Intelligent Data Engineering and Automated Learning - IDEAL 2009

Volume 5788 of the series Lecture Notes in Computer Science pp 586-593

A Soft Discretization Technique for Fuzzy Decision Trees Using Resampling

  • Taimur QureshiAffiliated withLaboratory ERIC, University of Lyon 2
  • , D. A. ZighedAffiliated withLaboratory ERIC, University of Lyon 2

* Final gross prices may vary according to local VAT.

Get Access


Decision trees generate classifiers from training data through a process of recursively splitting the data space. In the case of training on continuous-valued data, the associated attributes must be discretized into several intervals using a set of crisp cut points. One drawback of decision trees is their instability, i.e., small data deviations may require a significant reconstruction of the decision tree. Here, we present a novel soft decision tree method that uses soft of fuzzy discretization instead of traditional crisp cuts. We use a resampling based technique to generate soft discretization points and demonstrate the advantages of using our resampling based soft discretization over traditional crisp methods.


Crisp discretization resampling fuzzy discretization