Identifying Emerging Research Topics in Computer Science Using Overlapping Community Detection on Graph Neural Network Predicted Graphs

Constantinos Djouvas, Christos Christodoulou, Antonis Charalampous, Nicolas Tsapatsoulis

Abstract: Identifying emerging research topics is a challenging task. This study presents a novel methodology for the future prediction of emerging research topics, employing Graph Neural Networks (GNNs). The proposed framework is based on the construction of co-keyword graphs, serving as input for our innovative machine learning model designed to forecast keyword frequencies in forthcoming time periods. The proposed methodology is validated by applying it to forecast emerging topics and comparing the results against an observed year, 2022 in our case. Our findings underscore the potential of employing sophisticated computational methods to uncover emergent themes and offer valuable information for researchers, policymakers, and practitioners seeking to anticipate and navigate the dynamics of contemporary research.

External IDs:doi:10.1109/picom64201.2024.00028