Discovering research key terms as temporal patterns of importance indices for text mining

Authors:
Hidenao Abe;Shusaku Tsumoto
Affiliations:
Department of Medical Informatics, Shimane University, School of Medicine, Izumo, Shimane, Japan;Department of Medical Informatics, Shimane University, School of Medicine, Izumo, Shimane, Japan
Venue:
KES'10 Proceedings of the 14th international conference on Knowledge-based and intelligent information and engineering systems: Part III
Year:
2010

Citing 3
Cited 0

Detecting Temporal Trends of Technical Phrases by Using Importance Indices and Linear Regression

ISMIS '09 Proceedings of the 18th International Symposium on Foundations of Intelligent Systems
Detecting Temporal Patterns of Importance Indices about Technical Phrases

KES '09 Proceedings of the 13th International Conference on Knowledge-Based and Intelligent Information and Engineering Systems: Part II
Clustering of time series data-a survey

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

For researchers, it is important to continue discovering and understanding key topics on their own fields. However, the analysis is almost depended on their experiences. In order to support for discovering emergent key topics as key terms in given textual datasets, we propose a method based on temporal patterns in several data-driven indices for text mining. The method consists of an automatic term extraction method in given documents, three importance indices, and temporal patterns based on results of clustering and linear trends of their centroids. Empirical studies show that the three importance indices are applied to the titles of two academic conferences about artificial intelligence field as sets of documents. After extracting the temporal patterns of automatically extracted terms, we compared the trends of the technical terms among the titles of the conferences.