Academic conference analysis for understanding country-level research topics using text mining

Kwanho Kim, Sue Kyung Lee, Heemin Park, Jinseok Chae

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

The importance of academic conferences is getting intensively larger as a way to publish the up-to-date research results on each particular research topics in a fast manner unlike journals, and the number of conferences tends to be increased year by year. Moreover, since a conference information, mostly accessible on the Internet, contains not only topics but also geographical areas where the conference was held, these are considered as a valuable source to understand the research trends according to countries. In this paper, we aim to develop methods for analyzing country-level research trends and the relationships among the countries by using text mining and clustering techniques. Specifically, we collected conference information from 8,957 websites from 2015 to 2017, and we found three clusters of countries according to their distributions of topics and the similarities among them. The experimental results show that some countries focus on various topics ranging from social science and medicine, while the others mainly concentrated on some particular topics such as engineering. Moreover, we found country groups that show quite similar in terms of topics. For instance, the following three country groups are found (Philippines, Indonesia, Thailand), (China, Japan, Hong Kong), and (Austria, Czech Republic, Netherlands).

Original languageEnglish
Pages (from-to)1-16
Number of pages16
JournalInternational Journal of Computer Information Systems and Industrial Management Applications
Volume11
StatePublished - 2019

Keywords

  • Academic conference analysis
  • Big data analysis
  • Country clustering
  • Data mining
  • Text mining
  • Topic analysis

Fingerprint

Dive into the research topics of 'Academic conference analysis for understanding country-level research topics using text mining'. Together they form a unique fingerprint.

Cite this