Computer forensics based on particle swarm optimization in cloud computing

Recent years have witnessed cloud computing as a utility means for providing resources. As more and more enterprises and individuals applications are migrated into cloud platform, however, how to analyze the evidence of computer crime caused by malicious users is not a trivial question. In this paper, we conduct computer forensics based on particle swarm algorithm combining fuzzy k-means clustering. Specifically, the data analysis is the key link of computer forensics. Given a set of application logs, we use particle swarm algorithm to cluster users' actions into different classes. The extensive experiments present the practicality of the proposed mechanism with low computation overhead. Also low false positives and false negatives are demonstrated.