Peer-Reviewed Publications from NortonLifeLock Research Group
ACM Transactions on Information and System Security (TISSEC) (Volume 16 Issue 4, April 2014)
We present an extended version of Exposure and the experimental results on 17 months of its deployment on real data.
In Proceedings of the 2nd IEEE International Conference on Big Data 2014 (IEEE BigData 2014)
We introduce a new framework called MR-TRIAGE leveraging multi-criteria data clustering (MCDC) to perform scalable data clustering on large security data sets and further implement a set of efficient algorithms in a 3-stage MapReduce paradigm.
Journal of Neural Computing and Applications, Volume 25, Issue 7–8, December 2014
We proposed a new approach which applies the mass assignment-based fuzzy association rules mining (MASS-FARM) algorithm to Twitter data analysis, for the first time, to automatically extract useful and meaningful knowledge from large-scale data set.
In Proceedings of the 8th Workshop on Syntax, Semantics and Structure in Statistical Translation (SSST-8)
In Proceedings of the 35th IEEE Symposium on Security and Privacy Workshops (SP ‘14)
In Proceedings of the 9th ACM Symposium on Information, Computer and Communications Security (ASIA CCS '14)
We present a comprehensive study on the effectiveness of risk prediction based only on the web browsing behavior of users.
In Proceedings of the 25th International Conference on Computational Linguistics (COLING 2014)
In Proceedings of the 43rd IEEE “International Conference on Communications: Communications and Information Systems Security Symposium (ICC 2014)
This paper discusses the challenges of Internet routing anomalies and BGP hijacks investigations. With the help of a real-world potential BGP hijack case study, we describe our investigation process and highlight the challenges and limitations faced.
In Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘14)
We present AESOP, a scalable algorithm that identifies malicious executable files by leveraging a novel combination of locality-sensitive hashing and belief propagation. AESOP attained early labeling of 99% of benign files and 79% of malicious files with a 0.9961 true positive rate at 0.0001 false positive rate.