Electronic Theses and Dissertations

Identifier

950

Date

2013

Document Type

Thesis

Degree Name

Master of Science

Major

Computer Science

Committee Chair

Dipankar Dasgupta

Committee Member

Scott Fleming

Committee Member

King-Ip Lin

Abstract

In this information era, social media and online social networks have become a huge data source. The social network perspective provides a clear way of analyzing the structure of whole social entities. These social media and online social networks are a virtual representation of real life as they represent real life relations between social actors (people). The primary focus of this study is to propose an algorithm and its implementation for clustering of multi-characteristic dynamic graphs in general, and multi-characteristic dynamic online social networks in specific. Social networks are typically stored as graph data (edges lists mostly), and dynamically changes with time either by expanding or shrinking. The topology of the graph data also changes along with the values for the relationships between nodes. Several algorithms were proposed for clustering, but only few of them deals with multi-characteristic and dynamic networks. Most of the proposed algorithms work for static networks or small networks and a very small number of algorithms work for huge and dynamic networks. In this study a practical algorithm is proposed which uses a combination of multi-objective evolutionary algorithms, distributed file systems and nested hybrid-indexing techniques to cluster the multi-characteristic dynamic huge social networks. The results of this work show a fast clustering system that is adaptive to dynamic interactions in social networks also provides a reliable distributed framework for BIG data analysis

Comments

Data is provided by the student.

Library Comment

Dissertation or thesis originally submitted to the local University of Memphis Electronic Theses & dissertation (ETD) Repository.

Download

Share

COinS