Graph-Driven Topic Discovery: Extracting Insights from Citation Networks

This paper presents a strategy for word generation that uses Latent Dirichlet Allocation (LDA) to filter community citation graphs. Using LDA in a citation network setting, the main aim is to identify the hidden patterns in academic communities. In addition to simplifying meaningful word extraction, this innovative method sheds light on the topical dynamics and patterns of knowledge distribution at work in academic cooperation. An exciting new direction for improving our awareness of interrelated academic issues and the spread of information could be the combination of LDA and community citation graphs.