INCREMENTAL LEARNING DECISION TREE ALGORITHM FOR KNOWLEDGE DISCOVERY
Mohammed Moulana, Mohammed Ali Hussain · PONTE International Scientific Researchs Journal · 2016
A prominent learning discovery procedure is Data Mining. Decision trees are of the basic and intense decision making models in data mining. A single constraint in decision trees is the unpredictability and error rate. Motivated by human learning techniques, we suggest a decision tree structure which impersonates human adapting by performing consistent enhanced learning. In this paper, we propose a novel Incremental Learning Decision Tree (ILDT) technique taking into account human learning procedure. Far reaching trials, utilizing decision tree C4.5 as base classifier, demonstrate that the exactness of our system is similar to best in class systems.