multimodal Multi-image Fake News Detection

  • Giachanou, Anastasia
  • Zhang, Guobiao
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Recent years have seen a large increase in the amount of false information that is posted online. Fake news are created and propagated in order to deceive users and manipulate opinions and subsequently have a negative impact on the society. The automatic detection of fake news is very challenging since some of those news are created in sophisticated ways containing text or images that have been deliberately modified. Combining information from different modalities can be very useful for determining which of the online articles are fake. In this paper, we propose a multimodal multi-image system that combines information from different modalities in order to detect fake news posted online. In particular, our system combines textual, visual and semantic information. For the textual representation we use the Bidirectional Encoder Representations from Transformers (BERT) to better capture the underlying semantic and contextual meaning of the text. For the visual representation we extract image tags from multiple images that the articles contain using the VGG-16 model. The semantic representation refers to the text-image similarity calculated using the cosine similarity between the title and image tags embeddings. Our experimental results on a real world dataset show that combining features from the different modalities is effective for fake news detection. In particular, our multimodal multi-image system significantly outperforms the BERT baseline by 4.19% and SpotFake by 5.39%., Anastasia Giachanou is supported by the SNSF Early Postdoc Mobility grant under the project Early Fake News Detection on Social Media, Switzerland (P2TIP2 181441). Guobiao Zhang is funded by China Scholarship Council (CSC) from the Ministry of Education of P.R. China. The work of Paolo Rosso is partially funded by the Spanish MICINN under the research project MISMISFAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018- 096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Minería de argumentación en el Referéndum del 1 de Octubre de 2017, Argument Mining in the October 1, 2017 Referendum

  • Esteve, Marcos
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
  • Rosso, Paolo|||0000-0002-8922-1242
[ES] La minería de argumentación permite, mediante herramientas software, obtener cuáles son los argumentos que expresan los autores en un determinado texto. En este artículo se pretende realizar un análisis de la argumentación expresada por los usuarios en Twitter en relación al referendum del 1 de octubre de 2017. Se utilizar á para ello, el dataset MultiStanceCat proporcionado en la tarea organizada en el IberEval 2018. Dado que las herramientas de minería de argumentación trabajan en su mayoría en inglés, será necesario construir un sistema de traducción neuronal con postedición que permita realizar una traducción de los tweets del español y el catalán al inglés. Los resultados al realizar la minería de argumentación sobre los tweets traducidos ha demostrado obtener un porcentaje muy reducido de argumentación en todas las comunidades., [EN] Argument mining allows, through software tools, to obtain which are the arguments expressed by the authors in a given text. This article aims to make an analysis of the arguments expressed by users on Twitter in relation to the referendum of October 1, 2017 using the MultiStanceCat dataset provided in the shared task organized at IberEval 2018. Since the tools of argumentation mining work mostly in English, it was necessary to build a neural translation system with post-editing that allows to translate of tweets from Spanish and Catalan to English. The results of argumentation mining on the translated tweets have shown to obtain a minimum percentage of argumentation in all the communities., Dicho trabajo ha sido desarrollado en el marco de la asignatura Traducción Automática del Máster Universitario en Inteligencia Artificial, Reconocimiento de Formas e Imagen Digital (MIARFID) de la Universitat Politécnica de València. El trabajo de los ultimos dos autores se ha desarrollado en el marco del proyecto Misinformation and Miscommunication in social media: FAKE news and HATE speech (MISMIS-FAKEnHATE) del Ministerio de Ciencia, Innovación y Universidades (PGC2018-096212-B-C31) y del proyecto PROMETEO/2019/121 (DeepPattern) de la Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Detection of conspiracy propagators using psycho-linguistic characteristics

  • Giachanou, Anastasia
  • Ghanem, Bilal
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] The rise of social media has offered a fast and easy way for the propagation of conspiracy theories and other types of disinformation. Despite the research attention that has received, fake news detection remains an open problem and users keep sharing articles that contain false statements but which they consider real. In this article, we focus on the role of users in the propagation of conspiracy theories that is a specific type of disinformation. First, we compare profile and psycho-linguistic patterns of online users that tend to propagate posts that support conspiracy theories and of those who propagate posts that refute them. To this end, we perform a comparative analysis over various profile, psychological and linguistic characteristics using social media texts of users that share posts about conspiracy theories. Then, we compare the effectiveness of those characteristics for predicting whether a user is a conspiracy propagator or not. In addition, we propose ConspiDetector, a model that is based on a convolutional neural network (CNN) and which combines word embeddings with psycho-linguistic characteristics extracted from the tweets of users to detect conspiracy propagators. The results show that ConspiDetector can improve the performance in detecting conspiracy propagators by 8.82% compared with the CNN baseline with regard to F1-metric., The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The work of the first author was supported by the SNSF Early Postdoc Mobility grant P2TIP2_181441 under the project Early Fake News Detection on Social Media, Switzerland. The work of the third author was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in Social Media: Fake News and Hate Speech (PGC2018096212-B-C31) and by the European Cooperation in Science and Technology under the COST Action 17124 DigForAsp.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Tesis doctoral (DoctoralThesis)

New Avenues in Computational Irony Detection in Social Media

  • Ortega Bueno, Reynier
[ES] La presente tesis doctoral, presentada como compendio de artículos científicos, aborda el complejo problema de la detección de ironía. Esta figura del lenguaje es un fenómeno figurativo y pragmático que elude el análisis lingüístico del significado literal debido a su dependencia del contexto y la intencionalidad. La dificultad del problema se intensifica en contextos de redes sociales, donde la ironía se manifiesta en dimensiones multilingües y multimodales, complicando su modelación computacional. En este sentido, esta investigación está motivada en dos direcciones: primero, las aplicaciones potenciales de la detección de ironía más allá del análisis de sentimientos, especialmente en la mitigación de problemas de comunicación y desinformación; segundo, las limitaciones existentes de los métodos automáticos para la detección eficaz de la ironía. Esta tesis explora tres direcciones principales. Inicialmente, se analiza la ironía y la sátira a través de variantes del español, introduciendo un nuevo método que integra conocimiento lingüístico y representaciones profundas dentro de un modelo recurrente neuronal dotado con mecanismos de atención, superando el rendimiento del estado del arte. Posteriormente, se extiende la investigación al escenario multimodal, proponiendo una nueva arquitectura que combina transformers textuales con transformers visuales, mejorando la precisión en la detección de ironía en contenidos de redes sociales basados en texto e imagen. Finalmente, se investiga el sesgo temático en corpus de ironía y su impacto en la generalización de modelos, adoptando dos estrategias: la identificación y la mitigación del sesgo. El análisis realizado revela que los corpus de ironía exhiben sesgos que tienen un impacto negativo en la generalización de los modelos. Consecuentemente, la mitigación de dicho sesgo en los datos de entrenamiento produce mejoras de rendimiento en diversos dominios e idiomas. Además, el estudio examina la sensibilidad de los modelos al sesgo temático, cuando este es modificado mediante variaciones lingüísticas en los datos de evaluación, confirmando una significativa vulnerabilidad de los modelos analizados. Experimentos complementarios exploran las implicaciones sociales de la ironía y el humor, subrayando sus efectos negativos cuando se utilizan indebidamente para atacar a grupos minoritarios o individuos. Al abordar dimensiones computacionales, lingüísticas y sociales, esta tesis avanza en la detección computacional de la ironía y su implicación en la comunicación digital., [CA] La present tesi doctoral, presentada com a compendi d'articles científics, aborda el complex problema de la detecció de la ironia. Aquesta figura del llenguatge és un fenomen figuratiu i pragmàtic que eludeix l'anàlisi lingüística del significat literal a causa de la seua dependència del context i la intencionalitat. La dificultat del problema s'intensifica en contextos de xarxes socials, on la ironia es manifesta en dimensions multilingües i multimodals, complicant la seua modelització computacional. En aquest sentit, aquesta investigació està motivada en dues direccions: primer, les aplicacions potencials de la detecció de la ironia més enllà de l'anàlisi de sentiments, especialment en la mitigació de problemes de comunicació i desinformació; segon, les limitacions existents dels mètodes automàtics per a la detecció eficaç de la ironia. Aquesta tesi explora tres direccions principals. Inicialment, s'analitza la ironia i la sàtira a través de variants de l'espanyol, introduint un nou mètode que integra coneixement lingüístic i representacions profundes dins d'un model neuronal recurrent dotat amb mecanismes d'atenció, superant el rendiment de l'estat de l'art. Posteriorment, s'estén la investigació a l'escenari multimodal, proposant una nova arquitectura que combina transformers textuals amb transformers visuals, millorant la precisió en la detecció de la ironia en continguts de xarxes socials basats en text i imatge. Finalment, s'investiga el biaix temàtic en corpus d'ironia i el seu impacte en la generalització dels models, adoptant dues estratègies: la identificació i la mitigació del biaix. L'anàlisi realitzada revela que els corpus d'ironia exhibeixen biaixos que tenen un impacte negatiu en la generalització dels models. Conseqüentment, la mitigació d'aquest biaix en les dades d'entrenament produeix millores de rendiment en diversos dominis i idiomes. A més, l'estudi examina la sensibilitat dels models al biaix temàtic, quan aquest és modificat mitjançant variacions lingüístiques en les dades d'avaluació, confirmant una significativa vulnerabilitat dels models analitzats. Experiments complementaris exploren les implicacions socials de la ironia i l'humor, subratllant els seus efectes negatius quan s'utilitzen indegudament per a atacar grups minoritaris o individus. En abordar dimensions computacionals, lingüístiques i socials, aquesta tesi avança en la detecció computacional de la ironia i la seua implicació en la comunicació digital., [EN] This doctoral thesis, presented as a compendium of articles, addresses the complex challenge of irony detection. Irony is a figurative and pragmatic phenomenon that eludes straightforward linguistic analysis due to its reliance on context and intent. The difficulty intensifies in social media contexts, where irony manifests itself in multilingual and multimodal dimensions, complicating computational modelling. The motivation of this research is twofold: first, the potential applications of irony detection beyond sentiment analysis, notably in mitigating miscommunication and misinformation; second, the persistent computational gap in accurately identifying irony. The thesis explores three main directions. Initially, it analyses irony and satire across Spanish variants, introducing a novel method that integrates linguistic features and deep representations within an attentive LSTM model, surpassing state-of-the-art performance in irony and satire detection. Subsequently, it extends the investigation to multimodal scenarios, proposing an end-to-end architecture that fuses textual transformers with visual transformers, enhancing accuracy in detecting irony across text- and image-based social media content. Finally, it investigates topic bias in irony corpora and its impact on model generalisation, considering two perspectives: bias identification and mitigation. Analysis reveals that irony corpora exhibit bias, which impairs model generalisation; mitigating this bias in training data yields performance improvements across domains and languages. Furthermore, the study examines the model sensitivity to bias altered by linguistic perturbations in the evaluation data, confirming a significant vulnerability. Complementary experiments explore the social implications of irony and humour, underscoring their negative effects when misused to target minorities or individuals. By addressing linguistic, computational, and social dimensions, this thesis advances computational irony detection and its broader implications in digital communication., The work of the first two authors was in the framework of the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31), funded by Spanish Ministry of Science and Innovation, and DeepPattern (PROMETEO/2019/121), funded by the Generalitat Valenciana. The work of Paolo Rosso was in the framework of the research project on Fairness and Transparency for equitable NLP applications in social media (grant PID2021-124361OB-C31 funded by MCIN/AEI/10.13039/501100011033 and by ERDF/EU)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Overview of PAN 2020: Authorship Verification, Celebrity Profiling, Profiling Fake News Spreaders on Twitter, and Style Change Detection

