Data Stream Mining Based Dynamic Link Anomaly Analysis Using Paired Sliding Time Window Data
Abstract
Dynamic network analysis for network security is challenging because it is computationally expensive to extract knowledge structures for quantifying the security levels of dynamic networks. There has been an increased interest in dynamic network analysis for network security and it is an emergent scientific field in network science. In this report, we introduce network analytics metrics and sliding time window data structures for data stream mining in order to incorporate link anomaly detection into the dynamic network analysis. The proposed dynamic link anomaly detection framework provides the capability to construct effective knowledge structures by measuring the security levels of dynamic networks, and filtering anomalous or suspicious links from network flow data. In addition, the sliding time window based method produces useful processed stream data for generalized dynamic network analysis.
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 2014
- Accession Number
- ADA613504
Entities
People
- Keesook Han
- Qi Liao
- Zhang Tao
Organizations
- Air Force Research Laboratory