Sankar K. Pal

Soft computing methods for data mining.

Different characteristics of soft computing and its relation with machine intelligence and pattern recognition are explained.


Emergence of data  mining  and knowledge discovery from pattern recognition point  of view is illustrated. Significance of  integrating various soft computing tools for efficient learning in this regard is described.

Two examples, demonstrating such integration of fuzzy sets, artificial neural networks, genetic algorithms and rough sets for (i) efficient classification, rule generation and rule evaluation, and (ii) granular case generation in case based reasoning problems, are provided along with  their application specific merits. Different features of these methodologies are extensively demonstrated along with comparisons on different real life data sets.

 The talk concludes explaining the relation of rough-fuzzy computing with computational theory of perceptions (CTP), challenges of data mining and the future scope of  research in bioinformatics and web mining.