Quantifying Individual Research's Distance from the Trends based on Dynamic Topic Modeling. Issue 1 (14th October 2022)

• Record Type: Journal Article
• Authors: Meng, Jie; Lou, Wen; He, Jiangen
• Is Part Of: Proceedings of the Association for Information Science and Technology. Volume 59:Issue 1(2022)
• Page Start: 762
• Page End: 763
• Publication Date: 2022-10-14
• Subjects: Dynamic topic modeling -- Hot topics -- NLP -- Research behavior -- Research trends; Information science -- Congresses; Information technology -- Congresses; Information science; Information technology; Conference papers and proceedings; 020
• DOI: 10.1002/pra2.718
• Languages: English
• ISSNs: 2373-9231

Abstract

ABSTRACT: Research trends are the keys for researchers to decide their research agenda. However, only few works have tried to quantify how scholars follow the trends. This paper addresses this problem by proposing a novel measurement for quantifying how a scientific entity (paper or researcher) follows the hot topics in a research field. Specifically, the topic evolution and papers are vectorizing by dynamic topic modeling. Then the degree of hotness tracing is explored from three different perspectives: mainstream, short‐term direction, long‐term direction. Papers and researchers in the field of Computer Vision from 2006 to 2017 were selected to evaluate our method. Further study will show the results of topic evolution patterns and researchers' clusters.