Communities serve as fundamental building blocks in understanding the structure and dynamics of complex systems, often exhibiting strong internal cohesion and weak external connections, with applications ranging from social networks and biological systems to information networks, and beyond. In this study, the authors explore the application of different techniques for community detection on a network comprising research articles. The network is built on top of essential metadata such as titles, publication years, and keywords. The authors employ state-of-the-art community detection methods in this work. The methodology involves the formation of networks using BERT embeddings [5], with varying thresholds of cosine similarity. The resulting networks are then analyzed using the applied community detection techniques to identify cohesive communities within them. The study aims to provide insights into the effectiveness of different community detection algorithms in the context of research article datasets, offering valuable implications for information retrieval and knowledge discovery.