Title :
Multi-objective cost-sensitive attribute reduction
Author :
Bingxin Xu ; Huiping Chen ; Zhu, Wei ; Xiaozhong Zhu
Author_Institution :
Coll. of IOT Eng., Hohai Univ., Changzhou, China
Abstract :
Cost-sensitive learning is both hot and difficult in data mining and machine learning applications. Some research considers only one type of cost. Others convert two or more types of cost into the same unit, and then deal with a single-objective optimization problem. However, in many cases different types of cost cannot be converted. In this paper, we define and tackle multi-objective attribute reduct problem with multiple types of test cost. First, we compute all reducts of a decision system. Then, we separately calculate the money cost and time cost of these reducts and compare them according to the two kinds of test cost. Finally, the worse ones are removed. The remaining reducts form a Pareto optimal solution set. We tested our algorithm with three representative cost distributions on four UCI datasets. Experimental results indicate that a Pareto optimal solution set is usually very small compared with the size of all reducts. Hence our approach is effective in filtering out worse solutions and helping users in scheme selection.
Keywords :
Pareto optimisation; data mining; learning (artificial intelligence); Pareto optimal solution set; UCI datasets; cost sensitive learning; data mining; decision system; machine learning applications; multiobjective cost sensitive attribute reduction; single-objective optimization problem; Approximation methods; Cognition; Decision trees; Entropy; Pareto optimization; Rough sets; Cost-sensitive learning; attribute reduction; money cost; rough sets; time cost;
Conference_Titel :
IFSA World Congress and NAFIPS Annual Meeting (IFSA/NAFIPS), 2013 Joint
Conference_Location :
Edmonton, AB
DOI :
10.1109/IFSA-NAFIPS.2013.6608602