Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 10
Encontrada(s) 1 página(s)

e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Contribución de congreso (ConferenceOutput)

Unique on Facebook: Formulation and Evidence of (Nano)targeting Individual Users with non-PII Data

  • González Cabañas, José
  • Cuevas Rumín, Ángel
  • Cuevas Rumín, Rubén
  • López-Fernández, Juan
  • García, David
Proceedings of: ACM Internet Measurement Conference (IMC '21), November 2-4, 2021, Virtual Event, USA., The privacy of an individual is bounded by the ability of a third party to reveal their identity. Certain data items such as a passport ID or a mobile phone number may be used to uniquely identify a person. These are referred to as Personal Identifiable Information (PII) items. Previous literature has also reported that, in datasets including millions of users, a combination of several non-PII items (which alone are not enough to identify an individual) can uniquely identify an individual within the dataset. In this paper, we define a data-driven model to quantify the number of interests from a user that make them unique on Facebook. To the best of our knowledge, this represents the first study of individuals’ uniqueness at the world population scale. Besides, users’ interests are actionable non-PII items that can be used to define ad campaigns and deliver tailored ads to Facebook users. We run an experiment through 21 Face-book ad campaigns that target three of the authors of this paper to prove that, if an advertiser knows enough interests from a user, the Facebook Advertising Platform can be systematically exploited to deliver ads exclusively to a specific user. We refer to this practice as nanotargeting. Finally, we discuss the harmful risks associated with nanotargeting such as psychological persuasion, user manipulation, or blackmailing, and provide easily implementable countermea-sures to preclude attacks based on nanotargeting campaigns on Facebook., This research received funding from the European Union’s Horizon 2020 innovation action programme under the PIMCITY project (Grant 871370) and the TESTABLE project (Grant 101019206); the Ministerio de Economía, Industria y Competitividad, Spain, and the European Social Fund(EU), under the Ramón y Cajal programme (Grant RyC-2015-17732); the Ministerio de Educación, Cultura y Deporte, Spain, through the FPU programme (Grant FPU16/05852); the Agencia Estatal de Investigación (AEI) under the ACHILLES project (Grant PID2019-104207RB-I00/AEI/10.13039/501100011033); the Community of Madrid synergic project EMPATIA-CM (Grant Y2018/TCS-5046); the Fundación BBVA under the project AERIS; and the Vienna Science and Technology Fund through the project “Emotional Well-Being in the Digital Society” (Grant VRG16-005).
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Estimating ideology and polarization in European countries using Facebook data

  • Caravaca Crespo, Francisco
  • González Cabañas, José
  • Cuevas Rumín, Ángel
  • Cuevas Rumín, Rubén
Researchers have studied political ideology and polarization in many different contexts since their effects are usually closely related to aspects and actions of individuals and societies. Hence, being able to estimate and measure the changes in political ideology and polarization is crucial for researchers, stakeholders, and the general public. In this paper, we model the ideology and polarization of 28 countries (the 27 EU member states plus the UK) using Facebook public posts from political parties’ Facebook pages. We collected a three-year dataset from 2019 to 2021 with information from 234 political parties’ Facebook pages and took advantage of the EU parliament elections of May 2019 to create our models. Our methodology works across 28 countries and benefits from being a low-cost running process that measures ideology and polarization at a high-resolution time scale. The results show our models are pretty accurate when validating them against 19 individual countries’ elections as ground truth. Moreover, to make our results available to the research community, stakeholders, and individuals interested in politics, the last contribution of our paper is a website including detailed information about the political parties in our dataset. It also includes the temporal evolution of our ideology and polarization estimations. Therefore, our work delivers a novel tool that uses Facebook public data to create country metrics useful for different purposes. To the best of our knowledge, there is no prior work in the literature offering a solution that measures the ideology and polarization of all EU + UK countries., This research received funding from the Madrid Government (Comunidad de Madrid-Spain) under the Multiannual Agreement with UC3M ("Fostering Young Doctors Research", UE-MEASURE-CM-UC3M) the agreement between the Community of Madrid and the Universidad Carlos III de Madrid. The Community of Madrid and Universidad Carlos III de Madrid for the funding of research projects on SARS-CoV-2 and COVID-19 disease, project name "Multi-source and multi-method prediction to support COVID-19 policy decision making", which was supported with REACT-EU funds from the European regional development fund "a way of making Europe". The European Union's Horizon 2020 programme under TESTABLE project (Grant 101019206). The Agencia the Estatal de Investigación (AEI) under the ENTRUDIT project (Grant TED2021-130118B-I00). The Fundación BBVA under the project AERIS.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Contribución de congreso (ConferenceOutput)

