An interpretation method for classification trees in bio-data mining

  • Authors:

  • Shigeki Kozakura;Hisashi Ogawa;Hirokazu Miura;Noriyuki Matsuda;Hirokazu Taki;Satoshi Hori;Norihiro Abe

  • Affiliations:

  • Graduate School of Wakayama University, Wakayama, Japan;Graduate School of Wakayama University, Wakayama, Japan;Faculty of Systems Engineering, Wakayama University, Wakayama, Japan;Faculty of Systems Engineering, Wakayama University, Wakayama, Japan;Faculty of Systems Engineering, Wakayama University, Wakayama, Japan;Monotukuri Institute of Technologist;Kyusyu Institute of Technology

  • Venue:
  • KES'06 Proceedings of the 10th international conference on Knowledge-Based Intelligent Information and Engineering Systems - Volume Part II
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This research describes the analysis of a decision tree to interactively make rules from data that includes noise. A decision tree is often used in data mining technology because classification rules generated by the decision tree form new knowledge for the domain. However, it is difficult for a non-specialist user of data analysis to discover knowledge even if the decision tree is presented to the user. Moreover, the target data for mining may have both discrete values and continuous values, and these may include a lot of noise and exceptions. To understand the rules, the user needs the intermediate results of mining. Our system has interactive functions to realize interactive mining, and in this research, we propose a data mining technique with an interpretation function so that the user can understand analyzed data from which a required rule is derived.