Date of Award

6-22-2022

Publication Type

Thesis

Degree Name

M.Sc.

Department

Computer Science

Supervisor

Pooya Moradian Zadeh

Supervisor

Ziad Kobti

Abstract

Human migration is one of the challenging issues facing today’s world. It can happen for different reasons and at different levels, from one country to another or between societies. As a result of social collaboration and knowledge exchange, a migrant’s opinion about a topic can change over time. These collaborations lead to network evolution, which affects each individual’s decisions in society. In this thesis, we propose a novel computational model to study the opinion dynamics of a migrated individual in a multi-population social network. In this model, an individual migrates from one society to another with different beliefs and values. Each individual is able to take actions and has different opinions regarding various topics. We use Social Network Analysis (SNA) to investigate the opinion evolution of the migrated individual. Two algorithms, Belief-based and Learning-based, have been proposed to consider the opinion dynamics in which the migrated individual is under the effect of their neighbors’ opinions, actions, and their origin and new society’s social norms. Over time, the migrated individual tries to adapt to the new society by receiving feedback from their actions and collaborating with other people. The main objective of this thesis is to propose a new computational framework to investigate the effect of different factors on a migrated individual’s opinions and actions in multi-population social networks. We define various scenarios and situations to analyze the impact of different settings in the process of opinion dynamics. We have evaluated our proposed model by conducting several experiments on a couple of synthetic social networks. We have analyzed the impact of different selection mechanisms, the role of origin’s and destination’s social norms, as well as the neighbor’s opinion on the opinion dynamics of the migrated node. In addition, the role of learning and observation in the decision-making process has been studied. The results show that our model is capable of tracking the opinion dynamics of the migrated node in different scenarios and situations.

Download

Share

COinS