Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

ENSA dataset: a dataset of songs by non-superstar artists tested with an emotional analysis based on time-series

  • Ospitia-Medina, Yesid
  • Beltrán, José Ramón
  • Baldassarri, Sandra
This paper presents a novel dataset of songs by non-superstar artists in which a set of musical data is collected, identifying for each song its musical structure, and the emotional perception of the artist through a categorical emotional labeling process. The generation of this preliminary dataset is motivated by the existence of biases that have been detected in the analysis of the most used datasets in the field of emotion-based music recommendation. This new dataset contains 234 min of audio and 60 complete and labeled songs. In addition, an emotional analysis is carried out based on the representation of dynamic emotional perception through a time-series approach, in which the similarity values generated by the dynamic time warping (DTW) algorithm are analyzed and then used to implement a clustering process with the K-means algorithm. In the same way, clustering is also implemented with a Uniform Manifold Approximation and Projection (UMAP) technique, which is a manifold learning and dimension reduction algorithm. The algorithm HDBSCAN is applied for determining the optimal number of clusters. The results obtained from the different clustering strategies are compared and, in a preliminary analysis, a significant consistency is found between them. With the findings and experimental results obtained, a discussion is presented highlighting the importance of working with complete songs, preferably with a well-defined musical structure, considering the emotional variation that characterizes a song during the listening experience, in which the intensity of the emotion usually changes between verse, bridge, and chorus.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Use case cards: a use case reporting framework inspired by the European AI Act

  • Hupont, Isabelle
  • Fernández-Llorca, David
  • Baldassarri, Sandra
  • Gómez, Emilia
Despite recent efforts by the Artificial Intelligence (AI) community to move towards standardised procedures for documenting models, methods, systems or datasets, there is currently no methodology focused on use cases aligned with the risk-based approach of the European AI Act (AI Act). In this paper, we propose a new framework for the documentation of use cases that we call use case cards, based on the use case modelling included in the Unified Markup Language (UML) standard. Unlike other documentation methodologies, we focus on the intended purpose and operational use of an AI system. It consists of two main parts: firstly, a UML-based template, tailored to allow implicitly assessing the risk level of the AI system and defining relevant requirements, and secondly, a supporting UML diagram designed to provide information about the system-user interactions and relationships. The proposed framework is the result of a co-design process involving a relevant team of EU policy experts and scientists. We have validated our proposal with 11 experts with different backgrounds and a reasonable knowledge of the AI Act as a prerequisite. We provide the 5 use case cards used in the co-design and validation process. Use case cards allows framing and contextualising use cases in an effective way, and we hope this methodology can be a useful tool for policy makers and providers for documenting use cases, assessing the risk level, adapting the different requirements and building a catalogue of existing usages of AI.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Wearable Device for Measuring EDA in Constant Alternating Current

  • Asiain Ansorena, David
  • Ponce de León Vázquez, Jesús
  • Beltrán Blázquez, José Ramón
This article introduces a novel wearable device, the multisensor wearable wristband (MsW2), designed to measure electrodermal activity (EDA) and skin temperature using constant ac current instead of the usual constant ac voltage measurement method. Adapted from a previous headband model, MsW2 has been optimized for wrist-based applications. Its efficacy was validated through experiments with 25 participants, comparing its EDA measurements with the Empatica E4 device. The results demonstrated high coherence between the two systems. Additionally, a simplified skin impedance model was proposed and validated, providing a reliable means of analyzing EDA data. The wearable demonstrates significant potential for applications in emotion detection, health monitoring, and more.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Design of an AI-based Language correction system to improve older adults’ interaction with voice assistants

  • Dragut, Adnana-Catrinel
  • Lacuesta, Raquel
  • Gallardo, Jesús
  • Buades-Rubio, Jose-Maria
Voice assistants are in full expansion due to the numerous benefits they have demonstrated. However, it has been found that some users exhibit specific speech patterns that impede effective interaction with them. In this paper, we propose a language correction system powered by generative artificial intelligence, which addresses potential communication errors in interactions with voice assistants. This system is designed to improve the comprehension of commands on Alexa devices, so its goal is to correct simple voice inputs. During these interactions, there are instances where Alexa misinterprets or fails to understand certain commands; our system tries to infer the word the user really wanted to express. In this study, we conducted two types of tests: one involving evaluation with users aged 65 and older, and another simulating speech impairments or disorders. The approach presented would increase the understanding of an average of 73% of those commands that Alexa did not understand on the interaction with older adults, enhancing interaction quality. Two main factors affect Alexa’s comprehension: low voice intensity and pronunciation difficulties related to users’ advanced age. In the second test, when we simulated some disorders, only 41% of the words and sentences read achieved a similarity score above 0.9. This indicates that the Alexa device struggles to comprehend inputs from users with significant speech impairments. Moreover, in our second test with the second prompt, the correction accuracy of our system rose again to 76%, obtaining good results addressing and solving users’ errors. This improvement highlights the potential of our system to facilitate the communication between conversational agents and users.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Applicability of process mining in usability tests: a case study for identifying user mental models in geospatial search engines

  • Herrera-Murillo, Dagoberto José
  • Abad-Power, Paloma
  • López-Pellicer, Francisco J.
  • Baldassarri, Sandra
  • Nogueras-Iso, Javier
A fundamental challenge in digital product design is ensuring that actual mental models of users align with the intended functionality of the system. When discrepancies arise, usability can suffer, leading to inefficient interactions and reduced adoption. Usability testing lets development teams identify design problems in digital products by collecting qualitative and quantitative information. However, this technique is often not able to provide a panoramic view of the interaction with the system, especially when dealing with complex interfaces such as those used in geospatial search engines and we need to analyse the quantitative data compiled during testing. This work proposes to actively incorporate process mining into usability testing. We discuss the implications of process mining on usability testing using a case study performed at the National Geographic Institute of Spain, where a new geospatial search engine was under development. Twenty-one participants, ranging from novice to expert users, were recruited to perform a search task using the new geospatial search engine. The findings reveal that the mental model of users leans towards the archetype of a regular search engine rather than fully utilising the geographic functionalities provided by the platform, as intended by its designers.