ID3: Incorporation of Fuzziness and Generation of Network Architecture.

Date of Submission

December 1999

Date of Award

Winter 12-12-2000

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science


Machine Intelligence Unit (MIU-Kolkata)


Mitra, Sushmita (MIU-Kolkata; ISI)

Abstract (Summary of the Work)

A new method of generating fuzzy knowledge-based network is described. Crude domain knowledge is extracted using the ID3 algorithm. The ID3 approach to classification consists of a procedure for synthesizing an efficient discriminatory tree for classifying patterns that have non numeric values. One of the problems with ID3 is that it cannot deal with numeric (continuous) data, which most practical problems have. In our work a method to use ID3 in continuous attribute case is proposed. The rules are generated in linguistic form. They are mapped to a fuzzy neural network. Topology of the layered network is automatically determined and the net is finally pruned to generate an optimal architecture. The frequency of samples, representative of a rule, is also taken into consideration. Fuzzy membership concept is incorporated at the sample level to handle uncertainty. This involves changing the decision function at the node level. A novel approach to map confidence factors of unresolved/ ambiguous nodes directly into a fuzzy neural network is also described.The effectiveness of the algorithm is demonstrated on a vowel recognition problem.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


This document is currently not available here.