User modeling is a key topic in many applications, mainly social networks and information retrieval systems. To assess the effectiveness of a user modeling approach, its capability to classify personal characteristics (e.g., the gender, age, or consumption grade of the users) is evaluated. Due to the fact that some of the attributes to predict are multiclass (e.g., age usually encompasses multiple ranges), assessing fairness in user modeling becomes a challenge since most of the related metrics work with binary attributes. As a workaround, the original multiclass attributes are usually binarized to meet standard fairness metrics definitions where both the target class and sensitive attribute (such as gender or age) are binary. However, this alters the original conditions, and fairness is evaluated on classes that differ from those used in the classification. In this article, we extend the definitions of four existing fairness metrics (related to disparate impact and disparate mistreatment) from binary to multiclass scenarios, considering different settings where either the target class or the sensitive attribute includes more than two groups. Our work endeavors to bridge the gap between formal definitions and real use cases in bias detection. The results of the experiments, conducted on four real-world datasets by leveraging two state-of-the-art graph neural network-based models for user modeling, show that the proposed generalization of fairness metrics can lead to a more effective and fine-grained comprehension of disadvantaged sensitive groups and, in some cases, to a better analysis of machine learning models originally deemed to be fair. The source code and the preprocessed datasets are available at the following link: https://github.com/erasmopurif/toward-responsible-fairness-analysis.
@article{purificato_toward_2024,title={Toward a {Responsible} {Fairness} {Analysis}: {From} {Binary} to {Multiclass} and {Multigroup} {Assessment} in {Graph} {Neural} {Network}-{Based} {User} {Modeling} {Tasks}},volume={34},copyright={Creative Commons Attribution 4.0 International Licence (CC-BY)},issn={1572-8641},shorttitle={Toward a {Responsible} {Fairness} {Analysis}},url={https://doi.org/10.1007/s11023-024-09685-x},doi={10.1007/s11023-024-09685-x},language={en},number={3},urldate={2024-07-17},journal={Minds and Machines},abbr={MIND},author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},year={2024},keywords={User profiling, User modeling, Graph neural networks, Algorithmic fairness, Bias detection, Responsible artificial intelligence},pages={33},pdf={publications/2024_MIND.pdf},bibtex_show={true}}
arXiv
How Fair is Your Diffusion Recommender Model?
Malitesta, Daniele, Medda, Giacomo,
Purificato, Erasmo, Boratto, Ludovico, Malliaros, Fragkiskos D, Marras, Mirko, and De Luca, Ernesto William
Diffusion-based recommender systems have recently proven to outperform traditional generative recommendation approaches, such as variational autoencoders and generative adversarial networks. Nevertheless, the machine learning literature has raised several concerns regarding the possibility that diffusion models, while learning the distribution of data samples, may inadvertently carry information bias and lead to unfair outcomes. In light of this aspect, and considering the relevance that fairness has held in recommendations over the last few decades, we conduct one of the first fairness investigations in the literature on DiffRec, a pioneer approach in diffusion-based recommendation. First, we propose an experimental setting involving DiffRec (and its variant L-DiffRec) along with nine state-of-the-art recommendation models, two popular recommendation datasets from the fairness-aware literature, and six metrics accounting for accuracy and consumer/provider fairness. Then, we perform a twofold analysis, one assessing models’ performance under accuracy and recommendation fairness separately, and the other identifying if and to what extent such metrics can strike a performance trade-off. Experimental results from both studies confirm the initial unfairness warnings but pave the way for how to address them in future research directions.
@misc{malitesta2024fair,title={How Fair is Your Diffusion Recommender Model?},author={Malitesta, Daniele and Medda, Giacomo and Purificato, Erasmo and Boratto, Ludovico and Malliaros, Fragkiskos D and Marras, Mirko and De Luca, Ernesto William},year={2024},abbr={arXiv},archiveprefix={arXiv},primaryclass={cs.IR},bibtex_show={true},url={https://arxiv.org/abs/2409.04339},pdf={publications/2024_arXiv_2.pdf}}
UMAP
Paradigm Shifts in User Modeling: A Journey from Historical Foundations to Emerging Trends
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Adjunct Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization 2024
The presented tutorial aims to serve as a comprehensive roadmap for the UMAP community into the current user modeling research, focusing on the paradigm shifts that have transformed the research landscape in recent times. We will provide a complete overview of the large, long-standing, and ever-growing research fields of user modeling and user profiling, both from a historical and a technical point of view. We will then examine the definitions associated with each key term in this research domain, aiming to eliminate ambiguity and confusion in their usage. As the core of our tutorial, we present in-depth the paradigm shifts that have occurred in recent years, especially due to technological evolution, as well as the current research directions and novel trends in the field. In particular, we illustrate and discuss the advances in the following topics: implicit and explicit user profiling, user behavior modeling, user representation, and beyond-accuracy perspectives. The audience will be engaged in discussions during the whole presentation to foster the development of an interactive event. Detailed information and resources about the tutorial are available on the website: https://link.erasmopurif.com/tutorial-umap24.
