Leveraging one-class SVM and semantic analysis to detect anomalous content

Ozgur Yilmazel, Svetlana Symonenko, Niranjan Balasubramanian, Elizabeth D Liddy

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

Abstract

Experiments were conducted to test several hypotheses on methods for improving document classification for the malicious insider threat problem within the Intelligence Community. Bag-of-words (BOW) representations of documents were compared to Natural Language Processing (NLP) based representations in both the typical and one-class classification problems using the Support Vector Machine algorithm. Results show that the NLP features significantly improved classifier performance over the BOW approach both in terms of precision and recall, while using many fewer features. The one-class algorithm using NLP features demonstrated robustness when tested on new domains.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science
EditorsP. Kantor, G. Muresan, F. Roberts, D.D. Zeng, F.-Y. Wang, H. Chen, R.C. Merkle
Pages381-388
Number of pages8
Volume3495
StatePublished - 2005
EventIEEE International Conference on Intelligence and Security Informatics, ISI 2005 - Atlanta, GA, United States
Duration: May 19 2005May 20 2005

Other

OtherIEEE International Conference on Intelligence and Security Informatics, ISI 2005
CountryUnited States
CityAtlanta, GA
Period5/19/055/20/05

    Fingerprint

ASJC Scopus subject areas

  • Computer Science (miscellaneous)

Cite this

Yilmazel, O., Symonenko, S., Balasubramanian, N., & Liddy, E. D. (2005). Leveraging one-class SVM and semantic analysis to detect anomalous content. In P. Kantor, G. Muresan, F. Roberts, D. D. Zeng, F-Y. Wang, H. Chen, & R. C. Merkle (Eds.), Lecture Notes in Computer Science (Vol. 3495, pp. 381-388)