Real-time Anomaly Detection in High-Speed Time evolving Graphs
Abstract
The PI was successful in this research grant. The goal of this project was to 1) develop memory-efficient and accurate local triangle counting method in a multigraph stream using fixed/varying sampling rates, and 2) detect anomalies using triangle information. They created and tested two local triangle counting methods MASCOT and FURL. Experimental results demonstrate that FURL provides the best accuracy compared to the state-of-the-art algorithm in a memory-efficient way. The PI has 1 peer reviewed papery published and 1 currently in review as a direct result of this grant award.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 22, 2018
- Accession Number
- AD1069876
Entities
People
- U. Kang
Organizations
- Seoul National University