Treffer: Social network analytics with NetworkX and similar libraries. An exploratory and comparative study

Διπλωματική εργασία--Πανεπιστήμιο Μακεδονίας, Θεσσαλονίκη, 2024. ; This thesis explores the application and analysis of Social Network Analysis (SNA) using Python libraries, with a focus on link prediction algorithms applied to Twitter data from the 117th Congress. The study begins with an introduction to the significance of SNA in understanding complex networks and its broad range of applications, including marketing, political campaigning, and public health. Chapter 2 provides a comprehensive review of the theoretical foundations of SNA, including core concepts such as network structures, centrality measures, and community detection. These theories form the basis for analyzing and interpreting social networks and their dynamics. Chapter 3 examines various Python libraries utilized for SNA, with a particular focus on NetworkX, Gephi, and Python-igraph. The chapter details the functionalities of these libraries and their roles in constructing, visualizing, and analyzing social networks, highlighting their strengths and applications in research. In Chapter 4, a case study is presented, applying SNA to Twitter data from the 117th Congress. This chapter outlines the methodology for link prediction using different algorithms—Common Neighbors, Adamic-Adar Index, and Resource Allocation Index— and evaluates their performance in predicting potential connections within the network. The thesis concludes with Chapter 5, summarizing the findings and contributions of the research. It reflects on the effectiveness of the link prediction algorithms, discusses the implications of the results, and suggests potential areas for future research.