  • Bevendorff, Janek
  • Ghanem, Bilal Hisham Hasan
  • Giachanou, Anastasia
  • Kestemont, Mike
  • Manjavacas, Enrique
  • Markov, Ilia
  • Mayerl, Maximilian
  • Potthast, Martin
  • Rangel, Francisco
  • Specht, Günther
  • Stamatatos, Efstathios
  • Stein, Benno
  • Wiegmann, Matti
  • Zangerle, Eva
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] We briefly report on the four shared tasks organized as part of the PAN 2020 evaluation lab on digital text forensics and authorship analysis. Each tasks is introduced, motivated, and the results obtained are presented. Altogether, the four tasks attracted 230 registrations, yielding 83 successful submissions. This, and the fact that we continue to invite the submissions of software rather than its run output using the TIRA experimentation platform, marks for a good start into the second decade of PAN evaluations labs., We thank Symanto for sponsoring the ex aequo award for the two best performing systems at the author profiling shared task of this year on Profiling fake news spreaders on Twitter. The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018¿096212-B-C31). The work of Anastasia Giachanou is supported by the SNSF Early Postdoc Mobility grant under the project Early Fake News Detection on Social Media, Switzerland (P2TIP2_181441).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Classifier combination approach for question classification for Bengali question answering system

  • Banerjee, Somnath
  • Kumar Naskar, Sudip
  • Bndyopadhyay, Sivaji
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Question classification (QC) is a prime constituent of an automated question answering system. The work presented here demonstrates that a combination of multiple models achieves better classification performance than those obtained with existing individual models for the QC task in Bengali. We have exploited state-of-the-art multiple model combination techniques, i.e., ensemble, stacking and voting, to increase QC accuracy. Lexical, syntactic and semantic features of Bengali questions are used for four well-known classifiers, namely Naive Bayes, kernel Naive Bayes, Rule Induction and Decision Tree, which serve as our base learners. Single-layer question-class taxonomy with 8 coarse-grained classes is extended to two-layer taxonomy by adding 69 fine-grained classes. We carried out the experiments both on single-layer and two-layer taxonomies. Experimental results confirmed that classifier combination approaches outperform single-classifier classification approaches by 4.02% for coarse-grained question classes. Overall, the stacking approach produces the best results for fine-grained classification and achieves 87.79% of accuracy. The approach presented here could be used in other Indo-Aryan or Indic languages to develop a question answering system., Somnath Banerjee and Sudip Kumar Naskar are supported by Digital India Corporation (formerly Media Lab Asia), MeitY, Government of India, under the Visvesvaraya Ph.D. Scheme for Electronics and IT. The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project PGC2018-096212-B-C31.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Fake Opinion Detection: How Similar are Crowdsourced Datasets to Real Data?

  • Fornaciari, Tommaso
  • Cagnina, Leticia
  • Poesio, Massimo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Identifying deceptive online reviews is a challenging tasks for Natural Language Processing (NLP). Collecting corpora for the task is difficult, because normally it is not possible to know whether reviews are genuine. A common workaround involves collecting (supposedly) truthful reviews online and adding them to a set of deceptive reviews obtained through crowdsourcing services. Models trained this way are generally successful at discriminating between `genuine¿ online reviews and the crowdsourced deceptive reviews. It has been argued that the deceptive reviews obtained via crowdsourcing are very different from real fake reviews, but the claim has never been properly tested. In this paper, we compare (false) crowdsourced reviews with a set of `real¿ fake reviews published on line. We evaluate their degree of similarity and their usefulness in training models for the detection of untrustworthy reviews. We find that the deceptive reviews collected via crowdsourcing are significantly different from the fake reviews published online. In the case of the artificially produced deceptive texts, it turns out that their domain similarity with the targets affects the models¿ performance, much more than their untruthfulness. This suggests that the use of crowdsourced datasets for opinion spam detection may not result in models applicable to the real task of detecting deceptive reviews. As an alternative method to create large-size datasets for the fake reviews detection task, we propose methods based on the probabilistic annotation of unlabeled texts, relying on the use of meta-information generally available on the e-commerce sites. Such methods are independent from the content of the reviews and allow to train reliable models for the detection of fake reviews., Leticia Cagnina thanks CONICET for the continued financial support. This work was funded by MINECO/FEDER (Grant No. SomEMBED TIN2015-71147-C2-1-P). The work of Paolo Rosso was partially funded by the MISMIS-FAKEnHATE Spanish MICINN research project (PGC2018-096212-B-C31). Massimo Poesio was in part supported by the UK Economic and Social Research Council (Grant Number ES/M010236/1).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

#Brexit: Leave or Remain? The Role of User&apos, s Community and Diachronic Evolution on Stance Detection

  • Lai, Mirko
  • Patti, Viviana
  • Ruffo, Giancarlo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Interest has grown around the classification of stance that users assume within online debates in recent years. Stance has been usually addressed by considering users posts in isolation, while social studies highlight that social communities may contribute to influence users¿ opinion. Furthermore, stance should be studied in a diachronic perspective, since it could help to shed light on users¿ opinion shift dynamics that can be recorded during the debate. We analyzed the political discussion in UK about the BREXIT referendum on Twitter, proposing a novel approach and annotation schema for stance detection, with the main aim of investigating the role of features related to social network community and diachronic stance evolution. Classification experiments show that such features provide very useful clues for detecting stance., The work of P. Rosso was partially funded by the Spanish MICINN under the research projects MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech(PGC2018-096212-B-C31) and PROMETEO/2019/121 (DeepPattern) of the Generalitat Valenciana. The work of V. Patti and G. Ruffo was partially funded by Progetto di Ateneo/CSP 2016 Immigrants, Hate and Prejudice in Social Media (S1618 L2 BOSC 01).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

A Decade of Shared Tasks in Digital Text Forensics at PAN

  • Potthast, Martin
  • Stamatatos, Efstathios
  • Stein, Benno
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Digital text forensics aims at examining the originality and credibility of information in electronic documents and, in this regard, to extract and analyze information about the authors of these documents. The research field has been substantially developed during the last decade. PAN is a series of shared tasks that started in 2009 and significantly contributed to attract the attention of the research community in well-defined digital text forensics tasks. Several benchmark datasets have been developed to assess the state-of-the-art performance in a wide range of tasks. In this paper, we present the evolution of both the examined tasks and the developed datasets during the last decade. We also briefly introduce the upcoming PAN 2019 shared tasks., We are indebted to many colleagues and friends who contributed greatly to PAN's tasks: Maik Anderka, Shlomo Argamon, Alberto Barrón-Cedeño, Fabio Celli, Fabio Crestani, Walter Daelemans, Andreas Eiselt, Tim Gollub, Parth Gupta, Matthias Hagen, Teresa Holfeld, Patrick Juola, Giacomo Inches, Mike Kestemont, Moshe Koppel, Manuel Montes-y-Gómez, Aurelio Lopez-Lopez, Francisco Rangel, Miguel Angel Sánchez-Pérez, Günther Specht, Michael Tschuggnall, and Ben Verhoeven. Our special thanks go to PAN¿s sponsors throughout the years and not least to the hundreds of participants.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

