Abstract
Complex systems have features such as numerous nodes and edges, complicated and hierarchical relations and evolving with time. During their running time, complex systems are influenced by the internal and external factors which can lead to abnormal states. Finding out the outliers can effectively supervise the whole system. Here, we study a real-world complex dynamic complex system, observe the abnormal pattern based on entropy, and find out nodes which will lead to the system collapse by GROD algorithm.
Original language | English |
---|---|
Title of host publication | Proceedings of 2015 International Conference on Electrical and Information Technologies for Rail Transportation - Transportation |
Editors | Yong Qin, Limin Jia, Lijun Diao, Jianghua Feng, Min An |
Publisher | Springer Verlag |
Pages | 677-684 |
Number of pages | 8 |
ISBN (Print) | 9783662493687 |
DOIs | |
Publication status | Published - 2016 |
Event | 2nd International Conference on Electrical and Information Technologies for Rail Transportation, EITRT 2015 - Zhuzhou, China Duration: 28 Aug 2015 → 30 Aug 2015 |
Publication series
Name | Lecture Notes in Electrical Engineering |
---|---|
Volume | 378 |
ISSN (Print) | 1876-1100 |
ISSN (Electronic) | 1876-1119 |
Conference
Conference | 2nd International Conference on Electrical and Information Technologies for Rail Transportation, EITRT 2015 |
---|---|
Country/Territory | China |
City | Zhuzhou |
Period | 28/08/15 → 30/08/15 |
Keywords
- Complex system
- Data mining
- Dynamic graph
- Outlier detection
Fingerprint
Dive into the research topics of 'Outlier detection for time-evolving complex networks'. Together they form a unique fingerprint.Cite this
Zhang, H., Hu, C., & Wang, X. (2016). Outlier detection for time-evolving complex networks. In Y. Qin, L. Jia, L. Diao, J. Feng, & M. An (Eds.), Proceedings of 2015 International Conference on Electrical and Information Technologies for Rail Transportation - Transportation (pp. 677-684). (Lecture Notes in Electrical Engineering; Vol. 378). Springer Verlag. https://doi.org/10.1007/978-3-662-49370-0_70