MAPPING THE TERROR: A SOCIAL NETWORK ANALYSIS OF AL-QAEDA AND ITS AFFILIATES' OPERATIONS

Abstract

This dissertation presents an in-depth Social Network Analysis (SNA) of Al-Qaeda and its affiliates’ terrorist activities spanning from 2000 to 2020, offering new insights into the structural, temporal, and geospatial dimensions of transnational terrorism. Drawing on data collated from reputable sources—including RAND reports, the Global Terrorism Database (GTD), National Counterterrorism Center (NCTC) reports, and United Nations publications—the study constructs a comprehensive network capturing the multifaceted relationships among terrorist perpetrator groups, influential leaders, and targeted countries. Initially, an overall network is constructed using edge lists derived from both group-to leader and group-to-country interactions. This network is then analyzed using various centrality measures—degree, betweenness, closeness, and eigenvector centrality—to identify key nodes that serve as hubs and bridges within the network. By filtering the top 10% of nodes based on degree centrality, the analysis reveals that core entities such as Al Qaeda, AQAP, and AQIM consistently occupy central positions, facilitating communication, resource allocation, and operational coordination across diverse regions. Additionally, a bipartite network is developed that explicitly links perpetrator groups with the countries where terrorist incidents have occurred. Projections of this bipartite graph into single-mode networks uncover clusters of countries sharing similar terrorist threats and highlight overlapping operational domains among different groups. Temporal analysis is incorporated by segmenting the dataset into defined intervals, thereby illuminating the dynamic evolution of the network. This approach uncovers shifting centralities and the emergence or dissolution of sub-networks, reflecting the adaptive strategies employed by terrorist organizations in response to counterterrorism interventions and broader geopolitical shifts. Furthermore, the dissertation employs link prediction techniques, notably the Adamic/Adar index, to forecast potential future connections within the network. This predictive component is critical for anticipating emerging alliances and operational collaborations that may not yet be evident. Despite the absence of precise geographic coordinates, geospatial analysis is conducted at the country level by aggregating incident counts and mapping them onto global boundaries, which highlights regional hotspots and strategic operational zones. Collectively, the methodologies and findings of this study contribute to a deeper academic understanding of decentralized terrorist networks while providing practical recommendations for intelligence and counterterrorism agencies. The research underscores the resilience and adaptability of Al-Qaeda’s network and emphasizes the necessity of integrated, network-based strategies to effectively disrupt its operations.