Decision Tree Construction Algorithm Based on Multiscale Rough Set Model

Aiming at the repetitive sub tree, complicated structure, sensitive to noise and other problems for decision tree constructed by classical decision tree algorithms, this paper proposes a new decision tree construction algorithm based on multi scale rough set model. This algorithm brings in the concept of scale variable and scale function, which uses the index of approximate classification accuracy in different scales to select test attribute, using inhibitor to pruning the decision tree during the procedure of algorithm formation, avoiding the problem of too large size of decision tree generated by the proposed algorithm. The results show that decision tree structure constructed by this algorithm is simple, and has certain degree of anti-interference and can meet the decision accuracy requirements from different users.