Open Access Open Access  Restricted Access Subscription Access

Feature Selection with Multi-Cost Constraint

Jing-Kuan Li,
Hong Zhao,
William Zhu,


Cost-sensitive learning extends classical machine learning and data mining by considering various types of costs, of the data. Due to money limited, we also have a constraint on the cost for selecting feature and tradeoff between the test costs and misclassification costs. However, the precious works seldom involve the tradeoff problem under the multi-cost constraint. In this paper, we introduce the misclassification costs into the cost constraint problem firstly, and propose a quadratic heuristic algorithm to deal with the minimal feature selection with multi-cost constraint problem. The goal is to obtain a feature subset with minimal average total cost, which includes test costs and misclassification costs. Experimental results indicate the proposed algorithm is effective and efficient.


Feature selection; Granular computing; Misclassification costs; Multi-cost constraint; Heuristic algorithm

Citation Format:
Jing-Kuan Li, Hong Zhao, William Zhu, "Feature Selection with Multi-Cost Constraint," Journal of Internet Technology, vol. 17, no. 5 , pp. 981-991, Sep. 2016.

Full Text:



  • There are currently no refbacks.

Published by Executive Committee, Taiwan Academic Network, Ministry of Education, Taipei, Taiwan, R.O.C
JIT Editorial Office, Office of Library and Information Services, National Dong Hwa University
No. 1, Sec. 2, Da Hsueh Rd., Shoufeng, Hualien 974301, Taiwan, R.O.C.
Tel: +886-3-931-7314  E-mail: