A study of consensus building under two different hypotheses: truthful annotators (as a model for most voluntary citizen science projects) and self-interested annotators (as a model for paid crowdsourcing projects). read more of A tale of two consensus. Building consensus in collaborative and self-interested scenarios.

We envision a human-AI ecosystem in which AI-enabled devices act as proxies of humans and try to learn collectively a model. read more of Social AI gossiping

In this project, we enhance as basic algorithmic bias model by adding a biased interaction with media, to understand whether this facilitates polarisation. read more of Algorithmic bias and media effects

Development of an ethical dialogue system, with a case study on COVID-19 vaccine information. read more of Ethical Chatbots

The project delivered detailed analyses of the tweets around the USA elections and subsequent riots. read more of Normative behavior and extremism in facebook groups

Study of emergent collective phenomena at metropolitan level in personal navigation assistance systems with different recommendation policies, with respect to different collective optimization criteria, such as traffic load balance, safety risks and envir... read more of Impact of Navigation Systems on Urban Sustainability

This project investigated a data generation methodology that, given a data sample, can approximate the stochastic process that generated it. The methodology can be useful in many contexts where we need to share data while preserving user privacy. read more of A Graph-Based Drift-Aware Data Cloning Process

We used decision-making processes to create more explicable models that replicate human thinking. read more of Causality and Explainability in Temporal Data

We studied online optimization methods for hyper-parameter tuning. read more of Online Deep-AUTOML

This project investigated the role of social norms on misinformation in online communities. read more of Using Social Norms to counteract misinformation in online communities

This project conducted empirical research that explored the social and public attitudes of individuals towards AI and robots. read more of What idea of AI? Social and public perception of AI

The combination of mobility data can provide actionable insights about traffic mobility patterns read more of Improving air quality in large cities using mobile phone and IoT data

Creation of stories and narrative from data of Cultural Heritage read more of Venice

X-AI techniques applied to a DL model used to check real-time multivariate time series from connected black-boxes installed on cars. read more of Explaining Crash Predictions on Multivariate Time Series Data

Perform unbiased recommendation for news in the European domain. To achieve it, build a political leaning classifier on a EU-News Dataset read more of Pluralistic Recommendation in News

This project investigates the impact of dynamic social graphs on polarization with the Friedkin-Johnsen opinion dynamic model. The model can reproduce also special users in the network with extreme or neutral opinions and not willing to change their minds... read more of Polarization with the Friedkin-Johnsen model over a dynamic social network

Use simulated Connected Car data to train predictive models, in particular predict on-street parking space occupancies read more of Connected vehicle simulation for AI-based applications

There is a segmentation in a small “core” of active users responsible for large amounts of fake news and a larger "periphery" that mainly retweets. read more of Social norms to counteract misinformation in human-AI hybrid systems

The micro-project investigated recent machine and deep learning-based methods for modeling event interactions through temporal processes. read more of Human Behavior is a matter of Time! Modeling Events Interactions through Temporal Processes

We study how simple social structures affect the learning process of a fully decentralised AI system read more of Decentralized AI on simple social networks