Fundamenta Informaticae 98 (2010) 167–181 167 DOI 10.3233/FI-2010-222 IOS Press Feature Selection via Maximizing Fuzzy Dependency Qinghua Hu ∗ , Pengfei Zhu, Jinfu Liu, Yongbin Yang and Daren Yu Harbin Institute of Technology Harbin 150001, China huqinghua@hit.edu.cn Abstract. Feature selection is an important preprocessing step in pattern analysis and machine learning. The key issue in feature selection is to evaluate quality of candidate features. In this work, we introduce a weighted distance learning algorithm for feature selection via maximizing fuzzy de- pendency. We maximize fuzzy dependency between features and decision by distance learning and then evaluate the quality of features with the learned weight vector. The features deriving great weights are considered to be useful for classification learning. We test the proposed technique with some classical methods and the experimental results show the proposed algorithm is effective. Keywords: feature selection, distance learning, fuzzy rough sets, fuzzy dependency 1. Introduction In general classification learning tasks, samples are represented with a large number of features and a decision attribute. The aim is to construct discriminating functions based on the information hidden in these samples. Dependency between features and decision attributes lays the foundation for classification learning. It is easy to imagine that learning is useless if there is no dependency between the given features and decision. However, only few of features are relevant for predicting the decisions in some real-world tasks [7, 13, 21]. Given a large number of weakly relevant or redundant features, majority of learning techniques will be confused and the learning performance is reduced. Therefore feature selection is considered to be an important preprocessing step in pattern recognition and machine learning for improving learning performance and reducing training time [2, 6, 7, 8, 10, 15]. * Address for correspondence: Harbin Institute of Technology, Harbin 150001, China