André Panisson
YOU?
Author Swipe
View article: Size-adaptive Hypothesis Testing for Fairness
Size-adaptive Hypothesis Testing for Fairness Open
Determining whether an algorithmic decision-making system discriminates against a specific demographic typically involves comparing a single point estimate of a fairness metric against a predefined threshold. This practice is statistically…
View article: Learning Individual Behavior in Agent-Based Models with Graph Diffusion Networks
Learning Individual Behavior in Agent-Based Models with Graph Diffusion Networks Open
Agent-Based Models (ABMs) are powerful tools for studying emergent properties in complex systems. In ABMs, agent behaviors are governed by local interactions and stochastic rules. However, these rules are, in general, non-differentiable, l…
View article: Multi-Class and Multi-Task Strategies for Neural Directed Link Prediction
Multi-Class and Multi-Task Strategies for Neural Directed Link Prediction Open
Link Prediction is a foundational task in Graph Representation Learning, supporting applications like link recommendation, knowledge graph completion and graph generation. Graph Neural Networks have shown the most promising results in this…
View article: Disentangled and Self-Explainable Node Representation Learning
Disentangled and Self-Explainable Node Representation Learning Open
Node representations, or embeddings, are low-dimensional vectors that capture node properties, typically learned through unsupervised structural similarity objectives or supervised tasks. While recent efforts have focused on explaining gra…
View article: Value is in the Eye of the Beholder: A Framework for an Equitable Graph Data Evaluation
Value is in the Eye of the Beholder: A Framework for an Equitable Graph Data Evaluation Open
Proprietary data is a valuable asset used to develop predictive algorithms that benefit a wide range of users, including customers, business owners, and decision-makers. Consequently, there is a growing interest in developing safe and robu…
View article: DINE: Dimensional Interpretability of Node Embeddings
DINE: Dimensional Interpretability of Node Embeddings Open
Graphs are ubiquitous due to their flexibility in representing social and technological systems as networks of interacting elements. Graph representation learning methods, such as node embeddings, are powerful approaches to map nodes into …
View article: Evaluating Link Prediction Explanations for Graph Neural Networks
Evaluating Link Prediction Explanations for Graph Neural Networks Open
Graph Machine Learning (GML) has numerous applications, such as node/graph classification and link prediction, in real-world domains. Providing human-understandable explanations for GML models is a challenging yet fundamental task to foste…
View article: Beyond One-Hot-Encoding: Injecting Semantics to Drive Image Classifiers
Beyond One-Hot-Encoding: Injecting Semantics to Drive Image Classifiers Open
Images are loaded with semantic information that pertains to real-world ontologies: dog breeds share mammalian similarities, food pictures are often depicted in domestic environments, and so on. However, when training machine learning mode…
View article: Global Misinformation Spillovers in the Vaccination Debate Before and During the COVID-19 Pandemic: Multilingual Twitter Study
Global Misinformation Spillovers in the Vaccination Debate Before and During the COVID-19 Pandemic: Multilingual Twitter Study Open
Background Antivaccination views pervade online social media, fueling distrust in scientific expertise and increasing the number of vaccine-hesitant individuals. Although previous studies focused on specific countries, the COVID-19 pandemi…
View article: The Thin Ideology of Populist Advertising on Facebook during the 2019 EU Elections
The Thin Ideology of Populist Advertising on Facebook during the 2019 EU Elections Open
Social media has been an important tool in the expansion of the populist\nmessage, and it is thought to have contributed to the electoral success of\npopulist parties in the past decade. This study compares how populist parties\nadvertised…
View article: Revealing the determinants of gender inequality in urban cycling with large-scale data
Revealing the determinants of gender inequality in urban cycling with large-scale data Open
Cycling is an outdoor activity with massive health benefits, and an effective solution for sustainable urban transport. Despite these benefits and the recent rising popularity of cycling, most countries still have a negligible uptake. This…
View article: Annotation of Facebook ads in 2019 EU parliamentary elections
Annotation of Facebook ads in 2019 EU parliamentary elections Open