MSIR@FIRE: A Comprehensive Report from 2013 to 2016

  • Banerjee, Somnath
  • Choudhury, Monojit
  • Chakma, Kunal
  • Kumar Naskar, Sudip
  • Das, Amitava
  • Bandyopadhyay, Sivaji
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] India is a nation of geographical and cultural diversity where over 1600 dialects are spoken by the people. With the technological advancement, penetration of the internet and cheaper access to mobile data, India has recently seen a sudden growth of internet users. These Indian internet users generate contents either in English or in other vernacular Indian languages. To develop technological solutions for the contents generated by the Indian users using the Indian languages, the Forum for Information Retrieval Evaluation (FIRE) was established and held for the first time in 2008. Although Indian languages are written using indigenous scripts, often websites and user-generated content (such as tweets and blogs) in these Indian languages are written using Roman script due to various socio-cultural and technological reasons. A challenge that search engines face while processing transliterated queries and documents is that of extensive spelling variation. MSIR track was first introduced in 2013 at FIRE and the aim of MSIR was to systematically formalize several research problems that one must solve to tackle the code mixing in Web search for users of many languages around the world, develop related data sets, test benches and most importantly, build a research community focusing on this important problem that has received very little attention. This document is a comprehensive report on the 4 years of MSIR track evaluated at FIRE between 2013 and 2016., Somnath Banerjee and Sudip Kumar Naskar are supported by Media Lab Asia, MeitY, Government of India, under the Visvesvaraya PhD Scheme for Electronics & IT. The work of Paolo Rosso was partially supported by the MISMIS research project PGC2018-096212-B-C31 funded by the Spanish MICINN.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Categorizing Misogynistic Behaviours in Italian, English and Spanish Tweets, Categorización de comportamientos misóginos en tweets en italiano, inglés y español

  • Lazzardi, Silvia
  • Patti, Viviana
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Misogyny is a multifaceted phenomenon and can be linguistically manifested in numerous ways. The evaluation campaigns of EVALITA and IberEval in 2018 proposed a shared task of Automatic Misogyny Identification (AMI) based on Italian, English and Spanish tweets. Since the participating teams¿ results were pretty low in the misogynistic behaviour categorization, the aim of this study is to investigate the possible causes. We measured the overlap and the homogeneity of the clusters by varying the number of categories. This experiment showed that the clusters overlap. Finally, we tested several machine learning models both using the original data sets and merging together some categories according to their overlap, obtaining an increase in terms of macro F1., [ES] La misoginia es un fen¿omeno con m¿ultiples facetas y puede manifestarse ling¿u¿¿sticamente de muchas formas. Las campa¿nas de evaluaci¿on de EVALITA e IberEval en 2018 propusieron una tarea compartida de Identificaci¿on Autom¿atica de Misoginia (AMI) basada en tweets en italiano, ingl¿es y espa¿nol. Dado que los resultados de los equipos participantes fueron bastante bajos en la categorizaci¿on del comportamiento mis¿ogino, el objetivo de este estudio es investigar las posibles causas. Medimos el solape y la homogeneidad de los cl¿usteres variando el n¿umero de categor¿¿as. Este experimento mostr¿o que los grupos se solapan. Finalmente probamos varios modelos de aprendizaje autom¿atico utilizando los conjuntos de datos originales y fusionando algunas categor¿¿as de acuerdo con consideraciones basadas en medidas de similitud y las matrices de confusi¿on de los modelos, obteniendo un aumento de la F1 macro, The work of S. Lazzardi was partially carried out at the Universitat Politecnica de Valencia within the framework of the Erasmus+ program, Erasmus Traineeship 2018/19 funding. The work of P. Rosso was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31). The work of V. Patti was partially funded by the research projects "STudying European Racial Hoaxes and sterEOTYPES"(STERHEOTYPES, under the call "Challenges for Europe" of VolksWagen Stiftung and Compagnia di San Paolo) and "Be Positive!" (under the 2019 \Google.org Impact Challenge on Safety" call).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

RoMa at HAHA-2021: Deep Reinforcement Learning to Improve a Transformed-based Model for Humor Detection

  • Rodriguez, Mariano
  • Ortega-Bueno, Reynier
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] In this paper, we describe our system we participated in the shared task ¿Humor Analysis based on Human Annotation (HAHA) at IberLEF-2021 with. Our system relies on data representations learned through fine-tuned neural language models. The representations are used to train a Siamese Neural Network (SNN) which learns to verify whether or not a pair of tweets belong to the same or distinct classes. A key point in our model is the heuristic used to create the pair of messages in the training and test phases. For that, we used a Deep Reinforcement Learning (DRL) strategy that aims at identifying a set of optimal prototypes in each class. In general, the results achieved are encouraging and give us a starting point for further improvements., The work of the second and third authors was in the framework of the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31), funded by Spanish Ministry of Science and Innovation, and DeepPattern (PROMETEO/2019/121), funded by the Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Overview of PAN 2019: Bots and Gender Profiling, Celebrity Profiling, Cross-domain Authorship Attribution and Style Change Detection

  • Daelemans, Walter
  • Kestemont, Mike
  • Manjavacas, Enrique
  • Potthast, Martin
  • Rangel, Francisco
  • Specht, Günther
  • Stamatatos, Efstathios
  • Stein, Benno
  • Tschuggnall, Michael
  • Wiegmann, Matti
  • Zangerle, Eva
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] We briefly report on the four shared tasks organized as part of the PAN 2019 evaluation lab on digital text forensics and authorship analysis. Each task is introduced, motivated, and the results obtained are presented. Altogether, the four tasks attracted 373 registrations, yielding 72 successful submissions. This, and the fact that we continue to invite the submission of software rather than its run output using the TIRA experimentation platform, demarcates a good start into the second decade of PAN evaluations labs., The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212- B-C31). Our special thanks goes to all PAN participants for providing high-quality submission, to Symanto (https://www.symanto.net) for sponsoring the PAN Lab 2019 and to The Logic Value (https://thelogicvalue.com) for sponsoring the author profiling shared task award.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Tesis doctoral (DoctoralThesis)