@inproceedings{purificato2024paradigm,author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},title={Paradigm Shifts in User Modeling: A Journey from Historical Foundations to Emerging Trends},year={2024},isbn={9798400704666},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3631700.3653062},doi={10.1145/3631700.3653062},booktitle={Adjunct Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization},pages={13--16},numpages={4},keywords={Paradigm Shifts, User Modeling, User Profiling},location={Cagliari, Italy},series={UMAP Adjunct '24},abbr={UMAP},pdf={publications/2024_UMAP.pdf},bibtex_show={true},slides={https://www.slideshare.net/slideshow/paradigm-shifts-in-user-modeling-a-journey-from-historical-foundations-to-emerging-trends/270005861}}
ECIR
First International Workshop on Graph-Based Approaches in Information Retrieval (IRonGraphs 2024)
In the dynamic field of information retrieval, the adoption of graph-based approaches has become a notable research trend. Fueled by the growing research on Knowledge Graphs and Graph Neural Networks, these approaches rooted in graph theory have shown significant promise in enhancing the effectiveness and relevance of information retrieval results. With this motivation in mind, this workshop serves as a platform, bringing together researchers and practitioners from diverse backgrounds, to delve into and discuss the integration of modern graph-based methodologies into information retrieval methods. The workshop website is available at https://irongraphs.github.io/ecir2024/.
@inproceedings{boratto2024irongraphs,title={First International Workshop on Graph-Based Approaches in Information Retrieval (IRonGraphs 2024)},author={Boratto, Ludovico and Malitesta, Daniele and Marras, Mirko and Medda, Giacomo and Musto, Cataldo and Purificato, Erasmo},booktitle={European Conference on Information Retrieval},abbr={ECIR},pages={415--421},year={2024},organization={Springer},bibtex_show={true},url={https://link.springer.com/chapter/10.1007/978-3-031-56069-9_56},pdf={publications/2024_ECIR.pdf}}
arXiv
User Modeling and User Profiling: A Comprehensive Survey
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
The integration of artificial intelligence (AI) into daily life, particularly through information retrieval and recommender systems, has necessitated advanced user modeling and profiling techniques to deliver personalized experiences. These techniques aim to construct accurate user representations based on the rich amounts of data generated through interactions with these systems. This paper presents a comprehensive survey of the current state, evolution, and future directions of user modeling and profiling research. We provide a historical overview, tracing the development from early stereotype models to the latest deep learning techniques, and propose a novel taxonomy that encompasses all active topics in this research area, including recent trends. Our survey highlights the paradigm shifts towards more sophisticated user profiling methods, emphasizing implicit data collection, multi-behavior modeling, and the integration of graph data structures. We also address the critical need for privacy-preserving techniques and the push towards explainability and fairness in user modeling approaches. By examining the definitions of core terminology, we aim to clarify ambiguities and foster a clearer understanding of the field by proposing two novel encyclopedic definitions of the main terms. Furthermore, we explore the application of user modeling in various domains, such as fake news detection, cybersecurity, and personalized education. This survey serves as a comprehensive resource for researchers and practitioners, offering insights into the evolution of user modeling and profiling and guiding the development of more personalized, ethical, and effective AI systems.
@misc{purificato2024user,title={User Modeling and User Profiling: A Comprehensive Survey},author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},year={2024},eprint={2402.09660},archiveprefix={arXiv},abbr={arXiv},primaryclass={cs.AI},bibtex_show={true},url={https://arxiv.org/abs/2402.09660},pdf={publications/2024_arXiv.pdf}}
2023
AIxIA
Recent Advances in Fairness Analysis of User Profiling Approaches in E-Commerce with Graph Neural Networks
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Proceedings of the Discussion Papers - 22nd International Conference of the Italian Association for Artificial Intelligence (AIxIA 2023 DP), 2023
User profiling is a critical procedure for e-commerce applications that captures online users’ attributes, understands user models, supports the provision of tailor-made goods and services, and improves user satisfaction. With the advent of novel technologies like Graph Neural Networks (GNNs), the performance of user profiling approaches has improved by leaps and bounds, in step with the growing concern about data and algorithmic fairness. This paper provides an overview of recent advances in the fairness analysis of GNN-based models for user profiling in the e-commerce domain. We present the results of our recent works addressing the need for an accurate analysis of state-of-the-art models and the lack of a unified tool for enabling any user to perform a fairness analysis on a specific dataset by leveraging the most performing models in this context. Our goal is to foster discussions on the potential implications of our work within the community, not only from a technical view but also from domain experts’ perspective.
@inproceedings{purificato2023recent,title={Recent Advances in Fairness Analysis of User Profiling Approaches in E-Commerce with Graph Neural Networks},author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},year={2023},booktitle={Proceedings of the Discussion Papers - 22nd International Conference of the Italian Association for Artificial Intelligence (AIxIA 2023 DP),},abbr={AIxIA},pages={47--56},organization={CEUR},bibtex_show={true},url={https://ceur-ws.org/Vol-3537/paper6.pdf},pdf={publications/2023_AIxIA.pdf},slides={https://www.slideshare.net/ErasmoPurificato2/recent-advances-in-fairness-analysis-of-user-profiling-approaches-in-ecommerce-with-graph-neural-networks}}
CIKM
Leveraging Graph Neural Networks for User Profiling: Recent Advances and Open Challenges
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management 2023
The proposed tutorial aims to familiarise the CIKM community with modern user profiling techniques that utilise Graph Neural Networks (GNNs). Initially, we will delve into the foundational principles of user profiling and GNNs, accompanied by an overview of relevant literature. We will subsequently systematically examine cutting-edge GNN architectures specifically developed for user profiling, highlighting the typical data utilised in this context. Furthermore, ethical considerations and beyond-accuracy perspectives, e.g. fairness and explainability, will be discussed regarding the potential applications of GNNs in user profiling. During the hands-on session, participants will gain practical insights into constructing and training recent GNN models for user profiling using open-source tools and publicly available datasets. The audience will actively explore the impact of these models through case studies focused on bias analysis and explanations of user profiles. To conclude the tutorial, we will analyse existing and emerging challenges in the field and discuss future research directions.
