Colloquium "Care through Social Network Analysis: A Study on Social Isolation Detection and Recommender Systems" By: Bahareh Rahmati

Friday, February 9, 2024 - 11:00 to 12:00

The School of Computer Science at the University of Windsor is pleased to present…

 

Colloquium Presentation By: Bahareh Rahmati

 

Care through Social Network Analysis: A Study on Social Isolation Detection and Recommender Systems

 

 

Date: Friday, February 9th, 2024

Time: 11:00 am – 12:00 pm

Location: Erie Hall, Room 3123

 

Abstract: Social network analysis (SNA) is a powerful tool for understanding and modelling the relationships between entities in a social system. This talk explores the application of social network analysis, particularly in graph-based recommendation systems. We explore the intersection of data science and artificial intelligence, focusing on developing AI models.

One significant application discussed in this work is the detection of social isolation within community-based palliative care networks. Social isolation is a critical public health concern, significantly impacting older adults and palliative patients, often worsened by factors such as disease and limited daily capabilities. In response to this challenge, we propose a framework that views social isolation as an outlier detection problem within community-based social graphs. Our approach involves mapping the network to an attributed weighted social graph, where each patient is intricately linked to informal and formal care providers. Using formulae and indices, the societal norm is extracted by evaluating structural connections and quality-of-life features. The algorithm is validated using real-life data from the Windsor Essex Compassion Care Community (WECCC) and synthetic social graphs.

Future work will focus on developing a recommender system to suggest community resources to individuals to mitigate the impact of social isolation. We discuss the potential integration of deep learning and sequential recommender systems. The main goal is to identify and recommend the optimal combination of community support events and practices based on multiple objectives and criteria.

 

Keywords: Social Network Analysis, Recommender Systems, Graph-based Recommendation Systems, Data Science, Artificial Intelligence, Social Isolation Detection, Palliative Care Networks, Sequential Recommender System

 

Biography: Bahareh Rahmati is an enthusiastic Ph.D. candidate who started her program at the School of Computer Science in January 2021. Her research is in the field of data science and AI, with a focus on graph-based recommendation systems and explainable AI. She has currently published multiple papers in top-tier conferences.