CLASSIFICATION ON MULTI-LABEL DATASET USING RULE MINING TECHNIQUE

Most recent work has been focused on associative classification technique. Most research work of classification has been done on single label data. But it is not appropriate for some real world application like scene classification, bioinformatics, and text categorization. So that here we proposed multi label classification to solve the issues arise in single label classification. That is very useful in decision making process. Multi-label classification is an extension of single-label classification, and its generality makes it more difficult to solve compare to single label. Also we proposed classification based on association rule mining so that we can accumulate the advantages of both techniques. We can get the benefit of discovering interesting rules from data using rule mining technique and using rule ranking and rule pruning technique we can classified that rules so that redundant rules can be reduced. So that Here proposed work is done on multi label dataset that is classified using rule mining algorithm. Here proposed approach is an accurate and effective multi label classification technique, highly competitive and scalable than other traditional and associative classification approaches.