@inproceedings{purificato2023leveraging,author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},title={Leveraging Graph Neural Networks for User Profiling: Recent Advances and Open Challenges},year={2023},isbn={9798400701245},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3583780.3615292},doi={10.1145/3583780.3615292},booktitle={Proceedings of the 32nd ACM International Conference on Information and Knowledge Management},pages={5216–5219},numpages={4},keywords={explainability, graph neural networks, fairness, user profiling},location={Birmingham, United Kingdom},series={CIKM '23},abbr={CIKM},bibtex_show={true},pdf={publications/2023_CIKM.pdf}}
CHItaly
First Workshop on User Perspectives in Human-Centred Artificial Intelligence (HCAI4U)
De Luca, Ernesto William,
Purificato, Erasmo, Boratto, Ludovico, Marrone, Stefano, and Sansone, Carlo
In Proceedings of the 15th Biannual Conference of the Italian SIGCHI Chapter 2023
The emerging concept of Human-Centred Artificial Intelligence (HCAI) involves the amplification, augmentation, empowerment, and enhancement of individuals. The goal of HCAI is to ensure that AI meets our needs while also operating transparently, delivering fair and equitable outcomes, and respecting privacy, all while preserving human control. This approach involves multiple stakeholders, such as researchers, developers, business leaders, policy makers, and users, who are affected in various ways by the implementation and evaluation of AI systems. The primary focus of the First Workshop on User Perspectives in Human-Centred Artificial Intelligence (HCAI4U) is to examine the potential positive and negative impacts of automated decision-making systems on end-users, as well as how their interaction with AI is influenced by human-centred aspects of reliability, safety, and fairness. The workshop aims to facilitate discussion and exchange of ideas among the community on advances in developing trustworthy, fair, and privacy-preserving systems, as well as user interfaces that are explainable, with a specific focus on the users’ perception in real-world scenarios rather than solely on the algorithmic and model performance. Additionally, HCAI4U aims to foster cross-disciplinary and interdisciplinary discussions between experts from various research fields, such as computer science, psychology, sociology, law, medicine, business, etc., to discuss problems and synergies in this exciting research topic.
@inproceedings{deluca2023hcai4u,author={De Luca, Ernesto William and Purificato, Erasmo and Boratto, Ludovico and Marrone, Stefano and Sansone, Carlo},title={First Workshop on User Perspectives in Human-Centred Artificial Intelligence (HCAI4U)},year={2023},isbn={9798400708060},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3605390.3610829},doi={10.1145/3605390.3610829},booktitle={Proceedings of the 15th Biannual Conference of the Italian SIGCHI Chapter},articleno={36},numpages={3},keywords={Reliability, Explainability, Human-Centred Artificial Intelligence, Trustworthiness, Human-Computer Interaction, Fairness, User Perspectives, Artificial Intelligence},location={Torino, Italy},series={CHItaly '23},abbr={CHItaly},bibtex_show={true},pdf={publications/2023_CHItaly.pdf}}
SIGIR
FairUP: A Framework for Fairness Analysis of Graph Neural Network-Based User Profiling Models
Abdelrazek, Mohamed,
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval 2023
Modern user profiling approaches capture different forms of interactions with the data, from user-item to user-user relationships. Graph Neural Networks (GNNs) have become a natural way to model these behaviours and build efficient and effective user profiles. However, each GNN-based user profiling approach has its own way of processing information, thus creating heterogeneity that does not favour the benchmarking of these techniques. To overcome this issue, we present FairUP, a framework that standardises the input needed to run three state-of-the-art GNN-based models for user profiling tasks. Moreover, given the importance that algorithmic fairness is getting in the evaluation of machine learning systems, FairUP includes two additional components to (1) analyse pre-processing and post-processing fairness and (2) mitigate the potential presence of unfairness in the original datasets through three pre-processing debiasing techniques. The framework, while extensible in multiple directions, in its first version, allows the user to conduct experiments on four real-world datasets. The source code is available at https://link.erasmopurif.com/FairUP-source-code, and the web application is available at https://link.erasmopurif.com/FairUP.
