Rough set based decision rule generation to find behavioural patterns of customers

Research paper by L Sumalatha, P Uma Sankar, B Sujatha

Indexed on: 18 Aug '16Published on: 18 Aug '16Published in: Sadhana


Rough sets help in finding significant attributes of large data sets and generating decision rules for classifying new instances. Though multiple regression analysis, discriminant analysis, log-it analysis and several other techniques can be used for predicting results, they consider insignificant information also for processing which may lead to false positives and false negatives. In this study, we proposed rough set based decision rule generation framework to find reduct and to generate decision rules for predicting the Decision class. We conducted experiments over data of Portuguese Banking institution. From the proposed method, the dimensionality of data is reduced and decision rules are generated which predicts deposit nature of customers by 90% accuracy.