Welcome to the InfoSci Platform
Spam Image Clustering for Identifying Common Sources of Unsolicited Emails

Spam Image Clustering for Identifying Common Sources of Unsolicited Emails

Chengcui Zhang, Xin Chen, Wei-Bang Chen, Lin Yang, Gary Warner
Copyright: © 2011 |Pages: 17
ISBN13: 9781609605155|ISBN10: 1609605152|EISBN13: 9781609605162
DOI: 10.4018/978-1-60960-515-5.ch006
Cite Chapter Cite Chapter

MLA

Zhang, Chengcui, et al. "Spam Image Clustering for Identifying Common Sources of Unsolicited Emails." New Technologies for Digital Crime and Forensics: Devices, Applications, and Software, edited by Chang-Tsun Li and Anthony T. S. Ho, IGI Global, 2011, pp. 87-103. https://doi.org/10.4018/978-1-60960-515-5.ch006

APA

Zhang, C., Chen, X., Chen, W., Yang, L., & Warner, G. (2011). Spam Image Clustering for Identifying Common Sources of Unsolicited Emails. In C. Li & A. Ho (Eds.), New Technologies for Digital Crime and Forensics: Devices, Applications, and Software (pp. 87-103). IGI Global. https://doi.org/10.4018/978-1-60960-515-5.ch006

Chicago

Zhang, Chengcui, et al. "Spam Image Clustering for Identifying Common Sources of Unsolicited Emails." In New Technologies for Digital Crime and Forensics: Devices, Applications, and Software, edited by Chang-Tsun Li and Anthony T. S. Ho, 87-103. Hershey, PA: IGI Global, 2011. https://doi.org/10.4018/978-1-60960-515-5.ch006

Export Reference

Mendeley
Favorite

Abstract

In this article, we propose a spam image clustering approach that uses data mining techniques to study the image attachments of spam emails with the goal to help the investigation of spam clusters or phishing groups. Spam images are first modeled based on their visual features. In particular, the foreground text layout, foreground picture illustrations and background textures are analyzed. After the visual features are extracted from spam images, we use an unsupervised clustering algorithm to group visually similar spam images into clusters. The clustering results are evaluated by visual validation since there is no prior knowledge as to the actual sources of spam images. Our initial results show that the proposed approach is effective in identifying the visual similarity between spam images and thus can provide important indications of the common source of spam images.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.