International Journal of Information Dissemination and Technology
  • Year: 2020
  • Volume: 10
  • Issue: 3

Sentiment analysis research: A scientometrics assessment of global publications for the period 2003-2020

1Formerly with CSIR-NISTADS, New Delhi

2Formerly with CSIR-NPL, New Delhi

*Corresponding Author, B.M. Gupta bmgupta1@gmail.com

Online published on 6 January, 2021.

Abstract

This study has sought to evaluate the global sentiment analysis research (14025 records) on quantitative and qualitative metrics to understand the current status of research in the subject at global, national, institutional, and individual author level. The data for the study was sourced from the Scopus database covering the period 2003-2020. The global research in the subject registered an 81.15% annual average growth and averaged 11.91 citations per paper. India, China, and USA lead the global ranking of the top 10 most productive countries, with (18.37%, 17.71%, and 13.36%) respectively as their global publications share. The study identifies the most productive and most impactful research organizations and authors in the subject, profiles the highly cited papers, the share of international collaborative papers and the most productive source journals, and the most productive organizations as well as the leading authors. The study concludes that international collaboration research contributes significantly to the quality of research in sentiment analysis studies.

Keywords

Sentiment Analysis, Machine Learning, Global Publications, Scientometrics, Bibliometrics