Visualization and Analysis of the Data Mining Domain of Scientific Medical Publications in the Web of Science Database Using Co-word Analysis

Abstract

Background: The present study aims at visualizing and analyzing the scientific productions of data mining in medical sciences in the Web of Science (WoS) database to identify knowledge structure in this field. Methods: The present study was carried out using co-word analysis, which is a scientometrics method. Also, the analysis of the WoS approach was used for the visualization and analysis of word co-occurrence networks in the data mining domain. The research population consisted of 14,430 records related to data mining published in WoS from 2008 to 2016. The data were analyzed after retrieval using VoSviewer scientometric software. Results: The research findings showed that the average annual growth rate of scientific productions in the data mining field had an upward trend and reached its peak in 2016. The findings also showed that the USA was the most productive country, with 8123 degrees, claiming the first rank by 27.4% of the total scientific production, and Iran was in the 13th rank with 678 degrees. Among universities and institutions, the Chinese Academy of Sciences (692 degrees), the University of California (621 degrees), and the National Center for Scientific Research (CNSR) (514 degrees) were ranked first to third globally. Among Iranian institutions and universities, Islamic Azad University (IAU), Tehran University, and Iran University of Technological Sciences were ranked 41st, 94th, and 200th in the world with 140, 105, and 66 degrees, respectively. The results of the current research also showed that the field of data mining was formed based on four vocabulary clusters with 108, 82, 73, and 42 words, respectively. Conclusions: The recognition of Iran as a successful country in the field of data mining should encourage researchers from other countries to communicate with Iranian scientists.