Research on fuzzy semanteme of decision trees algorithms

Decision trees algorithms that have emerged based on semanteme have rigid division defects, so we research on fuzzy semanteme of decision trees algorithms. This paper proposes a new decision trees algorithm based on fuzzy semanteme named SFID3. By utilizing concept trees and fuzzy c-means algorithm to get memberships of continuous attributes values, and taking advantage of cloud model to obtain accuracies of memberships simultaneously, we solve problems that fuzziness is not taken into account in decision trees algorithms based on semanteme and fuzziness is not thorough. Among which, in order to make full use of each value of continuous attributes, we use unweighted pair-group method with arithmetic means (UPGMA) to promote hierarchies when generating concept trees, which can make hierarchies of concept trees much more reasonable. The experiment results prove that the new algorithm SFID3 is feasible and effective.