Influenced Citation Analysis Using Modified Word Movers Distance (MWMD)

Abstract

Citation analysis is an essential part of the research to inculcate the potentiality of the work and giving credit to the researcher. Generally, citations are used by researchers for many purposes, however mainly it is used to show that the references influenced author's work. Now a day's most of the references influenced one way or other. The proposed model is used to find influenced citations in the scientific documents using a semantic approach rather than simple keyword matching. Generally, every scientific document contains references or a bibliography at the end of the document. The goal of the proposed technique is to discover the relevancy among scientific document and a list of reference papers in the same document in a semantic way. In this paper, the proposed work developed using the proposed semantic similarity measure called Modified Word Movers Distance (MWMD) applied to the deep learning model. Finally, the proposed work efficiently finds influenced citations percentage in the scientific documents.