Analysis and implementation of nanotargeting on LinkedIn based on publicly available non-PII

  • Merino Hernández, Ángel
  • González Cabañas, José
  • Cuevas Rumín, Ángel
  • Cuevas Rumín, Rubén
Proceedings of: 2024 CHI Conference on Human Factors in Computing Systems (CHI'24), Honolulu, USA, 11-16 May 2024, The literature has shown that combining a few non-Personal Identifiable Information (non-PII) is enough to make a user unique in a dataset including millions of users. This work demonstrates that a combination of a few non-PII items can be activated to nanotarget users. We demonstrate that the combination of the location and 5 rare (13 random) skills in a LinkedIn profile is enough to become unique in a user base of ∼ 970M users with a probability of 75%. The novelty is that these attributes are publicly accessible to anyone registered on LinkedIn and can be activated through advertising campaigns. We ran an experiment configuring ad campaigns using the location and skills of three of the paper’s authors, demonstrating how all the ads using ≥ 13 skills were delivered exclusively to the targeted user. We reported this vulnerability to LinkedIn, which initially ignored the problem, but fixed it as of November 2023., This work has been partially funded by the following projects: the project TESTABLE (Grant 101019206) funded by European Union’s Horizon 2020 programme; the project AUDINT (Grant TED2021-132076B-I00) funded by the MCIN/AEI/ 10.13039/501100011033 and the EU FEDER funds; the A2 PRIVCOMP funded by the Ministerio de Asuntos Económicos y Transformación Digital and the European Union-NextGenerationEU.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Contribución de congreso (ConferenceOutput)

Overprofiling Analysis on Major Internet Players

  • Caravaca Crespo, Francisco
  • González Cabañas, José
  • Cuevas Rumín, Ángel
  • Cuevas Rumín, Rubén
Many Internet services obtain their revenue through the delivery of online advertisements based on the commercial exploitation of users’ profiles. The accuracy and size of these profiles have important implications in terms of advertisers’ campaign performance and users’ privacy. Despite the importance of auditing the profiling accuracy, very little effort has been devoted both in industry and academia. This paper presents the most comprehensive auditing effort to understand the profiling accuracy of four major online advertising platforms: Google, Facebook, Twitter, and LinkedIn. Our work unveils that less than 50% of the assigned interests are relevant. Moreover, platforms can distinguish what interests within the assigned ones are more relevant but hide this information from users and advertisers. Finally, we have proposed a very simple solution that only uses 25 general interests per user. This proposal outperforms all the analyzed platforms in terms of profile accuracy while improving users’ privacy., This research has received funding from: the project ENTRUDIT (Grant TED2021-130118B-I00) funded by the MCIN/AEI/10.13039/ 501100011033 and the NextGeneration EU/PRTR funds; the European Union’s Horizon 2020 innovation action programme under the TESTABLE project (Grant 101019206); and the Ministerio de Asuntos Económicos y Transformacion Digital and the European Union-NextGenerationEU through the project PRTIACLOUD; the project PRTR-INCIBE ANTICIPA
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