@inproceedings{abdelrazek2023fairup,author={Abdelrazek, Mohamed and Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},title={FairUP: A Framework for Fairness Analysis of Graph Neural Network-Based User Profiling Models},year={2023},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3539618.3591814},doi={10.1145/3539618.3591814},booktitle={Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval},pages={3165--3169},numpages={5},keywords={algorithmic fairness, graph neural networks, user profiling},location={Taipei, Taiwan},series={SIGIR '23},abbr={SIGIR},bibtex_show={true},pdf={publications/2023_SIGIR.pdf}}
HCII
Usability Studies in Times of Pandemic: Different Solutions for the Remote Usability Tests of Research Digital Tools
Purificato, Erasmo, Wehnert, Sabine, and De Luca, Ernesto William
In International Conference on Human-Computer Interaction, 2023
In this paper, we present the usability evaluation of three digital tools of the Leibniz Institute for Educational Media | Georg Eckert Institute, namely Curricula Workstation, GEI-Digital and International TextbookCat, compared with a Meta Search Engine (MSE) search.gei.de which shall replace the use of the individual tools. Due to the lockdown measures enforced by the German government at the end of 2021, we developed different solutions to adapt our usability test plans to remote settings. First, the MSE was compared with Curricula Workstation using recordings performed via Zoom. Second, we compared the MSE with GEI-Digital by leveraging mouse tracking data using a combination of Zoom with OBS Studio for recording the screen. Third, a comparison was made between International TextBookCat and the MSE with data collected via CamStudio. The experimental results showed that the individual tools are perceived as better than MSE mainly in terms of intuitive design and ease of learning, while MSE is more satisfying for users.
@inproceedings{purificato2023usability,title={Usability Studies in Times of Pandemic: Different Solutions for the Remote Usability Tests of Research Digital Tools},author={Purificato, Erasmo and Wehnert, Sabine and De Luca, Ernesto William},booktitle={International Conference on Human-Computer Interaction,},pages={666--673},year={2023},organization={Springer},url={https://doi.org/10.1007/978-3-031-35989-7_85},doi={10.1007/978-3-031-35989-7_85},abbr={HCII},location={Copenhagen, Denmark},bibtex_show={true},pdf={publications/2023_HCII.pdf}}
HCII
A Usability Study of a Research Institute Website with Eye-Tracking Devices
Wehnert, Sabine,
Purificato, Erasmo, and De Luca, Ernesto William
In International Conference on Human-Computer Interaction, 2023
In this paper, we present the results of our study conducted at the Leibniz Institute for Educational Media | Georg Eckert Institute to assess the usability of the institute’s website before the re-design of the same and the subsequent development of the new version. In particular, four specific pages are evaluated, i.e. Home, Institute, Departments and Publications. The aim of the presented usability studies is to uncover positive and negative usability findings in order to properly plan the potential corrective actions for the upcoming restyling. The experimental outcomes are displayed in form of aggregated heat maps and mainly focus on the ease of use of the different analysed sections.
@inproceedings{wehnert2023usability,title={A Usability Study of a Research Institute Website with Eye-Tracking Devices},author={Wehnert, Sabine and Purificato, Erasmo and De Luca, Ernesto William},booktitle={International Conference on Human-Computer Interaction,},pages={702--711},year={2023},organization={Springer},url={https://doi.org/10.1007/978-3-031-35989-7_90},doi={10.1007/978-3-031-35989-7_90},abbr={HCII},location={Copenhagen, Denmark},bibtex_show={true},pdf={publications/2023_HCII_2.pdf}}
UMAP
Tutorial on User Profiling with Graph Neural Networks and Related Beyond-Accuracy Perspectives
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization, 2023
The proposed tutorial aims to introduce the UMAP community to modern user profiling approaches leveraging graph neural networks (GNNs). We will begin by discussing the conceptual foundations of user profiling and GNNs and providing a literature review of the two topics. We will then present a systematic overview of the state-of-the-art GNN architectures designed for user profiling, including the types of data that are typically used for this purpose. We will also discuss ethical considerations and beyond-accuracy perspectives (i.e. fairness and explainability), which can arise within the potential applications of adopting GNNs for user profiling. In the practical session of the tutorial, attendees will have the opportunity to understand concretely how recent GNN models for user profiling are built and trained with open-source tools and publicly available datasets. The audience will also be engaged in investigating the impact of the presented models on case studies involving bias detection and mitigation, as well as user profiles explanations. The tutorial will end with an analysis of existing and emerging open challenges in the field and their future research directions.
@inproceedings{purificato2023tutorialumap,author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},title={Tutorial on User Profiling with Graph Neural Networks and Related Beyond-Accuracy Perspectives},year={2023},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3565472.3595616},doi={10.1145/3565472.3595616},booktitle={Proceedings of the 31st ACM Conference on User Modeling, Adaptation and Personalization,},abbr={UMAP},pages={309-312},numpages={4},keywords={Fairness, Graph Neural Networks, Explainability, User Profiling},location={Limassol, Cyprus},series={UMAP '23},bibtex_show={true},slides={https://www.slideshare.net/ErasmoPurificato2/tutorial-on-user-profiling-with-graph-neural-networks-and-related-beyondaccuracy-perspectives},pdf={publications/2023_UMAP.pdf}}
BIAS
What Are We Missing in Algorithmic Fairness? Discussing Open Challenges for Fairness Analysis in User Profiling with Graph Neural Networks
Purificato, Erasmo, and De Luca, Ernesto William
In International Workshop on Algorithmic Bias in Search and Recommendation 2023
Due to the rising importance of human-centred perspectives in artificial intelligence and all related fields, algorithmic fairness is currently a key topic when dealing with research on novel machine learning models and applications. However, in most cases, in the context of fairness analysis, we are commonly facing situations in which the fairness metrics are applied only in binary classification scenarios, and the capability of a model to produce fair results is evaluated considering the absolute difference of the scores of the two sensitive groups considered. In this paper, we aim to discuss these two open challenges and illustrate our position from an ethical perspective. To support our arguments, we present a case study on two recent scientific contributions exploiting Graph Neural Networks models for user profiling, which are considered state-of-the-art technologies in many domains. With the presented work, our goal is also to create a valuable debate in the community about the raised questions.
