Title: Detection of research trends from bibliographical data

Authors: Hidenao Abe; Shusaku Tsumoto

Addresses: Faculty of Information and Communications, Bunkyo University, 1100 Namegaya, Chigasaki, Kanagawa, 253-8550, Japan. ' Department of Medical Informatics, School of Medicine, Shimane University, 89-1, Enya-cho, Izumo, Shimane 693-8501, Japan

Abstract: In this paper, we propose a method for detecting temporal linear trends of technical terms based on importance indices. Recent years, electrical documents are published hourly, daily, monthly, annually, and irregularly for each purpose. Although the purposes of each set of documents are not changed, roles of terms and the relationship among them in the documents change temporally. In text mining, importance indices of terms such as simple frequency, document frequency including the terms, and TF-IDF of the terms, play a key role for finding valuable patterns in the documents with cross sectional manner. In order to detect such temporal changes, we combined an automatic term extraction method, importance indices of the extracted terms, and trend identification based on linear regression analysis. After implementing this strategy, our method detected emergent and subsiding linear trends of the extracted terms in a corpus of a research domain. By comparing this method with the existing burst detection method, we discuss the linear trends of terms including the several burst words.

Keywords: text mining; trend detection; term frequency; inverse document frequency; TF-IDF; Jaccard; matching coefficient; linear regression; technical terms; term extraction; burst detection.

DOI: 10.1504/IJDMMM.2012.048107

International Journal of Data Mining, Modelling and Management, 2012 Vol.4 No.3, pp.255 - 266

Published online: 23 Aug 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article