Detecting Deception, Partisan, and Social Biases

  • Sánchez Junquera, Juan Javier
[ES] En la actualidad, el mundo político tiene tanto o más impacto en la sociedad que ésta en el mundo político. Los líderes o representantes de partidos políticos hacen uso de su poder en los medios de comunicación, para modificar posiciones ideológicas y llegar al pueblo con el objetivo de ganar popularidad en las elecciones gubernamentales.A través de un lenguaje engañoso, los textos políticos pueden contener sesgos partidistas y sociales que minan la percepción de la realidad. Como resultado, los seguidores de una ideología, o miembros de una categoría social, se sienten amenazados por otros grupos sociales o ideológicos, o los perciben como competencia, derivándose así una polarización política con agresiones físicas y verbales. La comunidad científica del Procesamiento del Lenguaje Natural (NLP, según sus siglas en inglés) contribuye cada día a detectar discursos de odio, insultos, mensajes ofensivos, e información falsa entre otras tareas computacionales que colindan con ciencias sociales. Sin embargo, para abordar tales tareas, es necesario hacer frente a diversos problemas entre los que se encuentran la dificultad de tener textos etiquetados, las limitaciones de no trabajar con un equipo interdisciplinario, y los desafíos que entraña la necesidad de soluciones interpretables por el ser humano. Esta tesis se enfoca en la detección de sesgos partidistas y sesgos sociales, tomando como casos de estudio el hiperpartidismo y los estereotipos sobre inmigrantes. Para ello, se propone un modelo basado en una técnica de enmascaramiento de textos capaz de detectar lenguaje engañoso incluso en temas controversiales, siendo capaz de capturar patrones del contenido y el estilo de escritura. Además, abordamos el problema usando modelos basados en BERT, conocidos por su efectividad al capturar patrones sintácticos y semánticos sobre las mismas representaciones de textos. Ambos enfoques, la técnica de enmascaramiento y los modelos basados en BERT, se comparan en términos de desempeño y explicabilidad en la detección de hiperpartidismo en noticias políticas y estereotipos sobre inmigrantes. Para la identificación de estos últimos, se propone una nueva taxonomía con fundamentos teóricos en sicología social, y con la que se etiquetan textos extraídos de intervenciones partidistas llevadas a cabo en el Parlamento español. Los resultados muestran que los enfoques propuestos contribuyen al estudio del hiperpartidismo, así como a identif i car cuándo los ciudadanos y políticos enmarcan a los inmigrantes en una imagen de víctima, recurso económico, o amenaza. Finalmente, en esta investigación interdisciplinaria se demuestra que los estereotipos sobre inmigrantes son usados como estrategia retórica en contextos políticos., [CA] Avui, el món polític té tant o més impacte en la societat que la societat en el món polític. Els líders polítics, o representants dels partits polítics, fan servir el seu poder als mitjans de comunicació per modif i car posicions ideològiques i arribar al poble per tal de guanyar popularitat a les eleccions governamentals. Mitjançant un llenguatge enganyós, els textos polítics poden contenir biaixos partidistes i socials que soscaven la percepció de la realitat. Com a resultat, augmenta la polarització política nociva perquè els seguidors d'una ideologia, o els membres d'una categoria social, veuen els altres grups com una amenaça o competència, que acaba en agressions verbals i físiques amb resultats desafortunats. La comunitat de Processament del llenguatge natural (PNL) té cada dia noves aportacions amb enfocaments que ajuden a detectar discursos d'odi, insults, missatges ofensius i informació falsa, entre altres tasques computacionals relacionades amb les ciències socials. No obstant això, molts obstacles impedeixen eradicar aquests problemes, com ara la dif i cultat de tenir textos anotats, les limitacions dels enfocaments no interdisciplinaris i el repte afegit per la necessitat de solucions interpretables. Aquesta tesi se centra en la detecció de biaixos partidistes i socials, prenent com a cas pràctic l'hiperpartidisme i els estereotips sobre els immigrants. Proposem un model basat en una tècnica d'emmascarament que permet detectar llenguatge enganyós en temes polèmics i no polèmics, capturant pa-trons relacionats amb l'estil i el contingut. A més, abordem el problema avaluant models basats en BERT, coneguts per ser efectius per capturar patrons semàntics i sintàctics en la mateixa representació. Comparem aquests dos enfocaments (la tècnica d'emmascarament i els models basats en BERT) en termes de rendiment i les seves solucions explicables en la detecció de l'hiperpartidisme en les notícies polítiques i els estereotips d'immigrants. Per tal d'identificar els estereotips dels immigrants, proposem una nova tax-onomia recolzada per la teoria de la psicologia social i anotem un conjunt de dades de les intervencions partidistes al Parlament espanyol. Els resultats mostren que els nostres models poden ajudar a estudiar l'hiperpartidisme i identif i car diferents marcs en què els ciutadans i els polítics perceben els immigrants com a víctimes, recursos econòmics o amenaces. Finalment, aquesta investigació interdisciplinària demostra que els estereotips dels immigrants s'utilitzen com a estratègia retòrica en contextos polítics., [EN] Today, the political world has as much or more impact on society than society has on the political world. Political leaders, or representatives of political parties, use their power in the media to modify ideological positions and reach the people in order to gain popularity in government elections. Through deceptive language, political texts may contain partisan and social biases that undermine the perception of reality. As a result, harmful political polarization increases because the followers of an ideology, or members of a social category, see other groups as a threat or competition, ending in verbal and physical aggression with unfortunate outcomes. The Natural Language Processing (NLP) community has new contri-butions every day with approaches that help detect hate speech, insults, of f ensive messages, and false information, among other computational tasks related to social sciences. However, many obstacles prevent eradicating these problems, such as the dif f i culty of having annotated texts, the limitations of non-interdisciplinary approaches, and the challenge added by the necessity of interpretable solutions. This thesis focuses on the detection of partisan and social biases, tak-ing hyperpartisanship and stereotypes about immigrants as case studies. We propose a model based on a masking technique that can detect deceptive language in controversial and non-controversial topics, capturing patterns related to style and content. Moreover, we address the problem by evalu-ating BERT-based models, known to be ef f ective at capturing semantic and syntactic patterns in the same representation. We compare these two approaches (the masking technique and the BERT-based models) in terms of their performance and the explainability of their decisions in the detection of hyperpartisanship in political news and immigrant stereotypes. In order to identify immigrant stereotypes, we propose a new taxonomy supported by social psychology theory and annotate a dataset from partisan interventions in the Spanish parliament. Results show that our models can help study hyperpartisanship and identify dif f erent frames in which citizens and politicians perceive immigrants as victims, economic resources, or threat. Finally, this interdisciplinary research proves that immigrant stereotypes are used as a rhetorical strategy in political contexts., This PhD thesis was funded by the MISMIS-FAKEnHATE research project (PGC2018-096212-B-C31) of the Spanish Ministry of Science and Innovation.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Irony detection via sentiment-based transfer learning

  • Zhang, Shiwei
  • Zhang, Xiuzhen
  • Chan, Jeffrey
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Irony as a literary technique is widely used in online texts such as Twitter posts. Accurate irony detection is crucial for tasks such as effective sentiment analysis. A text's ironic intent is defined by its context incongruity. For example in the phrase "I love being ignored", the irony is defined by the incongruity between the positive word "love" and the negative context of "being ignored". Existing studies mostly formulate irony detection as a standard supervised learning text categorization task, relying on explicit expressions for detecting context incongruity. In this paper we formulate irony detection instead as a transfer learning task where supervised learning on irony labeled text is enriched with knowledge transferred from external sentiment analysis resources. Importantly, we focus on identifying the hidden, implicit incongruity without relying on explicit incongruity expressions, as in "I like to think of myself as a broken down Justin Bieber my philosophy professor." We propose three transfer learning-based approaches to using sentiment knowledge to improve the attention mechanism of recurrent neural models for capturing hidden patterns for incongruity. Our main findings are: (1) Using sentiment knowledge from external resources is a very effective approach to improving irony detection; (2) For detecting implicit incongruity, transferring deep sentiment features seems to be the most effective way. Experiments show that our proposed models outperform state-of-the-art neural models for irony detection.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

The Role of Personality and Linguistic Patterns in Discriminating Between Fake News Spreaders and Fact Checkers

  • Giachanou, Anastasia
  • Ríssola, Esteban A.
  • Ghanem, Bilal
  • Crestani, Fabio
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Users play a critical role in the creation and propagation of fake news online by consuming and sharing articles with inaccurate information either intentionally or unintentionally. Fake news are written in a way to confuse readers and therefore understanding which articles contain fabricated information is very challenging for non-experts. Given the di culty of the task, several fact checking websites have been developed to raise awareness about which articles contain fabricated information. As a result of those platforms, several users are interested to share posts that cite evidence with the aim to refute fake news and warn other users. These users are known as fact checkers. However, there are users who tend to share false information, who can be characterised as potential fake news spreaders. In this paper, we propose the CheckerOrSpreader model that can classify a user as a potential fact checker or a potential fake news spreader. Our model is based on a Convolutional Neural Network (CNN) and combines word embeddings with features that represent users' personality traits and linguistic patterns used in their tweets. Experimental results show that leveraging linguistic patterns and personality traits can improve the performance in di erentiating between checkers and spreaders., The work of the first author is supported by the SNSF Early Postdoc Mobility grant under the project Early Fake News Detection on Social Media, Switzerland (P2TIP2 181441). The work of Paolo Rosso is partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018- 096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

An Emotional Analysis of False Information in Social Media and News Articles

  • Ghanem, Bilal Hisham Hasan
  • Rangel, Francisco
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Fake news is risky since it has been created to manipulate the readers' opinions and beliefs. In this work, we compared the language of false news to the real one of real news from an emotional perspective, considering a set of false information types (propaganda, hoax, clickbait, and satire) from social media and online news articles sources. Our experiments showed that false information has different emotional patterns in each of its types, and emotions play a key role in deceiving the reader. Based on that, we proposed a LSTM neural network model that is emotionally-infused to detect false news., The work of the second author was partially funded by the Spanish MICINN under the research project MISMISFAKEnHATE on Misinformation and Miscommunication in social media: FAKEnews and HATE speech (PGC2018-096212B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Masking domain-specific information for cross-domain deception detection

  • Sánchez-Junquera, Javier
  • Villaseñor-Pineda, Luis
  • Montes-y-Gómez, Manuel
  • Stamatatos, Efstathios
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] The facilities provided by social media and computer-mediated communication make easy the dissemination of deceptive behavior, after which different entities or people could be affected. The deception detection by supervised learning has been widely studied; however, the scenario in which there is one domain of interest and the labeled data is in another domain has received poor attention. This paper presents, to our knowledge, the first domain adaptation approach for cross-domain deception detection in texts. Our proposal consists in modifying original texts from the source and target domains in a form in which common content and style information is maintained, but domain-specific information is masked. In order to adequately select domain-specific terms to be masked, the proposed method uses unlabeled instances from both domains. Our experiments demonstrate that the masking technique is a good idea for detecting deception in cross-domain scenarios; and the performance could be further improved if unlabeled information from the target domain is considered., The work of Javier Sanchez-Junquera and Paolo Rosso was funded by the MISMIS-FAKEnHATE Spanish MICINN research project (PGC2018-096212-B-C31). This work was parrtially supported also by CONACyT under grants CB-2015-01-257383, the Thematic Networks program, and the scholarship CONACyT-Mexico 613411.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