@inproceedings{purificato2023what,title={What Are We Missing in Algorithmic Fairness? Discussing Open Challenges for Fairness Analysis in User Profiling with Graph Neural Networks},author={Purificato, Erasmo and De Luca, Ernesto William},booktitle={International Workshop on Algorithmic Bias in Search and Recommendation},abbr={BIAS},pages={169--175},year={2023},organization={Springer},bibtex_show={true},pdf={publications/2023_BIAS.pdf}}
ECIR
FACADE: Fake Articles Classification and Decision Explanation
Purificato, Erasmo, Shahania, Saijal, Thiel, Marcus, and De Luca, Ernesto William
In Advances in Information Retrieval: 45th European Conference on Information Retrieval (ECIR 2023), Part III, 2023
The daily use of social networks and the resulting dissemination of disinformation over those media have greatly contributed to the rise of the fake news phenomenon as a global problem. Several manual and automatic approaches are currently in place to try to tackle and defuse this issue, which is becoming nearly uncontrollable. In this paper, we propose Facade, a fake news detection system that aims to provide a complete solution for classifying news articles and explain the motivation behind every prediction. The system is designed with a cascading architecture composed of two classification pipelines dealing with either low-level or high-level descriptors, with the overall goal of achieving a consistent confidence score on each outcome. In addition, the system is equipped with an explainable user interface through which fact-checkers and content managers can visualise in detail the features leading to a certain prediction and have the possibility for manual cross-checking.
@inproceedings{purificato2023facade,title={FACADE: Fake Articles Classification and Decision Explanation},author={Purificato, Erasmo and Shahania, Saijal and Thiel, Marcus and De Luca, Ernesto William},booktitle={Advances in Information Retrieval: 45th European Conference on Information Retrieval (ECIR 2023), Part III,},abbr={ECIR},pages={294--299},year={2023},organization={Springer},bibtex_show={true},url={https://link.springer.com/chapter/10.1007/978-3-031-28241-6_29},pdf={publications/2023_ECIR.pdf}}
2022
XAI.it
Tell Me Why It’s Fake: Developing an Explainable User Interface for a Fake News Detection System
Purificato, Erasmo, Shahania, Saijal, and De Luca, Ernesto William
In Proceedings of the 3rd Italian Workshop on Explainable Artificial Intelligence, co-located with 21th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2022), 2022
In this paper, we present the design and development of an explainable user interface for a fake news detection system. The problem of distinguishing real from fake articles gained a lot of popularity in the last few years, mainly due to the soaring diffusion of social networks and internet bots as means for propaganda and disinformation sharing. By leveraging various explainability methods, i.e. feature importance, partial dependence plots and SHAP values, we aim to show how the combination of different techniques embedded in an interactive user interface can lead to enhance trust in a detection system for a non-expert user, such as a fact-checker or a content manager. Through several examples, we describe all the explainability components along with the benefits and limitations they can provide to end users.
@inproceedings{purificato2022tell,title={Tell Me Why It's Fake: Developing an Explainable User Interface for a Fake News Detection System},author={Purificato, Erasmo and Shahania, Saijal and De Luca, Ernesto William},year={2022},booktitle={Proceedings of the 3rd Italian Workshop on Explainable Artificial Intelligence, co-located with 21th International Conference of the Italian Association for Artificial Intelligence (AIxIA 2022),},abbr={XAI.it},pages={51--63},organization={CEUR},bibtex_show={true},url={https://ceur-ws.org/Vol-3277/paper4.pdf},pdf={publications/2022_XAI-it.pdf},slides={https://www.slideshare.net/ErasmoPurificato2/tell-me-why-its-fake-developing-an-explainable-user-interface-for-a-fake-news-detection-system}}
CIKM
Do Graph Neural Networks Build Fair User Models? Assessing Disparate Impact and Mistreatment in Behavioural User Profiling
Purificato, Erasmo, Boratto, Ludovico, and De Luca, Ernesto William
In Proceedings of the 31st ACM International Conference on Information & Knowledge Management, 2022
Recent approaches to behavioural user profiling employ Graph Neural Networks (GNNs) to turn users’ interactions with a platform into actionable knowledge. The effectiveness of an approach is usually assessed with accuracy-based perspectives, where the capability to predict user features (such as gender or age) is evaluated. In this work, we perform a beyond-accuracy analysis of the state-of-the-art approaches to assess the presence of disparate impact and disparate mistreatment, meaning that users characterised by a given sensitive feature are unintentionally, but systematically, classified worse than their counterparts. Our analysis on two real-world datasets shows that different user profiling paradigms can impact fairness results. The source code and the preprocessed datasets are available at: https://github.com/erasmopurif/do_gnns_build_fair_models.