Expanding the measurement of culture with a sample of two billion humans

  • Obradovich, Nick
  • Ozak, Omer
  • Martín Martínez, Ignacio
  • Ortuño Ortín, Ignacio
  • Awad, Edmond
  • Cebrian, Manuel
  • Cuevas Rumín, Rubén
  • Desmet, Klaus
  • Rahwan, Iyad
  • Cuevas Rumín, Ángel
Culture has played a pivotal role in human evolution. Yet, the ability of social scientists to study culture is limited by the currently available measurement instruments. Scholars of culture must regularly choose between scalable but sparse survey-based methods or restricted but rich ethnographic methods. Here, we demonstrate that massive online social networks can advance the study of human culture by providing quantitative, scalable and high-resolution measurement of behaviourally revealed cultural values and preferences. We employ data across nearly 60 000 topic dimensions drawn from two billion Facebook users across 225 countries and territories. We first validate that cultural distances calculated from this measurement instrument correspond to traditional survey-based and objective measures of cross-national cultural differences. We then demonstrate that this expanded measure enables rich insight into the cultural landscape globally at previously impossible resolution. We analyse the importance of national borders in shaping culture and compare subnational divisiveness with gender divisiveness across countries. Our measure enables detailed investigation into the geopolitical stability of countries, social cleavages within small- and large-scale human groups, the integration of migrant populations and the disaffection of certain population groups from the political process, among myriad other potential future applications., Á.C. acknowledges funding from the European Union’s Horizon 2020 innovation action program under grant agreement no. 101019206 (TESTABLE project); and the Madrid Government (Comunidad de Madrid-Spain) under the Multiannual Agreement with UC3M (‘Fostering Young Doctors' Research’, UE-MEASURE-CM-UC3M). R.C. acknowledges funding from H2020 EU Project PIMCITY (grant no. 871370) and the Taptap Digital-UC3M Chair in Advanced AI and Data Science applied to Advertising and Marketing. Á.C., R.C, K.D, I.O.-O. and Ö.Ö. acknowledge funding from ECO2013-42710-P, MDM 2014-0431 and Fundacion BBVA. I.M. acknowledges funding from the Spanish Ministry of education with the FPU programme (FPU15/03518).
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

A new methodology to measure faultlines at scale leveraging digital traces

  • Mehrjoo, Amir Reza
  • Cuevas Rumín, Rubén
  • Cuevas Rumín, Ángel
The definition of society is tight with human group-level behavior. Group faultlines defined as hypothetical lines splitting groups into homogeneous subgroups based on members' attributes have been proposed as a theoretical method to identify conflicts within groups. For instance, crusades and women's rights protests are the consequences of strong faultlines in societies with diverse cultures. Measuring the presence and strength of faultlines represents an important challenge. Existing literature resorts in questionnaires as traditional tool to find group-level behavioral attributes and thus identify faultlines. However, questionnaire data usually come with limitations and biases, especially for large-scale human group-level research. On top of that, questionnaires limit faultline research due to the possibility of dishonest answers, unconscientious responses, and differences in understanding and interpretation. In this paper, we propose a new methodology for measuring faultlines in large-scale groups, which leverages data readily available from online social networks' marketing platforms. Our methodology overcomes the limitations of traditional methods to measure group-level attributes and group faultlines at scale. To prove the applicability of our methodology, we analyzed the faultlines between people living in Spain, grouped by geographical regions. We collected data on 67,270 interest topics from Facebook users living in Spain, France, Germany, Greece, Italy, Portugal, and the United Kingdom. We computed existing metrics to measure faultlines' distance and strenght using our data to identify potential faultlines existing among Spanish regions. The results reveal that the strongest faultlines in Spain belong to Spanish Islands (the Canary Islands and the Balearic Islands), Catalonia, and Basque regions. These findings are aligned with the historical secessionist movements and cultural diversity reports supporting the validity of our methodology., This research received funding from the European Union's Horizon 2020 innovation action program under the PIMCITY (Grant 871370) and TESTABLE (101019206) projects. The Agencia Estatal de Investigación (AEI) under the ACHILLES project (Grant PID2019-104207RB-I00/AEI/10.13039/501100011033). The Madrid Government (Comunidad de Madrid-Spain) under the Multiannual Agreement with UC3M ("Fostering Young Doctors Research"). Finally, the Fundación BBVA under the project AERIS.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Contribución de congreso (ConferenceOutput)

Analysis of User Uniqueness on LinkedIn Based on Publicly Available Non-PII

  • Merino Hernández, Ángel
  • González Cabañas, José
  • Cuevas Rumín, Ángel
  • Cuevas Rumín, Rubén
Proceedings of: The 2023 ACM Internet Measurement Conference (IMC '23), Montreal, Canada, 24-26 Oct. 2023, The literature has shown combining a few non-Personal Identifiable Information (non-PII) is enough to make a user unique in a dataset including millions of users. In this work, we demonstrate that the combination of the location and 6 rare (14 random) skills in a LinkedIn profile is enough to become unique in a user base of ~800M users with a probability of 75%. The novelty is these attributes are publicly accessible to anyone registered on LinkedIn and could be activated through advertising campaigns.
e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid
Publicación Artículo científico (JournalArticle)

