Abstract
We propose an online and local outlier detection technique with low resource consumption based on an unsupervised centered quarter-sphere support vector machine for wireless sensor networks. Using synthetic data, we demonstrate that our technique achieves better mining performance in terms of parameter selection using difference kernel functions compared to an earlier o²ine outlier detection technique.
Original language | Undefined |
---|---|
Title of host publication | Third IEEE European Conference on Smart Sensing and Context, EuroSSC 2008 |
Place of Publication | ETH Zurich |
Publisher | ETH Zurich |
Pages | 25-26 |
Number of pages | 2 |
ISBN (Print) | 978-3-909386-76-5 |
Publication status | Published - 29 Oct 2008 |
Event | 3rd European Conference on Smart Sensing and Context, EuroSSC 2008 - Zurich, Switzerland Duration: 29 Oct 2008 → 31 Oct 2008 Conference number: 3 |
Publication series
Name | |
---|---|
Publisher | ETH Zurich |
Number | 08HT8968C |
Conference
Conference | 3rd European Conference on Smart Sensing and Context, EuroSSC 2008 |
---|---|
Abbreviated title | EuroSSC |
Country/Territory | Switzerland |
City | Zurich |
Period | 29/10/08 → 31/10/08 |
Keywords
- EWI-13592
- IR-65032
- METIS-252053