• Funding : Autre
  • Start year :
  • 2022

Joined PhD with dMonastir University, Tunisia

The goal of the thesis is to develop novel methods and algorithms based on symbolic artificial intelligence (SAI) techniques to model the influence maximization problem in social networks. These algorithms must also be able to discover overlapping communities and provide alternatives to counter the influence or limit the spread of information in the network. From a data mining point of view, the challenge is to develop efficient SAI based encodings able to represent large size networks with the possibility of mining particular graphical structures encoded in the form of boolean constraints. A final objective of the thesis is to consider the notion of trust when querying data in social networks. This notion is widely studied in various fields, including psychology, philosophy, sociology, etc. In this context, the main idea is to model and to take into account the trust during the influence processing in networks.

Keywords: Social networks, Graph, Influence Maximization, Propositional Satisfiability