Bibliometrics of sentiment analysis literature

Abstract

This article provides a bibliometric study of the sentiment analysis literature based on Web of Science (WoS) until the end of 2016 to evaluate current research trends, quantitatively and qualitatively. We concentrate on the analysis of scientific documents, distribution of subject categories, languages of documents and languages that have been more investigated in sentiment analysis, most prolific and impactful authors and institutions, venues of publications and their geographic distribution, most cited and hot documents, trends of keywords and future works. Our investigations demonstrate that the most frequent subject categories in this field are computer science, engineering, telecommunications, linguistics, operations research and management science, information science and library science, business and economics, automation and control systems, robotics and social sciences. In addition, the most active venue of publication in this field is Lecture Notes in Computer Science ( LNCS). The United States, China and Singapore have the most prolific or impactful institutions. A keyword analysis demonstrates that sentiment analysis is a more accepted term than opinion mining. Twitter is the most used social network for sentiment analysis and Support Vector Machine ( SVM) is the most used classification method. We also present the most cited and hot documents in this field and authors’ suggestions for future works.

A bibliometric analysis of topic modelling studies (2000–2017)

“…First, it is limited by its query terms. If a pertinent article does not use any of the query terms, particularly in its title and abstract, it will not be included in our data, which might lead to the incompleteness of our data [10]. However, we believe that only a small number of articles are missed and it will not significantly affect the results of our study.…”

“…Bibliometrics is a statistical method used to organise thematic information, evaluate scientific development, determine research impact, compare research performance and identify emerging research fronts [9]. It has been widely applied to summarise the development of a discipline, such as natural language processing (NLP), big data and deep learning [10,11]. For example, in the field of NLP, many bibliometric studies have been conducted to evaluate the research quality and productivity of the field.…”

“…Chen et al [9] explored the statistical characteristics, geographical distributions and collaboration relationships of NLP-empowered mobile computing research with bibliometric data from the Web of Science. In addition, Piryani et al [12], Keramatfar and Amirkhani [10] and Mäntylä et al [13] investigated the studies concerning sentiment analysis with bibliometric databases such as the Web of Science or Scopus. Many bibliometric studies have also been published on topics such as big data and deep learning.…”

Twenty Years of Entropy Research: A Bibliometric Overview

“…Nunen et al [ 29 ] applied VOSviewer to safety culture to identify its major areas, key journals, prolific nations, collaborative characteristics, etc. For more articles, please refer to [ 30 , 31 , 32 , 33 , 34 , 35 , 36 ].…”

Apps-based Machine Translation on Smart Media Devices - A Review

“…Although translation software still has a few problems with many unclear translation outcomes, Internet users are forced to accept the existing translation result. In efforts to make the translation better and reliable, linguist and computer scientists should joint hands and work together to address the difficulties and challenge of improving machine translation systems to express correct and precise information [2].…”