Detecting Research Fronts Using Neural Network Model for Weighted Citation Network Analysis

We investigated the performance of different types of weighted citation networks to detect emerging research fronts by comparing existing studies and constructed citation networks and divided them into clusters. We also applied measures to such weighted citations as the differences in publication years between citing and cited papers and the similarities of their keywords to effectively detect emerging research fronts. However, the functions that decide the edge weight in the citation networks were decided based on experiments. For automatically deciding the effective weight's functions that depend on the dataset characteristics, a learning method is important. In this paper, we propose a novel learning method based on neural networks for deciding the edge weights for citation networks. We evaluated our proposed method in three research domains: gallium nitride, complex networks, and nano-carbon. We demonstrate that our proposed method has better performance of each approach than the existing methods by the following measures of extracted research fronts: visibility, speed, and topological and field relevance.