COMMUNITY DETECTION FROM NETWORK DATA USING THE LOUVAINALGORITHM IN A WEB-BASED ENVIRONMENT

A fundamental component of network science, community detection provides a prism through which the structuralarrangement of intricate systems can be comprehended. Such analysis can provide important insights into theunderlying dynamics of biological, technological, and social systems by identifying clusters of nodes with highinternal connectivity in comparison to the rest of the network. Using the Louvain algorithm, a scalable andeffective approach to modularity optimisation, this paper introduces a web-based application designed forinteractive community detection. The system is written in Python and incorporates several essential libraries:Matplotlib for results visualisation, NetworkX for graph processing and community detection, and Flask forcreating an intuitive web interface. The Louvain method is used to process network data that users upload informats like edge lists or adjacency matrices. The output facilitates intuitive investigation of the network structureby providing community assignments and associated visual representations. This application's accessibility is oneof its unique features. This platform provides a simplified, interactive experience in contrast to traditional toolsthat require programming knowledge or command-line interaction. It is especially well-suited for people in nontechnical fields who need network analysis without high technical barriers, like educators, students, or researchersin the social and biological sciences. This work shows how scalable algorithms like Louvain can be incorporatedinto easily accessible platforms for wider interdisciplinary use, highlighting the usefulness of community detectionin real-world applications. Additionally, the system facilitates dynamic exploration and interpretation of intricatenetwork communities by enabling real-time interaction and visualisation.