Improved Monotonic Decision Trees Algorithms

The monotonic classification problem is commonly in real life. Different from the generalclassification problem, the monotonic classification classifiers should satisfy the monotonicconstraint: if all values of the condition attributes of the sample A is not less than the ones ofcorresponding condition attributes of B, then the values of decision attribute A is not less thanthe ones of decision attribute B.By analyzing the problems of the monotonic classification problems, in order to deal withsome of them, this paper proposed the improved monotonic decision tree algorithm. The mainworks include the following two aspects:（1）For the induction of decision tree for continuous-valued attributes, the selection ofcut should ensure that the decision tree meet the monotonic constraint. After the cut pointsplitting the decision tree, as long as ensuring that the values of the decision attribute of thesamples in the left branch is not more than the values of the decision attribute of the samplesin the right branch, the decision tree can be guaranteed to meet the monotonic constraints.Based on this idea and the theory of dominance rough set, this paper proposes a monotonicdecision tree algorithm. Experimental results show that the decision tree generated by ouralgorithm can meet the monotonic constraints. Compared with the classical decision tree, theproposed algorithm has higher classification accuracy.（2）In order to improve the generalization ability of the monotonic decision treealgorithm, based on the matrix, this paper proposed a extended model of monotonic decisiontree, which can deal with the monotonic classification problems with discrete values.