A deep dive into the accuracy of IP geolocation databases and its impact on online advertising

  • Callejo Pinardo, Patricia
  • Gramaglia, Marco
  • Cuevas Rumín, Rubén
  • Cuevas Rumín, Ángel
The quest for every time more personalized Internet experience relies on the enriched contextual information about each user. Online advertising also follows this approach. Among the context information that advertising stakeholders leverage, location information is certainly one of them. However, when this information is not directly available from the end users, advertising stakeholders infer it using geolocation databases, matching IP addresses to a position on earth. The accuracy of this approach has often been questioned in the past: however, the reality check on an advertising stakeholder shows that this technique accounts for a large fraction of the served advertisements. In this paper, we revisit the work in the field, that is mostly from almost one decade ago, through the lenses of big data. More specifically, we, i) benchmark two commercial Internet geolocation databases, evaluate the quality of their information using a ground-truth database of user positions containing over 2 billion samples, ii) analyze the internals of these databases, devising a theoretical upper bound for the quality of the Internet geolocation approach, and iii) we run an empirical study that unveils the monetary impact of this technology by considering the costs associated with a real-world ad impressions dataset., This work was supported in part by European Union's Horizon 2020 innovation action programme under the PIMCITY Project under Grant 871370, in part by TESTABLE Project under Grant 101019206, in part by Agencia Estatal de Investigacion (AEI) under the ACHILLES Project under Grants PID2019-104207RB-I00/AEI/10.13039/501100011033, in part by the Spanish Ministry of Economic Affairs and Digital Transformation and European Union-Next GenerationEU through the UNICO 5G I+D 6G-RIEMANN-FR, in part by the agreement between the Community of Madrid and the Universidad Carlos III de Madrid for the funding of research projects on SARS-CoV-2 and COVID-19 disease, through project name Multi-source and multi-method prediction to support COVID-19 policy decision making, which was supported with REACT-EU funds from the European regional development fund a way of making Europe, and in part by TAPTAP-UC3M Chair in advanced AI and Data Science applied to advertising and marketing.
Repositori Obert UdL
Publicación Artículo científico (JournalArticle)

Certified Adversarial Robustness of End-to-End Malware Detectors via (De)Randomized Smoothing

  • Gibert, Daniel
  • Demetrio, Luca
  • Zizzo, Giulio
  • Le, Quan
  • Planes Cid, Jordi
  • Biggio, Battista
End-to-end machine learning malware detectors are vulnerable to adversarial EXEmples, carefully-crafted malicious programs that evade detection through minimal perturbations. Such attacks typically operate by either replacing unused content (patch attacks), or injecting new patterns (content-injection attacks). To counter these attacks, recent research has focused on certification methods for end-to-end models that aim to prove robustness guarantees within bounded perturbation sizes. However, existing approaches (i) are not robust against content-injection manipulations and (ii) provide only probabilistic guarantees for perturbations that are negligible relative to the overall program size. Hence, in this article we address these limitations through a novel deterministic certification schema based on (de)randomized smoothing. Our defense splits each executable into non-overlapping chunks and classifies them independently. The final decision is obtained via majority voting across all chunks, ensuring that localized modifications, such as injected or patched code, influence only a limited subset of chunks and have minimal impact in the overall classification. This design guarantees that each chunk either contains or does not contain an adversarial perturbation, enabling us to (i) handle manipulations occurring at arbitrary locations within the program, and (ii) compute deterministic estimates of the perturbation magnitude required to evade detection. We demonstrate the effectiveness of our certification schema through extensive experimental analysis, comparing our defense against a range of state-of-the-art attacks and defenses. The results show that our approach achieves unmatched robustness across all tested attack scenarios, substantially outperforming competing defenses., D. Gibert was supported by grant RYC2023-043607-I funded by MICIU/AEI/10.13039/501100011033 and FSE+. This project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under the Marie Skodowska-Curie Grant Agreement”(as per Article 29.4 of the grant agreement). We would like to thank Cormac Doherty and UCD’s Centre for Cybersecurity and Cybercrime Investigation for their support. This project has also been partially support by Fondazione di Sardegna under the project “TrustML: Towards ML that Humans Can Trust’’, CUP: F73C22001320007; and European Union’s Horizon Europe research and innovation program under the project ELSA, grant agreement No 101070617; and European Union’s Horizon 2020 research and innovation programme under project TESTABLE, grant agreement No 101019206; and SERICS (PE00000014) under the MUR National Recovery and Resilience Plan funded by the European Union – NextGenerationEU, and by MCIN/AEI/10.13039/501100011033/FEDER, UE under the project PID2022-139835NB-C22. This research project was made possible through the access granted by the Galician Supercomputing Center (CESGA) to its supercomputing infrastructure. The supercomputer FinisTerrae III and its permanent data storage system have been funded by the NextGeneration EU 2021 Recovery, Transformation and Resilience Plan, ICT2021-006904, and also from the Pluriregional Operational Programme of Spain 2014-2020 of the European Regional Development Fund (ERDF), ICTS-2019-02-CESGA-3, and from the State Programme for the Promotion of Scientific and Technical Research of Excellence of the State Plan for Scientific and Technical Research and Innovation 2013-2016 State subprogramme for scientific and technical infrastructures and equipment of ERDF, CESG15-DE-3114
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Certified Adversarial Robustness of End-to-End Malware Detectors via (De)Randomized Smoothing

