A Concept Matrix-Based Approach for Research Paper Clustering

The research papers facing similar challenges are grouped. The researchers studying the finance sector’s challenges will be benefited by this approach by reducing the time and effort of reading multiple research papers. A concept matrix is for 47 research articles that represent the issues facing the field. Hierarchical Agglomerative Clustering (HAC) is used on the basis of 9 challenges. The HAC algorithm’s lowest entropy value is 0.31, purity is 0.74, and Entropy for the K-Means method ranges from 0.41 to 0.44 with purity ranging from 0.51 to 0.54. As a consequence, HAC is more effective and accurate than K-means clustering. The proposed study is useful to the researchers who are doing study related to the finance domain, the challenges-specific research papers. The clusters are formed based on the challenges such as portfolio management, stock prediction, credit score analysis, and foreign exchange rate.