A Novel Cross Over and Mutation with Concept Hierarchy on Classification Algorithms

Machine learning is the search for algorithms that reason from externally supplied instances to produce general hypotheses, which then make predictions about future instances, which is used to solve classification problems in many applications. this work perform the function by using OneR, Feature Selection, Attribute Oriented Induction (AOI). Concept hierarchies can be used to reduce the data collecting and replacing low-level concepts by higher level concepts. A new attribute induction paradigm and as improving from current attribute oriented induction. A novel star schema attribute induction will be examined with current attribute oriented induction based on characteristic rule and using cross over and mutation with concept hierarchy. Experimental result shows proposed method has high accuracy with less execution time using UCI repository datasets.