INVESTIGATING SELECTED EGOCENTRIC USERS ON SOCIAL MEDIA PLATFORMS USING SOCIAL NETWORK ANALYSIS IN MINING FORENSIC EVIDENCE FOR LAW ENFORCEMENT IN KENYA

Abstract

Investigation of social media using social network theory is a new powerful tool that will aid and ease law enforcement agencies in multi-faceted ways in this ever evolving digital landscape. It is against this backdrop that this study focused on identifying and investigating selected individuals on Facebook and Twitter social media platforms. In particular, selected respondents from University of Eldoret, Kibabii, Moi, Kisii and Rongo Universities were involved in the study. The objective of the study was to demonstrate how Social Network Analysis (SNA) can be employed as an investigate tool to mine, analyse data from selected online social media users and present digital forensic evidence to aid law enforcement in Kenya. Particularly, the study aimed at identifying high degree nodes in the network and their behavioural patterns and profiles using visualizations, network metrics and user profile/demographic information. Social network analysis experimental research design was employed in this study. The sample size of the respondents was arrived at by employing Yamane’s formula of calculating sample size. The respondents were guided to create pseudoonline parody accounts in various social media platforms which was used to carry out the online data mining from the selected respondents to aid in social network analysis. The significance of the study was to fill the knowledge gap that hitherto not been researched by previous scholars yet it is imperative area as far as cyber-security and law enforcement is concerned in Kenya. Data mining and analysis was done using NodeXL, an Add-in tool in Ms-Excel for social network analysis. Computation of centrality measures, network clusters, cliques were presented using both infographic visualizations and centrality metrics of the respondents on egocentric networks Focal communication paths through which information flows in the network were also depicted. The findings demonstrated that Social Network Analysis can be effectively used on social media platforms to mine, analyse and present digital forensic evidence of individuals under investigation. The outcome of this study gives a new insight and techniques that can help law enforcement agencies and related stakeholders to identify or detect important individuals, subgroups, interaction patterns between subgroups and roles they play in a given network. The findings presented in this research illustrates how social network analysis can be used to determine the interpersonal connections, importance of actors in a given social network and detect communities of people and principally how law enforcement agencies can utilize this technique in identifying and tracking suspicious characters and ultimately help in maintaining law and order. SNA ought to be embraced as a supplement of conventional investigation, not necessarily replacing it.