  • Gibert, Daniel
  • Demetrio, Luca
  • Zizzo, Giulio
  • Le, Quan
  • Planes Cid, Jordi
  • Biggio, Battista
End-to-end machine learning malware detectors are vulnerable to adversarial EXEmples, carefully-crafted malicious programs that evade detection through minimal perturbations. Such attacks typically operate by either replacing unused content (patch attacks), or injecting new patterns (content-injection attacks). To counter these attacks, recent research has focused on certification methods for end-to-end models that aim to prove robustness guarantees within bounded perturbation sizes. However, existing approaches (i) are not robust against content-injection manipulations and (ii) provide only probabilistic guarantees for perturbations that are negligible relative to the overall program size. Hence, in this article we address these limitations through a novel deterministic certification schema based on (de)randomized smoothing. Our defense splits each executable into non-overlapping chunks and classifies them independently. The final decision is obtained via majority voting across all chunks, ensuring that localized modifications, such as injected or patched code, influence only a limited subset of chunks and have minimal impact in the overall classification. This design guarantees that each chunk either contains or does not contain an adversarial perturbation, enabling us to (i) handle manipulations occurring at arbitrary locations within the program, and (ii) compute deterministic estimates of the perturbation magnitude required to evade detection. We demonstrate the effectiveness of our certification schema through extensive experimental analysis, comparing our defense against a range of state-of-the-art attacks and defenses. The results show that our approach achieves unmatched robustness across all tested attack scenarios, substantially outperforming competing defenses., D. Gibert was supported by grant RYC2023-043607-I funded by MICIU/AEI/10.13039/501100011033 and FSE+. This project has received funding from the European Union’s Horizon 2020 Research and Innovation programme under the Marie Skodowska-Curie Grant Agreement”(as per Article 29.4 of the grant agreement). We would like to thank Cormac Doherty and UCD’s Centre for Cybersecurity and Cybercrime Investigation for their support. This project has also been partially support by Fondazione di Sardegna under the project “TrustML: Towards ML that Humans Can Trust’’, CUP: F73C22001320007; and European Union’s Horizon Europe research and innovation program under the project ELSA, grant agreement No 101070617; and European Union’s Horizon 2020 research and innovation programme under project TESTABLE, grant agreement No 101019206; and SERICS (PE00000014) under the MUR National Recovery and Resilience Plan funded by the European Union – NextGenerationEU, and by MCIN/AEI/10.13039/501100011033/FEDER, UE under the project PID2022-139835NB-C22. This research project was made possible through the access granted by the Galician Supercomputing Center (CESGA) to its supercomputing infrastructure. The supercomputer FinisTerrae III and its permanent data storage system have been funded by the NextGeneration EU 2021 Recovery, Transformation and Resilience Plan, ICT2021-006904, and also from the Pluriregional Operational Programme of Spain 2014-2020 of the European Regional Development Fund (ERDF), ICTS-2019-02-CESGA-3, and from the State Programme for the Promotion of Scientific and Technical Research of Excellence of the State Plan for Scientific and Technical Research and Innovation 2013-2016 State subprogramme for scientific and technical infrastructures and equipment of ERDF, CESG15-DE-3114