The unbearable hurtfulness of sarcasm

  • Frenda, Simona
  • Cignarella, Alessandra Teresa
  • Basile, Valerio
  • Bosco, Cristina
  • Patti, Viviana
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] In the last decade, the need to detect automatically irony to correctly recognize the sentiment and hate speech involved in online texts increased the investigation on humorous figures of speech in NLP. The slight boundaries among various types of irony lead to think of irony as a linguistic phenomenon that covers sarcasm, satire, humor and parody joined by their trend to create a secondary or opposite meaning to the literal one expressed in the message. Although this commonality, in literature sarcasm is defined as a type of irony more aggressive with the intent to mock or scorn a victim without excluding the possibility to amuse. The aggressive tone and the intent of contempt suggest that sarcasm involves some peculiarities that make it a suitable type of irony to disguise negative messages. To investigate these peculiarities of sarcasm, we examined the dataset of the IronITA shared task. It consists of Italian tweets about controversial social issues, such as immigration, politics and other more general topics. Each tweet is annotated as ironic and non-ironic, and, at a deeper level, as sarcastic and non-sarcastic. Qualitative and quantitative analyses of the dataset showed how sarcasm tends to be expressed with hurtful language revealing the aggressive intention with which the author targets the victim. While irony is characterized by being offensive in hateful context and, in general, moved by negative emotions. For a better understanding of the impact of hurtful and affective language on the detection of irony and sarcasm, we proposed a transformer-based system, called AlBERToIS, combining pre-trained AlBERTo model with linguistic features. This approach obtained the best performances on irony and sarcasm detection on the IronITA dataset., The work of S. Frenda, A.T. Cignarella, C. Bosco and V. Patti was partially funded by VolksWagen Stiftung and Compagnia di San Paolo, Italy under the call "Challenges for Europe'' for the research projects "STudying European Racial Hoaxes and sterEOTYPES'' (STERHEOTYPES, S129542). The work of V. Basile, A.T. Cignarella, C. Bosco and V. Patti was partially funded by Google, Italy under the call "Google.org Impact Challenge on Safety'' for the project "Be Positive!''. Finally, the work of P. Rosso was partially funded by the Spanish Ministry of Science and Innovation, Spain under the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media "FAKE news and HATE speech'' (PGC2018-096212-BC31) and by the Generalitat Valenciana under DeepPattern, Spain (PROMETEO/2019/121).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

How Do You Speak about Immigrants? Taxonomy and StereoImmigrants Dataset for Identifying Stereotypes about Immigrants

  • Sánchez-Junquera, Javier
  • Chulvi-Ferriols, María Alberta
  • Ponzetto, Simone Paolo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Stereotype is a type of social bias massively present in texts that computational models use. There are stereotypes that present special difficulties because they do not rely on personal attributes. This is the case of stereotypes about immigrants, a social category that is a preferred target of hate speech and discrimination. We propose a new approach to detect stereotypes about immigrants in texts focusing not on the personal attributes assigned to the minority but in the frames, that is, the narrative scenarios, in which the group is placed in public speeches. We have proposed a fine-grained social psychology grounded taxonomy with six categories to capture the different dimensions of the stereotype (positive vs. negative) and annotated a novel StereoImmigrants dataset with sentences that Spanish politicians have stated in the Congress of Deputies. We aggregate these categories in two supracategories: one is Victims that expresses the positive stereotypes about immigrants and the other is Threat that expresses the negative stereotype. We carried out two preliminary experiments: first, to evaluate the automatic detection of stereotypes; and second, to distinguish between the two supracategories of immigrants¿ stereotypes. In these experiments, we employed state-of-the-art transformer models (monolingual and multilingual) and four classical machine learning classifiers. We achieve above 0.83 of accuracy with the BETO model in both experiments, showing that transformers can capture stereotypes about immigrants with a high level of accuracy., The work of the authors from the Universitat Politecnica de Valencia was funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31). The work of Paolo Rosso was done also in the framework of the research project PROMETEO/2019/121 (DeepPattern) funded by the Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Introduction to the Special Section on Computational Modeling and Understanding of Emotions in Conflictual Social Interactions

  • Damiano, Rossana
  • Patti, Viviana
  • Clavel, Chloé
  • Rosso, Paolo|||0000-0002-8922-1242
The editorial work of C. Clavel for this special issue was partially supported by a grant overseen by the French National Research Agency (ANR17-MAOI) and by the European project H2020 ANIMATAS (MSCA-ITN-ETN 7659552). The editorial work of V. Patti was partially funded by Progetto di Ateneo/CSP 2016 (Immigrants, Hate and Prejudice in Social Media, S1618_L2_BOSC_01). P. Rosso was partially funded by Spanish MICINN under the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Overview of the 8th Author Profiling Task at PAN 2020: Profiling Fake News Spreaders on Twitter

  • Rangel, Francisco
  • Giachanou, Anastasia
  • Ghanem, Bilal Hisham Hasan
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] This overview presents the Author Profiling shared task at PAN 2020. The focus of this year's task is on determining whether or not the author of a Twitter feed is keen to spread fake news. Two have been the main aims: (i) to show the feasibility of automatically identifying potential fake news spreaders in Twitter; and (ii) to show the difficulty of identifying them when they do not limit themselves to just retweet domain-specific news. For this purpose a corpus with Twitter data has been provided, covering the English and Spanish languages. Altogether, the approaches of 66 participants have been evaluated., First of all we thank the participants: 66 this year, record in terms of participants at PAN Lab since 2009! We have to thank also Martin Potthast, Matti Wiegmann, and Nikolay Kolyada to help with the 66 Virtual Machines in the TIRA platform. We thank Symanto for sponsoring the ex aequo award for the two best performing systems at the author profiling shared task of this year. The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31). The work of Anastasia Giachanou is supported by the SNSF Early Postdoc Mobility grant under the project Early Fake News Detection on Social Media, Switzerland (P2TIP2 181441).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Offensive keyword extraction based on the attention mechanism of BERT and the eigenvector centrality using a graph representation

  • Peña-Sarracén, Gretel Liz de la
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] The proliferation of harmful content on social media affects a large part of the user community. Therefore, several approaches have emerged to control this phenomenon automatically. However, this is still a quite challenging task. In this paper, we explore the offensive language as a particular case of harmful content and focus our study in the analysis of keywords in available datasets composed of offensive tweets. Thus, we aim to identify relevant words in those datasets and analyze how they can affect model learning. For keyword extraction, we propose an unsupervised hybrid approach which combines the multi-head self-attention of BERT and a reasoning on a word graph. The attention mechanism allows to capture relationships among words in a context, while a language model is learned. Then, the relationships are used to generate a graph from what we identify the most relevant words by using the eigenvector centrality. Experiments were performed by means of two mechanisms. On the one hand, we used an information retrieval system to evaluate the impact of the keywords in recovering offensive tweets from a dataset. On the other hand, we evaluated a keyword-based model for offensive language detection. Results highlight some points to consider when training models with available datasets.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Masking and BERT-based Models for Stereotype Identication, Modelos Basados en Enmascaramiento y en BERT para la Identificación de Estereotipos

  • Sánchez-Junquera, Juan Javier
  • Montes Gomez, Manuel
  • Chulvi-Ferriols, María Alberta
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Stereotypes about immigrants are a type of social bias increasingly present in the human interaction in social networks and political speeches. This challenging task is being studied by computational linguistics because of the rise of hate messages, offensive language, and discrimination that many people receive. In this work, we propose to identify stereotypes about immigrants using two different explainable approaches: a deep learning model based on Transformers; and a text masking technique that has been recognized by its capabilities to deliver good and human-understandable results. Finally, we show the suitability of the two models for the task and offer some examples of their advantages in terms of explainability, [ES] Los estereotipos sobre inmigrantes son un tipo de sesgo social cada vez m¿as presente en la interacci¿on humana en redes sociales y en los discursos pol¿¿ticos. Esta desafiante tarea est¿a siendo estudiada por la ling¿u¿¿stica computacional debido al aumento de los mensajes de odio, el lenguaje ofensivo, y la discriminaci¿on que reciben muchas personas. En este trabajo, nos proponemos identificar estereotipos sobre inmigrantes utilizando dos enfoques diametralmente opuestos prestando atenci¿on a la explicabilidad de los mismos: un modelo de aprendizaje profundo basado en Transformers; y una t¿ecnica de enmascaramiento de texto que ha sido reconocida por su capacidad para ofrecer buenos resultados a la vez que comprensibles para los humanos. Finalmente, mostramos la idoneidad de los dos modelos para la tarea, y ofrecemos algunos ejemplos de sus ventajas en t¿erminos de explicabilidad, The work of the authors from the Universitat Politecnica of Valencia was funded by the Spanish Ministry of Science and Innovation under the research project MISMISFAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31). Experiments were carried out on the GPU cluster at PRHLT thanks to the PROMETEO/2019/121 (DeepPattern) research project funded by the Generalitat Valenciana
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Multi-view informed attention-based model for Irony and Satire detection in Spanish variants

  • Ortega-Bueno, Reynier
  • Medina-Pagola, José E.
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Making machines understand language and reasoning on it has been one of the most challenging problems addressed by Artificial Intelligent researchers. This challenge increases when figurative language is used for communicating complex meanings, intentions, emotions and attitudes in creative and funny ways. In fact, sentiment analysis approaches struggle when facing irony, satire and other figurative languages, particularly those where the explanation of a prediction might arguably be as necessary as the prediction itself. This paper describes a new model MvAttLSTM based on deep learning for irony and satire detection in tweets written in distinct Spanish variants. The proposed model is based on an attentive-LSTM informed with three additional views learned from distinct perspectives. We investigate two strategies to pass these views into MvAttLSTM. We perform an extensive evaluation on three corpora, one for irony detection and two for satire detection. Moreover, in order to study the robustness of our proposed model, we investigate its performance on humor recognition. Experiments confirm that the proposed views help our model to improve its performance. Moreover, they show that affective information benefits our model to detect irony and satire. In particular, a first analysis of the results highlights the discriminating power of emotional features obtained from SenticNet and SEL lexicon. Overall, our system achieves the state-of-the-art performance in irony and satire detection in Spanish variants and competitive results in humor recognition., The work of the first two authors was in the framework of the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31) , funded by Spanish Ministry of Science and Innovation, and DeepPattern (PROMETEO/2019/121) , funded by the Generalitat Valenciana, Spain.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Multimodal Fake News Detection with Textual, Visual and Semantic Information

