Discovery of Influential Publications Using Research Article's Usage Context
Faisal Fayyaz Kiyani, Bushra Hamid, Memona Humayun, Siva Raja Sindiramutty, Sohini Chowdhury · 2024
Identification of influential research articles has been an area of interest for quite a long time. Usually, the importance of an article is measured by the number of citations it has gained. There are different techniques available to identify the importance of a given paper based on their number of citations and authors popularity (h-index), but these techniques do not truly reflect the importance of a paper with respect to their usage context. In this paper we are proposing a technique to identify the importance of a research article from it usage perspective and not just relying on its number of citations or H-Index. To identify the importance of a given paper we calculate and use the paper's context, in what sense a paper is cited. We employ machine learning techniques to classify a citation into positive, negative or neutral and based on these criteria we calculate paper importance. As a result of our research work, swe also produced and published a citations dataset manually classified into positive, negative and neutral classes.