Communications of the ACM
Rough-Fuzzy Hybridization: A New Trend in Decision Making
Rough-Fuzzy Hybridization: A New Trend in Decision Making
Data mining, rough sets and granular computing
Data mining, rough sets and granular computing
Hi-index | 0.00 |
Rough Set theory and its extensions are found to be very effective in Knowledge Discovery in Data Bases. A number of tools are available in RST to solve Data Mining tasks such as clustering, rule mining, handling missing values and elimination of redundant data. In Data Mining handling of information tables with missing data values plays a very important role as missing values reduces the quality of information extracted. In this paper we discuss three different approaches to handle missing values. First method is called RSFit approach which predicts the missing attribute values based on a distance function.Second method, called Characteristic set based approach, provides decision rules from incomplete information systems. Finally a novel approach is introduced for constructing decision rules. This is based on a similarity relation. Experiment with small data set shows that the new approach is slightly better than the second method.