Multiple-options decision tree mining concept-drifting data streams
Ye Ai · Microcomputer & its Applications · 2010
This paper focuses on the concept-drifting data streams mining,and based on the CVFDT algorithm improvements proposed a multiple-choice decision-tree algorithm mCVFDT.In this algorithm multi-attribute selection mechanism is added to the node structure in an effort to overcome the CVFDT not automatically detect defects in the concept-drifting,while avoiding duplication of tree traversal algorithm to improve the classification accuracy and efficiency.Experimental results show that the algorithm increases with the number of examples in the classification accuracy than CVFDT algorithm has better performance.