Worst-case discriminative feature selection

Shuangli Liao, Quanxue Gao, Feiping Nie, Yang Liu, Xiangdong Zhang

科研成果: 书/报告/会议事项章节会议稿件同行评审

11 引用 (Scopus)

摘要

Feature selection plays a critical role in data mining, driven by increasing feature dimensionality in target problems. In this paper, we propose a new criterion for discriminative feature selection, worst-case discriminative feature selection (WDFS). Unlike Fisher Score and other methods based on the discriminative criteria considering the overall (or average) separation of data, WDFS adopts a new perspective called worst-case view which arguably is more suitable for classification applications. Specifically, WDFS directly maximizes the ratio of the minimum of between-class variance of all class pairs over the maximum of within-class variance, and thus it duly considers the separation of all classes. Otherwise, we take a greedy strategy by finding one feature at a time, but it is very easy to implement and effective. Moreover, we utilize the correlation between features to help reduce the redundancy, and then WDFS is extended to uncorrelated WDFS (UWDFS). To evaluate the effectiveness of the proposed algorithm, we conduct classification experiments on many real data sets. In the experiment, we respectively use the original features and the score vectors of features over all class pairs to calculate the correlation coefficients, and analyze the experimental results in these two ways. Experimental results demonstrate the effectiveness of WDFS and UWDFS.

源语言英语
主期刊名Proceedings of the 28th International Joint Conference on Artificial Intelligence, IJCAI 2019
编辑Sarit Kraus
出版商International Joint Conferences on Artificial Intelligence
2973-2979
页数7
ISBN(电子版)9780999241141
DOI
出版状态已出版 - 2019
活动28th International Joint Conference on Artificial Intelligence, IJCAI 2019 - Macao, 中国
期限: 10 8月 201916 8月 2019

出版系列

姓名IJCAI International Joint Conference on Artificial Intelligence
2019-August
ISSN(印刷版)1045-0823

会议

会议28th International Joint Conference on Artificial Intelligence, IJCAI 2019
国家/地区中国
Macao
时期10/08/1916/08/19

指纹

探究 'Worst-case discriminative feature selection' 的科研主题。它们共同构成独一无二的指纹。

引用此