Adaptive Decision Tree Algorithm Based on Fusion of Decision Costs and Attribute Preference

Jia-jun CHEN,Duo-qian MIAO

DOI: https://doi.org/10.3969/j.issn.1000-1220.2018.06.016

2018-01-01

Abstract:In view of the fact that the existing decision tree models have not fully considered the decision accuracy,attribute preference behavior of decision makers and decision risk factors during the classification process,this paper proposes an adaptive decision tree algo-rithm based on fusion of decision costs and attribute preference. Combining decision-theoretic rough set and cost sensitive learning prob-lems,by introducing the misclassification decision costs into the probabilistic approximations of the target,the model of decision-theoretic rough set is then sensitive to cost. Introducing the decision loss function and the concept of user preference degree because different users tend to have different preferences for each attribute in the decision table ,and according to the Bayesian minimum-risk decision rules the decision costs are calculated. Through constructing adaptability-degree function as heuristic function is used to select test attributes,so as to establish a decision tree model. In the process of constructing the decision tree,the confidence factor is put forward to prune the deci-sion tree and to prevent the generation of decision tree too large. The experimental results show that the decision tree algorithm is effec-tive,which can fully take into account the decision maker's preference and the cost of misclassification due to the uncertainty of decision making,the experimental parameters can enhance the adaptability of the algorithm and meet the needs of different applications.