@inproceedings{purificato2022cikm,author={Purificato, Erasmo and Boratto, Ludovico and De Luca, Ernesto William},title={Do Graph Neural Networks Build Fair User Models? Assessing Disparate Impact and Mistreatment in Behavioural User Profiling},year={2022},isbn={9781450392365},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3511808.3557584},doi={10.1145/3511808.3557584},booktitle={Proceedings of the 31st ACM International Conference on Information & Knowledge Management,},pages={4399–4403},numpages={5},keywords={fairness, user profiling, user models, graph neural networks},location={Atlanta, GA, USA},series={CIKM '22},abbr={CIKM},bibtex_show={true},pdf={publications/2022_CIKM.pdf},poster={posters/2022_CIKM_poster.pdf},slides={https://www.slideshare.net/ErasmoPurificato2/do-graph-neural-networks-build-fair-user-models-assessing-disparate-impact-and-mistreatment-in-behavioural-user-profiling}}
UMAP
Beyond-Accuracy Perspectives on Graph Neural Network-Based Models for Behavioural User Profiling
Purificato, Erasmo
In Proceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization, 2022
The presented doctoral research aims to develop a behavioural user profiling framework focusing simultaneously on three beyond-accuracy perspectives: privacy, to study how to intervene on graph data structures of specific contexts and provide methods to make the data available in a meaningful manner without neither exposing personal user information nor corrupting the profiles creation and system performances; fairness, to provide user representations that are free of any inherited discrimination which could affect a downstream recommender by developing debiasing approaches to be applied on state-of-the-art GNN-based user profiling models; explainability, to produce understandable descriptions of the framework results, both for user profiles and recommendations, mainly in terms of interaction importance, by designing an adaptive and personalised user interface which provides tailored explanations to the end-users, depending on their specific user profiles.
@inproceedings{purificato2022umapdc,author={Purificato, Erasmo},title={Beyond-Accuracy Perspectives on Graph Neural Network-Based Models for Behavioural User Profiling},year={2022},isbn={9781450392075},publisher={Association for Computing Machinery},address={New York, NY, USA},url={https://doi.org/10.1145/3503252.3534361},doi={10.1145/3503252.3534361},booktitle={Proceedings of the 30th ACM Conference on User Modeling, Adaptation and Personalization,},abbr={UMAP},pages={311--315},numpages={5},keywords={Explainability, User Profiling, Fairness, Privacy, Graph Neural Networks, User Interfaces},location={Barcelona, Spain},series={UMAP '22},bibtex_show={true},pdf={publications/2022_UMAP.pdf},poster={posters/2022_UMAP_poster.pdf},slides={https://www.slideshare.net/ErasmoPurificato2/beyondaccuracy-perspectives-on-graph-neural-networkbased-models-for-behavioural-user-profiling}}
MTAP
Enriching Videos with Automatic Place Recognition in Google Maps
Fallucchi, Francesca, Di Stabile, Rosario,
Purificato, Erasmo, Giuliano, Romeo, and De Luca, Ernesto William
The availability of videos has grown rapidly in recent years. Finding and browsing relevant information to be automatically extracted from videos is not an easy task, but today it is an indispensable feature due to the immense number of digital products available. In this paper, we present a system which provides a process to automatically extract information from videos. We describe a system solution that uses a re-trained OpenNLP model to locate all the places and famous people included in a specific video. The system obtains information from the Google Knowledge Graph related to relevant named entities such as places or famous people. In this paper we will also present the Automatic Georeferencing Video (AGV) system developed by RAI (Radiotelevisione italiana, which is the national public broadcasting company of Italy, owned by the Ministry of Economy and Finance) Teche for the European Project “La Città Educante” (The Educating City: teaching and learning processes in cross-media ecosystem) Our system contributes to The Educating City project by providing the technological environment to create statistical models for automatic named entity recognition (NER), and has been implemented in the field of education, in Italian initially. The system has been applied to the learning challenges facing the world of educational media and has demonstrated how beneficial combining topical news content with scientific content can be in education.
@article{fallucchi2022enriching,title={Enriching Videos with Automatic Place Recognition in Google Maps},author={Fallucchi, Francesca and Di Stabile, Rosario and Purificato, Erasmo and Giuliano, Romeo and De Luca, Ernesto William},journal={Multimedia Tools and Applications,},abbr={MTAP},volume={81},number={16},pages={23105--23121},year={2022},publisher={Springer},doi={10.1007/s11042-021-11253-9},bibtex_show={true},url={https://link.springer.com/article/10.1007/s11042-021-11253-9},pdf={publications/2021_MTAP.pdf}}
IJHCI
The Use of Responsible Artificial Intelligence Techniques in the Context of Loan Approval Processes
Purificato, Erasmo, Lorenzo, Flavio, Fallucchi, Francesca, and De Luca, Ernesto William
In International Journal of Human–Computer Interaction, 2022
Despite the existing skepticism about the use of automatic systems in contexts where human knowledge and experience are considered indispensable (e.g., the granting of a mortgage, the prediction of stock prices, or the detection of cancers), our work aims to show how the use of explainability and fairness techniques can lead to the growth of a domain expert’s trust and reliance on an artificial intelligence (AI) system. This article presents a system, applied to the context of loan approval processes, focusing on the two aforementioned ethical principles out of the four defined by the High-Level Expert Group on AI in the document “Ethics Guidelines for Trustworthy AI,” published in April 2019, in which the key requirements that AI systems should meet to be considered trustworthy are identified. The presented case study is realized within a proprietary framework composed of several components for supporting the user throughout the management of the whole life cycle of a machine learning model. The main approaches, consisting of providing an interpretation of the model’s outputs and monitoring the model’s decisions to detect and react to unfair behaviors, are described in more detail to compare our system within state-of-the-art related frameworks. Finally, a novel Trust & Reliance Scale is proposed for evaluating the system, and a usability test is performed to measure the user satisfaction with the effectiveness of the developed user interface; results are obtained, respectively, by the submission of the mentioned novel scale to bank domain experts and the usability questionnaire to a heterogeneous group composed of loan officers, data scientists, and researchers.
