From my first year in university in 2014 to a Ph.D. student now, I am one of the pure products of Cy Cergy.
I did a Computer Science Bachelor’s degree from 2014 to 2017 and an Image and Big Data Master’s degree from 2017 to 2019 under the label CMI.
Now and since 2020 I work as a Ph.D. student on the subject “Explanations for missing recommendations in machine-learning-based recommenders” under the supervision of Aikaterini Tzompanaki and the direction of Dimitris Kotzinos
- Machine Learning
- Recommendation Systems
- Explanable Artificial Intelligence
- Why-Not Explanation
Recommenders suggest pertinent items to users from a vast variety of possibilities. However, it is crucial for the user and the system developer to understand why the system recommends certain items (Why), and why it does not recommend others that he/she might expect (Why-Not). In this thesis, we aim to explore explanations to the “Why-Not” problem, which is less explored, but equally important. This thesis falls in the wider area of explainable recommenders, by aiming to provide insights into why certain items are not recommended. Explanations on unseen items are primarily useful to developers or knowledgeable users that observe unexpectedly missing recommendations. In the case of developers, such explanations can further guide their system-repairing process. In this thesis, we are first planning to explore and define explanations for missing recommendations. Second, we are going to conceive techniques to compute them, depending on the category of the recommender system. In parallel, we are going to explore different visualization methods that would best convey the explanations. Finally, we are going to leverage the explanations to aid the developers auto-tune their recommenders.