PhD Dissertation Proposal by Bonaventure Chidube Molokwu:"A Framework for Link Dynamics in Social Networks using Graph Embeddings with Heuristics "

Thursday, December 10, 2020 - 11:00 to 12:30


The School of Computer Science is pleased to present… 

PhD Dissertation Proposal by: Bonaventure Chidube Molokwu 

Date:Thursday December 10, 2020 
Time: 11.00am - 12:30pm   
Passcode: If you are interested in attending this thesis event, please request this via the Graduate Secretary at 


Social Network Analysis (SNA) has become an appealing research topic, within the domain of Artificial Intelligence (AI), owing to the widespread diffusion of cybersocieties. These virtual communities are fuelled via interactions between actors that constitute the social-graph representation of the cybersocieties. Several literature in SNA have focused on studying and modelling the apparent and/or latent interactions within social graphs as an n-ary operation which yields binary/dyadic outputs comprising positives (friends, trusts, etc.) and negatives (foes, distrusts, etc.). Inasmuch as interactions constitute the bedrock of any given Social Network (SN) structure; there exist scenarios where an interaction, which was once considered a positive, transmutes into a negative as a result of one or more indicators which have affected the quality of the interaction. This transmutation at present, from an existent positive tie to a negative tie, is manually executed by the affected actors in the SN. In this regard, these manual transmutations can be quite inefficient, ineffective, and a catastrophic effect might have been incurred by the constituent actors and the SN structure prior to a resolution. The problem we attempt to solve herein differs from the conventional problems of link prediction and strength of ties. Strength of ties focuses on classifying ties either as strong or weak ties based on an influence metric; link prediction aims at forecasting newer ties; and our problem statement aims at flagging positive ties that should be considered for breakups/rifts (negative-tie state), as they tend to pose potential threats to actors and the SN. Therefore, we are proposing a unique multilayer framework with the capabilities of link prediction and breakup/rift prediction. 
Keywords: Breakup, Unlink, Dimensionality Reduction, Link Prediction, Representation Learning 

Thesis Committe:  

Internal Reader: Dr. Saeed Samet 
Internal Reader: Dr. Jianguo Lu   
External Reader: Dr. Dragana Martinovic                
Advisor: Dr. Ziad Kobti  

PhD Dissertation Proposal Announcement  


5113 Lambton Tower 401 Sunset Ave. Windsor ON, N9B 3P4 (519) 253-3000 Ext. 3716