REU Site on Integrated Machine Learning Systems:
Malware Detection


Terran Lane Dean F. Hougen

Collaborative filtering for malware detection (Lane, Hougen). Many classes of malicious software downloads (a.k.a., malware) propagate through user-initiated access to hostile web pages, trojan-infected emails, or downloads of malicious code masquerading as innocuous programs. Modern operating systems offer some hooks to block such accesses, typically through an approval requester, but users do not have enough information to make an informed decision. This project develops a collaborative filtering system that draws information from multiple systems to present more informative dialogs. We use anomaly detection methods to detect disruptions to users' systems and collaborative filtering methods to integrate these detections into a global view of the safety of downloadable objects.

REU Home
Back to Projects

fagg [[at]]

Last modified: Wed Feb 20 00:04:00 2008