A Comparison of Community Clustering Techniques: Fruchterman-Reingold and Wakita-Tsurumi
Abstract
The study of social network analysis depends on the relationships of people and online community. The relationships define who they are and how they act. Personality, educational background, race, and ethnicity, all of these interact with the patterns of relationship. Thus, by observing and analyzing such patterns, people can reveal and answer many questions about society. The relationship can be visualized in many ways, e.g. online community. Fruchterman-Reingold is a standard method force-directed algorithm or spring embedders place vertices by assigning forces according to the edges connecting the vertices. Meanwhile, Wakita-Tsurumi is a clustering algorithm used for cluster detection. It uses the metric of modularity (Q) as a quality measure of division in a network, based on the idea that networks with inherent community structure deviate from random networks and that networks with high modularity have denser connections inside a community, but fewer connections between nodes of different communities. The comparison shows the ability of both techniques to recognize the community based on sociometric and its contents. Both of the graph detected the same number of vertices (363), edges (5814) and density (0.02975511).
Downloads
Downloads
Published
Issue
Section
License
Copyright (c) 2017 Irma Yuliana, S Sukirman, S Sujalwo
This work is licensed under a Creative Commons Attribution 4.0 International License.