Detecting word substitutions: PMI vs. HMM

ABSTRACT

Those who want to conceal the content of their communications can do so by replacing words that might trigger attention. For example, instead of writing "The bomb is in position", a terrorist may chose to write "The flower is in position." The substituted sentence would sound a bit "odd" for a human reader and it has been shown in prior research that such oddity is detectable by text mining approaches. However, the importance of each component in the suggested oddity detection approach has not been thoroughly investigated. Also, the approach has not been compared with such an obvious candidate for the task as Hidden Markov Models (HMM). In this work, we explore further oddity detection algorithms reported earlier, specifically those based on pointwise mutual information (PMI) and Hidden Markov Models (HMM).

REFERENCES

• 1: Jeff A. Bilmes , Katrin Kirchhoff, Factored language models and generalized parallel backoff, Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology: companion volume of the Proceedings of HLT-NAACL 2003--short papers, p.4-6, May 27-June 01, 2003, Edmonton, Canada [doi>10.3115/1073483.1073485]
• 2: British National Corpus (BNC), 2004. www.natcorp.ox.ac.uk.
• 3: Lee, H. and Ng, A.Y. Spam deobfuscation using a Hidden Markov Model. In Proceedings of the Second Conference on Email and Anti-Spam, 2005.
• 4: S. Fong, D. Skillicorn, and D. Roussinov, "Measures to detect word substitution in intercepted communication," IEEE International Conference on Intelligence and Security Informatics, ISI 2006, San Diego, CA, USA, May 23-24, ser. LNCS 3975.
• 5: Dmitri Roussinov , Leon J. Zhao , Weiguo Fan, Mining context specific similarity relationships using the world wide web, Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, p.499-506, October 06-08, 2005, Vancouver, British Columbia, Canada [doi>10.3115/1220575.1220638]
• 6: Skillicorn, D.B. Beyond keyword filtering for message and conversation detection. In IEEE International Conference on Intelligence and Security Informatics (ISI2005), pages 231--243. Springer-Verlag Lecture Notes in Computer Science LNCS 3495, May 2005.
• 7: Peter D. Turney, Mining the Web for Synonyms: PMI-IR versus LSA on TOEFL, Proceedings of the 12th European Conference on Machine Learning, p.491-502, September 05-07, 2001

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval
          Subjects: Information filtering

General Terms:
Algorithms

Keywords:
detecting substitutions, security informatics, text mining, text oddity