We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Data fault detection algorithm based on multi-label classification in sensor network.
- Authors
ZHANG Zhen-hai; LI Shi-ning; LI Zhi-gang; ZUO Xue-wen
- Abstract
Multiple data faults may occur at the same time in sensor network. In order to detect these data faults simultaneously, this paper modeled the data fault detection problem as a multi-label classification task. To improve the performance of multi-label classifiers in detecting data faults, it proposed a feature selection method based on multi-label ReliefF and genetic algorithm (MLRG). The method extended the ReliefF to the multi-label ReliefF which could estimate the quality of feature subset. MLRG firstly searched for a feature subset and then evaluated the feature subset using the multi-label ReliefF1. It performed experiments on the MLRG using three multi-label classifiers and compared it with other feature reduction algorithms. The experimental results show that MLRG can promote the performance of multi-label classifiers significantly in data faults detection.
- Subjects
DEBUGGING; CLASSIFICATION algorithms; SENSOR networks; DATA analysis; GENETIC algorithms; DATA reduction
- Publication
Application Research of Computers / Jisuanji Yingyong Yanjiu, 2014, Vol 31, Issue 12, p3788
- ISSN
1001-3695
- Publication type
Article
- DOI
10.3969/j.issn.1001-3695.2014.12.069