Date of Award

3-10-2021

Publication Type

Master Thesis

Degree Name

M.Sc.

Department

Computer Science

First Advisor

Pooya Moradian Zadeh

Second Advisor

Saeed Samet

Keywords

Computational Model, Social Isolation, Social Network Analysis

Rights

info:eu-repo/semantics/openAccess

Abstract

The human being is a social creature and needs to communicate with others to share information, emotions, and fulfill its basic needs. Social isolation can be considered as a serious health risk issue which not only has unignorably negative impacts on the well-being and quality of life of individuals, but also it is harmful to healthy human development. In this research, a computational model and a couple of novel algorithms are proposed to address social isolation detection in social networks. In our model, a given community is represented by a weighted-directed social graph. An algorithm, SBSID (Structure-based Social Isolation Detection), is proposed to detect socially isolated individuals based on the graph's structure by finding the number of each individual's active friends and their influence on each other. On the other hand, each individual's demographic characteristics in our model are represented by a set of binary attributes. Consequently, another algorithm is proposed, FBSID (Feature-based Social Isolation Detection), to address social isolation based on the nodes' features in the social graph.We propose a couple of metrics and formulas to calculate society's norms based on the overall structure and attributes of the social graph. Structural characteristics and attributes of each individual are compared with the norm of society to identify socially isolated individuals. We have evaluated the performance of our proposed model and algorithms on a set of synthetic networks. The results show that our model is capable of finding socially isolated nodes in various sizes of graphs with high accuracy and efficiency.

Share

COinS