Study and development of a Deep-Learning Based Semantic Recommendation System for Web Social Networks

  • 07 Dec 2020
  • Ongoing Research - Informatics & Communication

Ammar Alnahhas

Researchers

Post Graduate Studies, Research & IR Council Meeting No. 1, 22/11/2020

Date of Acceptance


Web social networks has spread widely recently, causing a massive data to be exchanged thorough them, so it is now important to build personalized recommendation systems to help users reach most interesting information on these networks. In this research we propose to build a recommendation system based on conceptual representation of text in the social networks, as well as investigating the role of the context of the concepts in the text using deep learning techniques, specifically, recurrent neural networks. We will build the methodology, then we study the results of applying this methodology on a real data extracted from one of social networks, to test the effectiveness of this methodology. Then we should compare the results with the state-of-the-art researches in order to prove the efficiency of the proposal.

Abstract