Detecting difference of usage of terms as difference of structure

  • Authors:
  • Tetsuya Yoshida;Teruyuki Kondo;Shogo Nishida

  • Affiliations:
  • Graduate School of Information Science and Technology, Hokkaido University, West 9, North 14, Kita, Sapporo, Hokkaido 060-0814, Japan;Graduate School of Engineering and Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan;Graduate School of Engineering and Science, Osaka University, 1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan

  • Venue:
  • Cognitive Systems Research
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a method for detecting difference of usage of terms aiming at supporting collaboration of people with different backgrounds. Generally people tend to have different ways of conception, especially when they have different backgrounds and specialties. Since they usually communicate with each other using terms or symbols, such difference can be manifested as difference of usage of terms. In our approach people with different backgrounds are required to specify their knowledge as a set of classification cases. The specified cases play the role of their individual domain knowledge and they are structured into decision trees. Six algorithms are proposed for detecting difference based on difference of the structure of constructed decision trees and these are implemented on a prototype system. Based on the candidates suggested by the system, users of the system discuss each other on difference and modify their usage of terms so that difference can be gradually removed by repeating the human-computer interaction. Experiments were conducted by simulating difference using an artificial dataset and datasets from the University of California Irvine Data Repository. The results are encouraging.