Oana Balalau: Explainable recommendations and representation learning in graphs

Oana Balalau will present a talk entitled “Explainable recommendations and representation learning in graphs” at November 29, 2pm in the Thomas Flowers room of the Turing building. Oana joined our team in early November as Inria Starting Reasearcher.

Title: Explainable recommendations and representation learning in graphs

Abstract: In this talk, I will present two of my recent research projects.
In the first project, our goal is to provide explanations for recommendations
that users receive each day. We introduce a provider-side solution in a
graph-based recommender system, where an explanation is defined to be a
set of minimal actions performed by the user that, if removed, changes
the recommendation to a different item. Given a recommendation, we introduce
a polynomial-time optimal algorithm for finding this minimal set of a user’s
actions from an exponential search space.
In the second project, the objective is to learn subgraph embeddings.
Subgraph embeddings have many applications, such as community detection,
cascade prediction or question answering. In this work, we propose a
subgraph to subgraph proximity measure as a building block for a
subgraph embedding framework.

Comments are closed.