  • Giachanou, Anastasia
  • Zhang, Guobiao
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Recent years have seen a rapid growth in the number of fake news that are posted online. Fake news detection is very challenging since they are usually created to contain a mixture of false and real information and images that have been manipulated that confuses the readers. In this paper, we propose a multimodal system with the aim to di erentiate between fake and real posts. Our system is based on a neural network and combines textual, visual and semantic information. The textual information is extracted from the content of the post, the visual one from the image that is associated with the post and the semantic refers to the similarity between the image and the text of the post. We conduct our experiments on three standard real world collections and we show the importance of those features on detecting fake news., Anastasia Giachanou is supported by the SNSF Early Postdoc Mobility grant under the project Early Fake News Detection on Social Media, Switzerland (P2TIP2 181441). Guobiao Zhang is funded by China Scholarship Council (CSC) from the Ministry of Education of P.R. China. The work of Paolo Rosso is partially funded by the Spanish MICINN under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31)
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Killing me Softly: Creative and Cognitive Aspects of Implicitness in Abusive Language Online

  • Frenda, Simona
  • Patti, Viviana
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Abusive language is becoming a problematic issue for our society. The spread of messages that reinforce social and cultural intolerance could have dangerous effects in victims¿ life. State-of-the-art technologies are often effective on detecting explicit forms of abuse, leaving unidentified the utterances with very weak offensive language but a strong hurtful effect. Scholars have advanced theoretical and qualitative observations on specific indirect forms of abusive language that make it hard to be recognized automatically. In this work, we propose a battery of statistical and computational analyses able to support these considerations, with a focus on creative and cognitive aspects of the implicitness, in texts coming from different sources such as social media and news. We experiment with transformers, multi-task learning technique, and a set of linguistic features to reveal the elements involved in the implicit and explicit manifestations of abuses, providing a solid basis for computational applications.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Tesis doctoral (DoctoralThesis)

On the Keyword Extraction and Bias Analysis, Graph-based Exploration and Data Augmentation for Abusive Language Detection in Low-Resource Settings

  • Peña Sarracén, Gretel Liz de la
[ES] La detección del lenguaje abusivo es una tarea que se ha vuelto cada vez más importante en la era digital moderna, donde la comunicación se produce a través de diversas plataformas en línea. El aumento de las interacciones en estas plataformas ha provocado un aumento de la aparición del lenguaje abusivo. Abordar dicho contenido es crucial para mantener un entorno en línea seguro e inclusivo. Sin embargo, esta tarea enfrenta varios desafíos que la convierten en un área compleja y que demanda de continua investigación y desarrollo. En particular, detectar lenguaje abusivo en entornos con escasez de datos presenta desafíos adicionales debido a que el desarrollo de sistemas automáticos precisos a menudo requiere de grandes conjuntos de datos anotados. En esta tesis investigamos diferentes aspectos de la detección del lenguaje abusivo, prestando especial atención a entornos con datos limitados. Primero, estudiamos el sesgo hacia palabras clave abusivas en modelos entrenados para la detección del lenguaje abusivo. Con este propósito, proponemos dos métodos para extraer palabras clave potencialmente abusivas de colecciones de textos. Luego evaluamos el sesgo hacia las palabras clave extraídas y cómo se puede modificar este sesgo para influir en el rendimiento de la detección del lenguaje abusivo. El análisis y las conclusiones de este trabajo revelan evidencia de que es posible mitigar el sesgo y que dicha reducción puede afectar positivamente el desempeño de los modelos. Sin embargo, notamos que no es posible establecer una correspondencia similar entre la variación del sesgo y el desempeño de los modelos cuando hay escasez datos con las técnicas de reducción del sesgo estudiadas. En segundo lugar, investigamos el uso de redes neuronales basadas en grafos para detectar lenguaje abusivo. Por un lado, proponemos una estrategia de representación de textos diseñada con el objetivo de obtener un espacio de representación en el que los textos abusivos puedan distinguirse fácilmente de otros textos. Por otro lado, evaluamos la capacidad de redes neuronales convolucionales basadas en grafos para clasificar textos abusivos. La siguiente parte de nuestra investigación se centra en analizar cómo el aumento de datos puede influir en el rendimiento de la detección del lenguaje abusivo. Para ello, investigamos dos técnicas bien conocidas basadas en el principio de minimización del riesgo en la vecindad de instancias originales y proponemos una variante para una de ellas. Además, evaluamos técnicas simples basadas en el reemplazo de sinónimos, inserción aleatoria, intercambio aleatorio y eliminación aleatoria de palabras. Las contribuciones de esta tesis ponen de manifiesto el potencial de las redes neuronales basadas en grafos y de las técnicas de aumento de datos para mejorar la detección del lenguaje abusivo, especialmente cuando hay limitación de datos. Estas contribuciones han sido publicadas en conferencias y revistas internacionales., [CA] La detecció del llenguatge abusiu és una tasca que s'ha tornat cada vegada més important en l'era digital moderna, on la comunicació es produïx a través de diverses plataformes en línia. L'augment de les interaccions en estes plataformes ha provocat un augment de l'aparició de llenguatge abusiu. Abordar este contingut és crucial per a mantindre un entorn en línia segur i inclusiu. No obstant això, esta tasca enfronta diversos desafiaments que la convertixen en una àrea complexa i contínua de recerca i desenvolupament. En particular, detectar llenguatge abusiu en entorns amb escassetat de dades presenta desafiaments addicionals pel fet que el desenvolupament de sistemes automàtics precisos sovint requerix de grans conjunts de dades anotades. En esta tesi investiguem diferents aspectes de la detecció del llenguatge abusiu, prestant especial atenció a entorns amb dades limitades. Primer, estudiem el biaix cap a paraules clau abusives en models entrenats per a la detecció de llenguatge abusiu. Amb este propòsit, proposem dos mètodes per a extraure paraules clau potencialment abusives de col·leccions de textos. Després avaluem el biaix cap a les paraules clau extretes i com es pot modificar este biaix per a influir en el rendiment de la detecció de llenguatge abusiu. L'anàlisi i les conclusions d'este treball revelen evidència que és possible mitigar el biaix i que esta reducció pot afectar positivament l'acompliment dels models. No obstant això, notem que no és possible establir una correspondència similar entre la variació del biaix i l'acompliment dels models quan hi ha escassetat dades amb les tècniques de reducció del biaix estudiades. En segon lloc, investiguem l'ús de xarxes neuronals basades en grafs per a detectar llenguatge abusiu. D'una banda, proposem una estratègia de representació textual dissenyada amb l'objectiu d'obtindre un espai de representació en el qual els textos abusius puguen distingir-se fàcilment d'altres textos. D'altra banda, avaluem la capacitat de models basats en xarxes neuronals convolucionals basades en grafs per a classificar textos abusius. La següent part de la nostra investigació se centra en analitzar com l'augment de dades pot influir en el rendiment de la detecció del llenguatge abusiu. Per a això, investiguem dues tècniques ben conegudes basades en el principi de minimització del risc en el veïnatge d'instàncies originals i proposem una variant per a una d'elles. A més, avaluem tècniques simples basades en el reemplaçament de sinònims, inserció aleatòria, intercanvi aleatori i eliminació aleatòria de paraules. Les contribucions d'esta tesi destaquen el potencial de les xarxes neuronals basades en grafs i de les tècniques d'augment de dades per a millorar la detecció del llenguatge abusiu, especialment quan hi ha limitació de dades. Estes contribucions han sigut publicades en revistes i conferències internacionals., [EN] Abusive language detection is a task that has become increasingly important in the modern digital age, where communication takes place via various online platforms. The increase in online interactions has led to an increase in the occurrence of abusive language. Addressing such content is crucial to maintaining a safe and inclusive online environment. However, this task faces several challenges that make it a complex and ongoing area of research and development. In particular, detecting abusive language in environments with sparse data poses an additional challenge, since the development of accurate automated systems often requires large annotated datasets. In this thesis we investigate different aspects of abusive language detection, paying particular attention to environments with limited data. First, we study the bias toward abusive keywords in models trained for abusive language detection. To this end, we propose two methods for extracting potentially abusive keywords from datasets. We then evaluate the bias toward the extracted keywords and how this bias can be modified in order to influence abusive language detection performance. The analysis and conclusions of this work reveal evidence that it is possible to mitigate the bias and that such a reduction can positively affect the performance of the models. However, we notice that it is not possible to establish a similar correspondence between bias mitigation and model performance in low-resource settings with the studied bias mitigation techniques. Second, we investigate the use of models based on graph neural networks to detect abusive language. On the one hand, we propose a text representation framework designed with the aim of obtaining a representation space in which abusive texts can be easily distinguished from other texts. On the other hand, we evaluate the ability of models based on convolutional graph neural networks to classify abusive texts. The next part of our research focuses on analyzing how data augmentation can influence the performance of abusive language detection. To this end, we investigate two well-known techniques based on the principle of vicinal risk minimization and propose a variant for one of them. In addition, we evaluate simple techniques based on the operations of synonym replacement, random insertion, random swap, and random deletion. The contributions of this thesis highlight the potential of models based on graph neural networks and data augmentation techniques to improve abusive language detection, especially in low-resource settings. These contributions have been published in several international conferences and journals., This research work was partially funded by the Spanish Ministry of Science and Innovation under the research project MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31). The authors thank also the EU-FEDER Comunitat Valenciana 2014-2020 grant IDIFEDER/2018/025. This work was done in the framework of the research project on Fairness and Transparency for equitable NLP applications in social media, funded by MCIN/AEI/10.13039/501100011033 and by ERDF, EU A way of making EuropePI. FairTransNLP research project (PID2021-124361OB-C31) funded by MCIN/AEI/10.13039/501100011033 and by ERDF, EU A way of making Europe. Part of the work presented in this article was performed during the first author’s research visit to the University of Mannheim, supported through a Contact Fellowship awarded by the DAAD scholarship program “STIBET Doktoranden”.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Stance Polarity in Political Debates: a Diachronic Perspective of Network Homophily and Conversations on Twitter

