Comparing Temporal Behavior of Phrases on Multiple Indexes with a Burst Word Detection Method

Authors:
Hidenao Abe;Shusaku Tsumoto
Affiliations:
Department of Medical Informatics, Shimane University, School of Medicine, Shimane, Japan 693-8501;Department of Medical Informatics, Shimane University, School of Medicine, Shimane, Japan 693-8501
Venue:
RSFDGrC '09 Proceedings of the 12th International Conference on Rough Sets, Fuzzy Sets, Data Mining and Granular Computing
Year:
2009

Citing 5
Cited 1

Bursty and Hierarchical Structure in Streams

Data Mining and Knowledge Discovery
Discovering evolutionary theme patterns from text: an exploration of temporal text mining

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Numerical time-series pattern extraction based on irregular piecewise aggregate approximation and gradient specification

New Generation Computing
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
Clustering of time series data-a survey

Pattern Recognition

Evaluating a temporal pattern detection method for finding research keys in bibliographical data

Transactions on rough sets XIV

Quantified Score

Hi-index	0.00

Visualization

Abstract

In temporal text mining, some importance indices such as simple appearance frequency, tf-idf, and differences of some indices play the key role to identify remarkable trends of terms in sets of documents. However, most of conventional methods have treated their remarkable trends as discrete statuses for each time-point or fixed period. In order to find their trends as continuous statuses, we have considered the values of importance indices of the terms in each time-point as temporal behaviors of the terms. In this paper, we describe the method to identify the temporal behaviors of terms on several importance indices by using the linear trends. Then, we show a comparison between visualizations on each time-point by using composed indices with PCA and the trends of the emergent terms, which are detected the burst word detection method.