文献类型：会议论文

英文题名：Mutual Information-Based Feature Selection and Ensemble Learning for Classification

作者：Qi, Chengming[1,2];Zhou, Zhangbing[1,3];Wang, Qun[1];Hu, Lishuan[1,2]

第一作者：Qi, Chengming;亓呈明

通讯作者：Zhou, ZB[1];Zhou, ZB[2]

机构：[1]China Univ Geosci Beijing, Sch Informat Engn, Beijing, Peoples R China;[2]Beijing Union Univ, Sch Automat, Beijing, Peoples R China;[3]TELECOM SudParis, Comp Sci Dept, F-91001 Evry, France

第一机构：China Univ Geosci Beijing, Sch Informat Engn, Beijing, Peoples R China

通讯机构：[1]corresponding author), China Univ Geosci Beijing, Sch Informat Engn, Beijing, Peoples R China;[2]corresponding author), TELECOM SudParis, Comp Sci Dept, F-91001 Evry, France.

会议论文集：International Conference on Identification, Information and Knowledge in the Internet of Things (IIKI)

会议日期：OCT 20-21, 2016

会议地点：Beijing, PEOPLES R CHINA

语种：英文

外文关键词：Ensemble; Hyperspectral image; Multiple kernel learning; Mutual Information

摘要：Feature selection approaches aim to maximize relevance and minimize redundancy to the target by selecting a small subset of features in classification. This paper proposes a feature selection method based on mutual information (MI). We select a feature subset with minimal redundancy maximal relevance criteria. Multiple kernel learning (MKL) and ensemble learning (EL) have been applied in hyperspectral image classification. Our method applies Adaptive Boosting (AdaBoost) approach to learning multiple kernel-based classifier for multi-class classification problem. Classification experiments with a challenging Hyperspectral imaging (HSI) task demonstrate that our approach outperforms current state-of-the-art HSI classification methods.