Readings in information visualization: using vision to think
Readings in information visualization: using vision to think
On-line unsupervised outlier detection using finite mixtures with discounting learning algorithms
Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining
Information visualization in data mining and knowledge discovery
Information visualization in data mining and knowledge discovery
Data Mining and Knowledge Discovery
Detecting Hostile Accesses through Incremental Subspace Clustering
WI '03 Proceedings of the 2003 IEEE/WIC International Conference on Web Intelligence
Detecting Interesting Exceptions from Medical Test Data with Visual Summarization
ICDM '03 Proceedings of the Third IEEE International Conference on Data Mining
| Hi-index | 0.00 |
In this paper, we propose a visualization method based on probabilistic clustering in order to detect hostile accesses to a Web site. A system administrator is required to monitor a huge amount of access log data in order to detect novel types of hostile accesses. Our PrototypeLines is a visualization method based on probabilistic clustering with a single parameter that must be tuned and has been successful in medical domain. Thus we believe that PrototypeLines is more attractive than conventional hostile access detection methods based on machine learning since each of the latter methods typically has many parameters that must be tuned. We modify our PrototypeLines for hostile access detection and investigate its performance by experiments with real data. Experimental results show that our method is effective in detecting hostile accesses since it provides a display of a large amount of access sessions in a compact manner emphasizing hostile accesses with warm colors.