  • Lai, Mirko
  • Tambuscio, Marcella
  • Patti, Viviana
  • Ruffo, Giancarlo
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] In the last decade, social media gained a very significant role in public debates, and despite the many intrinsic difficulties of analyzing data streaming from on-line platforms that are poisoned by bots, trolls, and low-quality information, it is undeniable that such data can still be used to test the public opinion and overall mood and to investigate how individuals communicate with each other. With the aim of analyzing the debate in Twitter on the 2016 referendum on the reform of the Italian Constitution, we created an Italian annotated corpus for stance detection for automatically estimating the stance of a relevant number of users. We take into account a diachronic perspective to shed lights on users' opinion dynamics. Furthermore, different types of social network communities, based on friendships, retweets, quotes, and replies were investigated, in order to analyze the communication among users with similar and divergent viewpoints. We observe particular aspects of users' behavior. First, our analysis suggests that users tend to be less explicit in expressing their stances after the outcome of the vote; simultaneously, users who exhibit a high number of cross-stance relations tend to become less polarized or to adopt a more neutral style in the following phase of the debate. Second, despite social media networks are generally aggregated in homogeneous communities, we highlight that the structure of the network can strongly change when different types of social relations are considered. In particular, networks defined by means of reply-to messages exhibit inverse homophily by stance, and users use more often replies for expressing diverging opinions, instead of other forms of communication. Interestingly, we also observe that the political polarization increases forthcoming the election and decreases after the election day., The work of Viviana Patti and Giancarlo Ruffo was partially funded by the Fondazione CRT under research project the Hate Speech and Social Media (2016.0688), and the "Progetto di Ateneo/CSP 2016" under research project "Immigrants, Hate and Prejudice in Social Media" (S1618_L2_BOSC_01). The work of Paolo Rosso was partially funded by the Spanish MICINN under the research project "MISMIS-FAKEnHATE on Misinformation and Miscommunication in social media: FAKE news and HATE speech" (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Online Hate Speech against Women: Automatic Identification of Misogyny and Sexism on Twitter

  • Frenda, Simona
  • Ghanem, Bilal
  • Montes-y-Gómez, Manuel
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Patriarchal behavior, such as other social habits, has been transferred online, appearing as misogynistic and sexist comments, posts or tweets. This online hate speech against women has serious consequences in real life, and recently, various legal cases have arisen against social platforms that scarcely block the spread of hate messages towards individuals. In this difficult context, this paper presents an approach that is able to detect the two sides of patriarchal behavior, misogyny and sexism, analyzing three collections of English tweets, and obtaining promising results., The work of Simona Frenda and Paolo Rosso was partially funded by the Spanish MINECO under the research project SomEMBED (TIN2015-71147-C2-1-P). We also thank the support of CONACYT-Mexico (project FC-2410).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

DomainSenticNet: An Ontology and a Methodology Enabling Domain-aware Sentic Computing

  • Distante, Damiano
  • Faralli, Stefano
  • Rittinghaus, Steve
  • Samsami, Nima
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] In recent years, SenticNet and OntoSenticNet have represented important developments in the novel interdisciplinary field of research known as sentic computing, enabling the development of a variety of Sentic applications. In this paper, we propose an extension of the OntoSenticNet ontology, named DomainSenticNet, and contribute an unsupervised methodology to support the development of domain-aware Sentic applications. We developed an unsupervised methodology that, for each concept in OntoSenticNet, mines semantically related concepts from WordNet and Probase knowledge bases and computes domain distributional information from the entire collection of Kickstarter domain-specific crowdfunding campaigns. Subsequently, we applied DomainSenticNet to a prototype tool for Kickstarter campaign authoring and success prediction, demonstrating an improvement in the interpretability of sentiment intensities. DomainSenticNet is an extension of the OntoSenticNet ontology that integrates each of the 100,000 concepts included in OntoSenticNet with a set of semantically related concepts and domain distributional information. The defined unsupervised methodology is highly replicable and can be easily adapted to build similar domain-aware resources from different domain corpora and external knowledge bases. Used in combination with OntoSenticNet, DomainSenticNet may favor the development of novel hybrid aspect-based sentiment analysis systems and support further research on sentic computing in domain-aware applications., The work of Paolo Rosso was partially funded by the Spanish MICINN under the project PGC2018-096212-B-C31.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Profiling hate speech spreaders on twitter task at PAN 2021

  • Rangel, Francisco
  • Peña-Sarracén, Gretel Liz de la
  • Chulvi-Ferriols, María Alberta
  • Fersini, Elisabetta
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] This overview presents the Author Profiling shared task at PAN 2021. The focus of this year¿s task is on determining whether or not the author of a Twitter feed is keen to spread hate speech. The main aim is to show the feasibility of automatically identifying potential hate speech spreaders on Twitter. For this purpose a corpus with Twitter data has been provided, covering the English and Spanish languages. Altogether, the approaches of 66 participants have been evaluated., First of all, we thank the participants: again 66 this year, as the previous year on Profiling Fake News Spreaders! We have to thank also Martin Potthast, Matti Wiegmann, Nikolay Kolyada, and Magdalena Anna Wolska for their technical support with the TIRA platform. We thank Symanto for sponsoring again the award for the best performing system at the author profiling shared task. The work of Francisco Rangel was partially funded by the Centre for the Development of Industrial Technology (CDTI) of the Spanish Ministry of Science and Innovation under the research project IDI-20210776 on Proactive Profiling of Hate Speech Spreaders - PROHATER (Perfilador Proactivo de Difusores de Mensajes de Odio). The work of the researchers from Universitat Politècnica de València was partially funded by the Spanish MICINN under the project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31), and by the Generalitat Valenciana under the project DeepPattern (PROMETEO/2019/121). This article is also based upon work from the Dig-ForAsp COST Action 17124 on Digital Forensics: evidence analysis via intelligent systems and practices, supported by European Cooperation in Science and Technology.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Multilingual Stance Detection in Social Media Political Debates

  • Lai, Mirko
  • Cignarella, Alessandra Teresa
  • Hernandez-Farias, Delia Irazu
  • Bosco, Cristina
  • Patti, Viviana
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Stance Detection is the task of automatically determining whether the author of a text is in favor, against, or neutral towards a given target. In this paper we investigate the portability of tools performing this task across different languages, by analyzing the results achieved by a Stance Detection system (i.e. MultiTACOS) trained and tested in a multilingual setting. First of all, a set of resources on topics related to politics for English, French, Italian, Spanish and Catalan is provided which includes: novel corpora collected for the purpose of this study, and benchmark corpora exploited in Stance Detection tasks and evaluation exercises known in literature. We focus in particular on the novel corpora by describing their development and by comparing them with the benchmarks. Second, MultiTACOS is applied with different sets of features especially designed for Stance Detection, with a specific focus to exploring and combining both features based on the textual content of the tweet (e.g., style and affective load) and features based on contextual information that do not emerge directly from the text. Finally, for better highlighting the contribution of the features that most positively affect system performance in the multilingual setting, a features analysis is provided, together with a qualitative analysis of the misclassified tweets for each of the observed languages, devoted to reflect on the open challenges., Cristina Bosco and Viviana Patti are partially supported by Progetto di Ateneo/CSP 2016 (Immigrants, Hate and Prejudice in Social Media, S1618_L2_BOSC_01). The work of Paolo Rosso was partially funded bythe Spanish MICINN under the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

IberLEF 2021 Overview: Natural Language Processing for Iberian Languages

  • Gonzalo, Julio
  • Montes-y-Gómez, Manuel
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] IberLEF is a comparative evaluation campaign for Natural Language Processing Systems in Spanish and other Iberian languages. Its goal is to encourage the research community to organize competitive text processing, understanding and generation tasks in order to define new research challenges and set new state-of-the-art results in those languages. This paper summarizes the evaluation activities carried out in IberLEF 2021, which included twelve tasks dealing with emotions, stance and opinions, harmful information, health-related information extraction and discovery, humor and irony, and lexical acquisition. Overall, IberLEF activities were a remarkable collective effort involving 359 researchers from 22 countries in Europe, Asia and the Americas., The authors of this overview have been supported by the Spanish Government, Ministry of Science and Innovation, via research grants MISMIS (PGC2018- 096212-B), MISMIS-BIAS (PGC2018-096212-B-C32) and MISMISFAKEnHATE (PGC2018-096212-B-C31); and by CONACyT-Mexico project CB-2015-01- 257383 and the thematic networks program (Language Technologies Thematic Network).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

