计算机工程与应用2012，Vol.48Issue(22)：189-194,6.DOI:10.3778/j.issn.1002-8331.2012.22.038
高维空间中针对离群点检测的特征抽取
Feature extraction for outlier detection in high-dimensional spaces
张小燕 1胡昊 1苏勇1
作者信息
- 1. 江苏科技大学计算机科学与工程学院,江苏镇江212003
- 折叠
摘要
Abstract
This work addresses the problem of feature extraction for boosting the performance of outlier detectors in high-dimensional spaces. Recent years, the prominence of multidimensional data on which traditional detection techniques usually fail to work as expected due to the curse of dimensionality. This paper introduces an efficient feature extraction method can take advantage of both ERE and APCPA which brings nontrivial improvements in detection accuracy in outlier detection. Similar to APCDA, this approach performs engenspace decomposition as well as feature extraction on the weight-adjusted scatter matrices, and applies the strategy of ERE during the eigenspace reg-ularization process to preserve the discriminant information. Experiments carried out on real datasets demonstrate the feasibility of feature extraction in outlier detection.关键词
特征抽取/降维/离群点检测Key words
feature extraction/ dimensionality reduction/ outlier detection分类
信息技术与安全科学引用本文复制引用
张小燕,胡昊,苏勇..高维空间中针对离群点检测的特征抽取[J].计算机工程与应用,2012,48(22):189-194,6.