@article{purificato2022theuseofrai,author={Purificato, Erasmo and Lorenzo, Flavio and Fallucchi, Francesca and De Luca, Ernesto William},title={The Use of Responsible Artificial Intelligence Techniques in the Context of Loan Approval Processes},journal={International Journal of Human–Computer Interaction,},abbr={IJHCI},volume={0},number={0},pages={1--20},year={2022},publisher={Taylor & Francis},doi={10.1080/10447318.2022.2081284},bibtex_show={true},url={https://doi.org/10.1080/10447318.2022.2081284},pdf={publications/2022_IJHCI.pdf}}
IUI
First Workshop on Adaptive and Personalized Explainable User Interfaces (APEx-UI 2022)
Purificato, Erasmo, Musto, Cataldo, Lops, Pasquale, and De Luca, Ernesto William
In Proceedings of the 27th International Conference on Intelligent User Interfaces, 2022
Adaptation and personalization are crucial aspects of the design and development of successful Artificial Intelligence systems, from search engines and recommender systems to wearable devices. The increased desire for customization inevitably leads to the need for the end-user to understand the rationale behind displaying that specific tailored content. User interfaces play a central role to provide the right explanations to the end-users. While adaptive and personalized user interfaces are well-known and advanced research fields, a common issue we face in terms of explainability is finding intelligent user interfaces following the one-fits-all paradigm without considering the different peculiarities of individuals. The 1st Workshop on Adaptive and Personalized Explainable User Interfaces (APEx-UI 2022) aims to foster a cross-disciplinary and interdisciplinary discussion between experts from different fields (e.g. computer science, psychology, sociology, law, medicine, business, etc.) in order to answer a precise research question: How can we adapt and personalize explainable user interfaces to the needs, demands and requirements of different end-users, considering their distinct knowledge, background and expertise?
@inproceedings{purificato2022apexui,title={First Workshop on Adaptive and Personalized Explainable User Interfaces (APEx-UI 2022)},author={Purificato, Erasmo and Musto, Cataldo and Lops, Pasquale and De Luca, Ernesto William},booktitle={Proceedings of the 27th International Conference on Intelligent User Interfaces,},abbr={IUI},pages={1--3},year={2022},publisher={Association for Computing Machinery},doi={10.1145/3490100.3511168},bibtex_show={true},url={https://dl.acm.org/doi/abs/10.1145/3490100.3511168},pdf={publications/2022_IUI.pdf}}
2021
IntRS
Evaluating Explainable Interfaces for a Knowledge Graph-Based Recommender System
Purificato, Erasmo, Manikandan, Baalakrishnan Aiyer, Karanam, Prasanth Vaidya, Pattadkal, Mahantesh Vishvanath, and De Luca, Ernesto William
In Proceedings of the 8th Joint Workshop on Interfaces and Human Decision Making for Recommender Systems co-located with 15th ACM Conference on Recommender Systems (RecSys 2021), 2021
In this paper, we present the design and the implementation of a knowledge graph-based recommender system for research paper suggestion, along with two explainable interfaces which provide different types of explanations to the users interacting with the recommender. Our work, developed within the academic context of the Georg Eckert Institute for International Textbook Research, aims to assess the effectiveness of the explanation among the researchers of the institute and understand which characteristics of the interfaces themselves are perceived to be as most interpretable, leading to increase the trust and confidence in the recommender system and its credibility. We evaluated our work through a user study performed among different experts covering several research fields. All participants were asked to take part in an online survey, and a focus group answered some targeted interviews. This last qualitative evaluation aims better to understand the interaction patterns within the two explainable interfaces. The results show the greater effectiveness of the interface providing the explanation through a natural language sentence and displaying the graph path from the user to the recommended paper.
