Text Mining: Research Trend Analysis, Forecast and Citation Approach

Text Mining (TM) is one of the immerging areas of research, but there are limited studies from the view point of scientometric. Using the bibliometric approach, this paper analyses TM research trend, forecast and citation approach from 2000 to 2019 by locating headings “text mining”, “text clustering”, “text extraction” and “text categorization” in Web of Science database. The paper classified 5006 retrieved articles, using the following ten categories – publication year, citation, country, institution, type of document, language, subject, author, source title and key-word – for distribution status of different areas, in order to explore the trend of researches in this field during this period. According to K-S test, the result depicts that the set of data confirms to Lotka's Law is rejected at 0.01 level of significance. To do so, Pao’s formula and Least-square method are used. The research provides a roadmap for future researchers to follow, whether they can concentrate in the core categories where the possibility of success is lying.