This dataset contains manual annotation for ads from parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections and that were defined as "populist" by popu-list.org. For each ad, …
View article: Annotation of Facebook ads in 2019 EU parliamentary elections
Annotation of Facebook ads in 2019 EU parliamentary elections Open
This dataset contains manual annotation for ads from parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections and that were defined as "populist" by popu-list.org. For each ad, …
View article: Global misinformation spillovers in the online vaccination debate before and during COVID-19
Global misinformation spillovers in the online vaccination debate before and during COVID-19 Open
Anti-vaccination views pervade online social media, fueling distrust in scientific expertise and increasing vaccine-hesitant individuals. While previous studies focused on specific countries, the COVID-19 pandemic brought the vaccination d…
View article: Mapping urban socioeconomic inequalities in developing countries through Facebook advertising data
Mapping urban socioeconomic inequalities in developing countries through Facebook advertising data Open
Ending poverty in all its forms everywhere is the number one Sustainable Development Goal of the UN 2030 Agenda. To monitor the progress toward such an ambitious target, reliable, up-to-date and fine-grained measurements of socioeconomic i…
View article: Echoes through Time: Evolution of the Italian COVID-19 Vaccination Debate
Echoes through Time: Evolution of the Italian COVID-19 Vaccination Debate Open
Italy was the first European country to be hit by COVID-19 in the early 2020, since then losing over 100,000 people to the disease. By the end of the vaccination campaign of 2021, 81% of the public received at least one dose. These dramati…
View article: Facebook pages for parties in 2019 EU parliamentary elections
Facebook pages for parties in 2019 EU parliamentary elections Open
This dataset contains information about the parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections. For each party, we manually associated the corresponding Facebook page that…
View article: Facebook pages for parties in 2019 EU parliamentary elections
Facebook pages for parties in 2019 EU parliamentary elections Open
This dataset contains information about the parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections. For each party, we manually associated the corresponding Facebook page that…
View article: Facebook pages for parties in 2019 EU parliamentary elections
Facebook pages for parties in 2019 EU parliamentary elections Open
This dataset contains information about the parties from UK, Italy, Germany, Spain, and Poland that ran ads on Facebook during the 2019 EU parliamentary elections. For each party, we manually associated the corresponding Facebook page that…
View article: Echoes through Time: Evolution of the Italian COVID-19 Vaccination Debate
Echoes through Time: Evolution of the Italian COVID-19 Vaccination Debate Open
Twitter is one of the most popular social media platforms in the country, but pre-pandemic vaccination debate has been shown to be polarized and siloed into echo chambers. It is thus imperative to understand the nature of this discourse, w…
View article: Revealing the determinants of gender inequality in urban cycling with large-scale data
Revealing the determinants of gender inequality in urban cycling with large-scale data Open
Cycling is an outdoor activity with massive health benefits, and an effective solution towards sustainable urban transport. Despite these benefits and the recent rising popularity of cycling, most countries still have a negligible uptake. …
View article: GRAPHSHAP: Explaining Identity-Aware Graph Classifiers Through the Language of Motifs
GRAPHSHAP: Explaining Identity-Aware Graph Classifiers Through the Language of Motifs Open
Most methods for explaining black-box classifiers (e.g. on tabular data, images, or time series) rely on measuring the impact that removing/perturbing features has on the model output. This forces the explanation language to match the clas…
View article: Forecast of Distributed Energy Generation and Consumption in a Partially Observable Electrical Grid: A Machine Learning Approach
Forecast of Distributed Energy Generation and Consumption in a Partially Observable Electrical Grid: A Machine Learning Approach Open
With a radical energy transition fostered by the increased deployment of renewable non-programmable energy sources over conventional ones, the forecasting of distributed energy production and consumption is becoming a cornerstone to ensure…