@inproceedings{purificato2021evaluating,title={Evaluating Explainable Interfaces for a Knowledge Graph-Based Recommender System},author={Purificato, Erasmo and Manikandan, Baalakrishnan Aiyer and Karanam, Prasanth Vaidya and Pattadkal, Mahantesh Vishvanath and De Luca, Ernesto William},booktitle={Proceedings of the 8th Joint Workshop on Interfaces and Human Decision Making for Recommender Systems co-located with 15th ACM Conference on Recommender Systems (RecSys 2021),},abbr={IntRS},pages={73--88},year={2021},organization={CEUR},bibtex_show={true},url={http://ceur-ws.org/Vol-2948/paper5.pdf},pdf={publications/2021_IntRS.pdf},slides={https://www.slideshare.net/ErasmoPurificato2/evaluating-explainable-interfaces-for-a-knowledge-graphbased-recommender-system}}
Computers
Dynamic Privacy-Preserving Recommendations on Academic Graph Data
Purificato, Erasmo, Wehnert, Sabine, and De Luca, Ernesto William
In the age of digital information, where the internet and social networks, as well as personalised systems, have become an integral part of everyone’s life, it is often challenging to be aware of the amount of data produced daily and, unfortunately, of the potential risks caused by the indiscriminate sharing of personal data. Recently, attention to privacy has grown thanks to the introduction of specific regulations such as the European GDPR. In some fields, including recommender systems, this has inevitably led to a decrease in the amount of usable data, and, occasionally, to significant degradation in performance mainly due to information no longer being attributable to specific individuals. In this article, we present a dynamic privacy-preserving approach for recommendations in an academic context. We aim to implement a personalised system capable of protecting personal data while at the same time allowing sensible and meaningful use of the available data. The proposed approach introduces several pseudonymisation procedures based on the design goals described by the European Union Agency for Cybersecurity in their guidelines, in order to dynamically transform entities (e.g., persons) and attributes (e.g., authored papers and research interests) in such a way that any user processing the data are not able to identify individuals. We present a case study using data from researchers of the Georg Eckert Institute for International Textbook Research (Brunswick, Germany). Building a knowledge graph and exploiting a Neo4j database for data management, we first generate several pseudoN-graphs, being graphs with different rates of pseudonymised persons. Then, we evaluate our approach by leveraging the graph embedding algorithm node2vec to produce recommendations through node relatedness. The recommendations provided by the graphs in different privacy-preserving scenarios are compared with those provided by the fully non-pseudonymised graph, considered as the baseline of our evaluation. The experimental results show that, despite the structural modifications to the knowledge graph structure due to the de-identification processes, applying the approach proposed in this article allows for preserving significant performance values in terms of precision.
@article{purificato2021dynamic,title={Dynamic Privacy-Preserving Recommendations on Academic Graph Data},author={Purificato, Erasmo and Wehnert, Sabine and De Luca, Ernesto William},journal={Computers,},abbr={Computers},volume={10},number={9},pages={107},year={2021},publisher={Multidisciplinary Digital Publishing Institute},doi={10.3390/computers10090107},bibtex_show={true},url={https://www.mdpi.com/2073-431X/10/9/107},pdf={publications/2021_Computers.pdf}}
2018
ICCSA
A Multimodal Approach for Cultural Heritage Information Retrieval
Purificato, Erasmo, and Rinaldi, Antonio Maria
In International Conference on Computational Science and its Applications, 2018
The daily use of mobile devices and the expansion of the world-wide-web lead multimedia information to an uncontrolled growth. In this context, the use of smart interfaces and the combination of different features in the information retrieval process are crucial aspects. In particular, for a cultural heritage application it is important to consider that a digitized artwork is only a representation of a real object, represented under specific conditions (camera position, brightness, etc.). These issues could be causes of alterations during the features extraction task. In this paper we propose a multimodal approach for cultural heritage information retrieval combining geographic and visual data. Our approach has been implemented in a mobile system based on open source technologies. It is composed of three main parts related to image matching functionalities, Geographic Information Retrieval task, and a combination strategy for multimedia and geographic data integration. An Android application has been developed to give a user friendly interface and a case study together with some experimental results are presented to show the effectiveness of our approach for the user satisfaction.
@inproceedings{purificato2018multimodal,title={A Multimodal Approach for Cultural Heritage Information Retrieval},author={Purificato, Erasmo and Rinaldi, Antonio Maria},booktitle={International Conference on Computational Science and its Applications,},abbr={ICCSA},pages={214--230},year={2018},organization={Springer},doi={10.1007/978-3-319-95162-1_15},bibtex_show={true},url={https://link.springer.com/chapter/10.1007/978-3-319-95162-1_15},pdf={publications/2018_ICCSA.pdf}}
MTAP
Multimedia and geographic data integration for cultural heritage information retrieval
In this paper a system providing an efficient integration between Content-Based Image Retrieval (CBIR) and Geographic Information Retrieval (GIR) is presented. Over the years, many CBIR systems have been proposed to give a solution for an efficient use of multimedia/visual contents and other issues as performance, quality of retrieval, data heterogeneity, and multimodal information integration. The aim of the proposed approach is to prove that the use of geographic data can improve the results obtained by an image matching system based only on visual data. Our framework is composed of three parts, each of them described in detail in this paper: the first part is dedicated to CBIR, with an experimental comparison of a large number of different multimedia features to choose the one to use in the system implementation; in the second part the methodology to integrate geographic and multimedia data is showed; in the last part is presented a GIR system implementation using a “points of interest” search. An Android application has been developed for the client-side using Apache Solr as server side provider for the information retrieval functionalities. An experimental evaluation is carried out to demonstrate the effective improvement given by the combination of geographic and multimedia data. Our results have been obtained using a real dataset composed of artworks located in Naples’s museums.
@article{purificato2018multimedia,title={Multimedia and geographic data integration for cultural heritage information retrieval},author={Purificato, Erasmo and Rinaldi, Antonio Maria},journal={Multimedia Tools and Applications,},abbr={MTAP},volume={77},number={20},pages={27447--27469},year={2018},publisher={Springer},doi={10.1007/s11042-018-5931-7},bibtex_show={true},url={https://dl.acm.org/doi/10.1007/s11042-018-5931-7},pdf={publications/2018_MTAP.pdf}}