Modernizing historical documents: A user Study

  • Domingo-Ballester, Miguel|||0000-0002-7910-4536
  • Casacuberta Nolla, Francisco|||0000-0002-8497-5598
[EN] Accessibility to historical documents is mostly limited to scholars. This is due to the language barrier inherent in human language and the linguistic properties of these documents. Given a historical document, modernization aims to generate a new version of it, written in the modern version of the document's language. Its goal is to tackle the language barrier, decreasing the comprehension difficulty and making historical documents accessible to a broader audience. In this work, we proposed a new neural machine translation approach that profits from modern documents to enrich its systems. We tested this approach with both automatic and human evaluation, and conducted a user study. Results showed that modernization is successfully reaching its goal, although it still has room for improvement., The authors wish to thank the anonymous reviewers for their careful reading and in-depth criticisms and suggestions. The research leading to these results has received funding from the European Union through Programa Operativo del Fondo Europeo de Desarrollo Regional (FEDER) from Comunitat Valenciana (2014-2020) under project Sistemas de frabricacion inteligentes para la industria 4.0 (grant agreement IDIFEDER/2018/025); from Ministerio de Economia y Competitividad (MINECO) under project MISMIS-FAKEnHATE (grant agreement PGC2018-096212-B-C31); from Fundacion BBVA under project Carabela (grant agreement CARABELA); and from Generalitat Valenciana (GVA) under project DeepPattern (grant agreement PROMETEO/2019/121). We gratefully acknowledge the support of NVIDIA Corporation with the donation of a GPU used for part of this research, and Andres Trapiello and Ediciones Destino for granting us permission to use their book in our research. Additionally, we would like to thank all the volunteers that took part in the user study, and the scholars from Prolope that took part in the human evaluation.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

Deep Modeling of Latent Representations for Twitter Profiles on Hate Speech Spreaders Identification Task

  • Labadie Tamayo, Roberto
  • Castro Castro, Daniel
  • Ortega-Bueno, Reynier
[EN] In this paper, we describe the system proposed by UO-UPV team for addressing the task Profiling Hate Speech Spreaders on Twitter shared at PAN 2021. The system relies on a modular architecture, combining Deep Learning models with an introduced variant of the Impostor Method (IM). It receives a single profile composed of a fixed quantity of tweets. These posts are encoded as dense feature vectors using a fine-tuned transformer model and later combined to represent the whole profile. For classifying a new profile as hate speech spreader or not, it is compared by a similarity function with the Impostor Method with respect to random sampled prototypical profiles. In the final evaluation phase, our model achieved 74% and 82% of accuracy for English and Spanish languages respectively, ranking our team at 2¿¿ position and giving a starting point for further improvements., The work of the third author was in the framework of the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31), funded by Spanish Ministry of Science and Innovation, and DeepPattern (PROMETEO/2019/121), funded by the Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Contribución de congreso (ConferenceOutput)

UPV at CheckThat! 2021: Mitigating Cultural Differences for Identifying Multilingual Check-worthy Claims

  • Baris-Schlicht, Ipek
  • Magnossao de Paula, Angel Felipe
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Identifying check-worthy claims is often the first step of automated fact-checking systems. Tackling this task in a multilingual setting has been understudied. Encoding inputs with multilingual text representations could be one approach to solve the multilingual check-worthiness detection. However, this approach could suffer if cultural bias exists within the communities on determining what is check-worthy. In this paper, we propose a language identification task as an auxiliary task to mitigate unintended bias. With this purpose, we experiment joint training by using the datasets from CLEF-2021 CheckThat!, that contain tweets in English, Arabic, Bulgarian, Spanish and Turkish. Our results show that joint training of language identification and check-worthy claim detection tasks can provide performance gains for some of the selected languages., The work of P. Rosso was partially funded by the Spanish Ministry of Science and Innovation under the research project MISMIS-FAKEnHATE on MISinformation and MIScommunication in social media: FAKE news and HATE speech (PGC2018-096212-B-C31).
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publication Artículo científico (JournalArticle)

The Impact of Emotional Signals on Credibility Assessment

  • Anastasia Giachanou
  • Crestani, Fabio
  • Rosso, Paolo|||0000-0002-8922-1242
[EN] Fake news is considered one of the main threats of our society. The aim of fake news is usually to confuse readers and trigger intense emotions to them in an attempt to be spread through social networks. Even though recent studies have explored the effectiveness of different linguistic patterns for fake news detection, the role of emotional signals has not yet been explored. In this paper, we focus on extracting emotional signals from claims and evaluating their effectiveness on credibility assessment. First, we explore different methodologies for extracting the emotional signals that can be triggered to the users when they read a claim. Then, we present emoCred, a model that is based on a long-short term memory model that incorporates emotional signals extracted from the text of the claims to differentiate between credible and non-credible ones. In addition, we perform an analysis to understand which emotional signals and which terms are the most useful for the different credibility classes. We conduct extensive experiments and a thorough analysis on real-world datasets. Our results indicate the importance of incorporating emotional signals in the credibility assessment problem., Generalitat Valenciana, Grant/Award Number: DeepPattern (PROMETEO/2019/121); Ministerio de Ciencia e Innovacion, Grant/Award Number: PGC2018-096212-B-C31; Schweizerischer Nationalfonds zur Forderung der Wissenschaftlichen Forschung, Grant/Award Number: P2TIP2_181441
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Minería de argumentación en el Referéndum del 1 de Octubre de 2017, Argument Mining in the October 1, 2017 Referendum

  • Esteve, Marcos
  • Casacuberta Nolla, Francisco
  • Rosso, Paolo
La minería de argumentación permite, mediante herramientas software, obtener cuáles son los argumentos que expresan los autores en un determinado texto. En este artículo se pretende realizar un análisis de la argumentación expresada por los usuarios en Twitter en relación al referendum del 1 de octubre de 2017. Se utilizará para ello, el dataset MultiStanceCat proporcionado en la tarea organizada en el IberEval 2018. Dado que las herramientas de minería de argumentación trabajan en su mayoría en inglés, será necesario construir un sistema de traducción neuronal con postedición que permita realizar una traducción de los tweets del español y el catalán al inglés. Los resultados al realizar la minería de argumentación sobre los tweets traducidos ha demostrado obtener un porcentaje muy reducido de argumentación en todas las comunidades.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Categorizing Misogynistic Behaviours in Italian, English and Spanish Tweets, Categorización de comportamientos misóginos en tweets en italiano, inglés y español

  • Lazzardi, Silvia
  • Patti, Viviana
  • Rosso, Paolo
Misogyny is a multifaceted phenomenon and can be linguistically manifested in numerous ways. The evaluation campaigns of EVALITA and IberEval in 2018 proposed a shared task of Automatic Misogyny Identification (AMI) based on Italian, English and Spanish tweets. Since the participating teams' results were pretty low in the misogynistic behaviour categorization, the aim of this study is to investigate the possible causes. We measured the overlap and the homogeneity of the clusters by varying the number of categories. This experiment showed that the clusters overlap. Finally, we tested several machine learning models both using the original data sets and merging together some categories according to their overlap, obtaining an increase in terms of macro F1.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Masking and BERT-based Models for Stereotype Identification, Modelos Basados en Enmascaramiento y en BERT para la Identificación de Estereotipos

  • Sánchez-Junquera, Javier
  • Rosso, Paolo
  • Montes y Gómez, Manuel
  • Chulvi, Berta
Stereotypes about immigrants are a type of social bias increasingly present in the human interaction in social networks and political speeches. This challenging task is being studied by computational linguistics because of the rise of hate messages, offensive language, and discrimination that many people receive. In this work, we propose to identify stereotypes about immigrants using two different explainable approaches: a deep learning model based on Transformers; and a text masking technique that has been recognized by its capabilities to deliver good and human-understandable results. Finally, we show the suitability of the two models for the task and offer some examples of their advantages in terms of explainability.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

Overview of DETOXIS at IberLEF 2021: DEtection of TOXicity in comments In Spanish, Visión general de DETOXIS en IberLEF 2021: DEtección de TOXicidad en comentarios En Español

  • Taulé Delor, Mariona
  • Ariza, Alejandro
  • Nofre, Montserrat
  • Amigó Cabrera, Enrique
  • Rosso, Paolo
In this paper we present the DETOXIS task, DEtection of TOxicity in comments In Spanish, which took place as part of the IberLEF 2021 Workshop on Iberian Languages Evaluation Forum at the SEPLN 2021 Conference. We describe the NewsCom-TOX dataset used for training and testing the systems, the metrics applied for their evaluation and the results obtained by the submitted approaches. We also provide an error analysis of the results of these systems.
RUA. Repositorio Institucional de la Universidad de Alicante
Publication Otros (Other)

MISMIS: Misinformation and Miscommunication in social media: aggregating information and analysing language, MISMIS: Desinformación y agresividad en los medios de comunicación social: agregando información y analizando el lenguaje

  • Rosso, Paolo
  • Casacuberta Nolla, Francisco
  • Gonzalo Arroyo, Julio
  • Plaza Morales, Laura
  • Carrillo-de-Albornoz, Jorge
  • Amigó Cabrera, Enrique
  • Verdejo Maillo, María Felisa
  • Taulé Delor, Mariona
  • Salamó Llorente, Maria
  • Martí Antonín, Maria Antònia
The general objectives of the project are to address and monitor misinformation (biased and fake news) and miscommunication (aggressive language and hate speech) in social media, as well as to establish a high quality methodological standard for the whole research community (i) by developing rich annotated datasets, a data repository and online evaluation services; (ii) by proposing suitable evaluation metrics; and (iii) by organizing evaluation campaigns to foster research on the above issues.