research
A brief discussion about my main research interests.
During my carrier as a researcher, I have been working on several topics related to machine learning like:
- Federated Learning
- Explainable/Interpretable Machine Learning
- Cybersecurity
- Recommender Systems
- Kernel Methods
- Process Mining
In the last years, I am focusing my efforts on Federated Learning with several recent publications [C19, C20, C21, C25].
I am also active in the organization of workshops on this theme:
- 1st Workshop on Federated Learning Technologies @ TheWebConf 2023
- 1st Workshop on Advancement in Federated Learning @ ECML-PKDD 2023
as well as special sessions
and journal special issue:
- Federated Learning: Theoretical and Practical Advances - Frontiers in Big Data
Before diving into Federated Learning, in the past, I worked on Cybersecurity [C12] and security/privacy aspects related to recommender systems [C16,J11]. Recommender Systems have been a research topic of mine since my Ph.D. thesis [C14, W03, C17, W04, C24], and I am still relatively active in this area.
Another research area I am still interested in is Interpretable Machine Learning. I have been tackling this problem from different angles, e.g., via rule-set learning [C12], designing ad hoc methods based on game-theoretic concepts [C13, C15, J12], and using logic-based kernels [C05, C10, J06].
During my Ph.D., I specialized myself in the development of logic-based kernels for categorical data with applications to collaborative filtering [C20, C03, W02, J01, J02] and for the design of interpretable machine learning techniques [C05, C10, J06]. I also worked on theoretical aspects related to such family of kernels [C04, J03, J10].
However, my most cited works are related to Process Mining in particular machine learning applied to the prediction of the remaining time of a running business process instance [C01, C11, J04].