Content and expression-based copy recognition for intellectual property protection

ABSTRACT

Protection of copyrights and revenues of content owners in the digital world has been gaining importance in the recent years. This paper presents a way of fingerprinting text documents that can be used to identify content and expression similarities in documents, as a way of facilitating tracking of digital copies of works, to ensure proper compensation to content owners.The fingerprints we collected consist of surface, syntactic, and semantic features of documents. Because they reflect mostly how things are said, we call these features stylistic fingerprints. However, how things are said are not independent of what is said, therefore these features have predictive power with respect to both content and expression.We tested the ability of these stylistic fingerprints to identify content and expression similarities between documents using a corpus of translated novels. On this corpus, these fingerprints identified the source of a given book chapter (content) successfully 90% of the time and the translator of the chapter (expression) 67% of the time using ten-fold cross validation and decision trees.In comparison, fingerprints based on the vocabularies of documents recognized the source of a given book chapter accurately 93% of the time and the expression of a particular translator 61% of the time.We believe that the right fingerprints can identify modified and literal copies of works, securing revenues for content owners. Enabling the content owners to secure revenues from distribution of their works can alleviate the digital copyright problem and reduce the need to prevent distribution, giving a chance to solutions that promote uninhibited distribution and use of works by the public.

REFERENCES

• 1: Eric Brill, Some advances in transformation-based part of speech tagging, Proceedings of the twelfth national conference on Artificial intelligence (vol. 1), p.722-727, October 1994, Seattle, Washington, United States
• 2: Brinegar, C. S. 1963. Mark Twain and the Quintus Curtius Snodgrass Letters: A Statistical Test of Authorship, Journal of the American Statistical Association, 58, 85--96.
• 3: Buckley, C., Singhal, A., Mitra, M., and Salton, G. 1995. New retrieval approaches using SMART: TREC 4. TREC 1995.
• 4: Michael John Collins , Mitchell P. Marcus, Head-driven statistical models for natural language parsing, 1999
• 5: Stone, P. J. 2002. Descriptions of Inquirer Categories and Use of Inquirer Dictionaries. http://www.wjh.harvard.edu/~inquirer/homecat.htm
• 6: Gerovac, B. and Solomon, R. J. Protect Revenues, Not Bits: Identify Your Intellectual Property. http://www.cni.org/docs/ima.ip-workshop/Gerovac.Solomon.html
• 7: Glover, A. and Hirst, G. 1996. Detecting stylistic inconsistencies in collaborative writing. In: Sharples, Mike and van der Geest, Thea (editors), The new writing environment: Writers at work in a world of technology. London: Springer-Verlag.
• 8: Google.com
• 9: Hatzivassiloglou, V., Klavans, J., and Eskin E. 1999. Detecting Similarity by Applying Learning over Indicators. 37th Annual Meeting of the ACL, 1999.
• 10: Holmes, D. I. 1994. Authorship Attribution. Computers and the Humanities 28(2), 87--106.
• 11: McCallum, A. K. 1996. Bow: A toolkit for statistical language modeling, text retrieval, classification and clustering. http://www.cs.cmu.edu/~mccallum/bow.
• 12: Mendenhall, T. C. 1901. A mechanical solution of a literary problem. The Popular Science Monthly, Vol.LX no.7, pp.97--105.
• 13: Morton, A. Q. 1965. The Authorship of Greek Prose. Journal of the Royal Statistical Society, Series A, 128, 169--233.
• 14: Mosteller, F. and Wallace, D. L. 1963. Inference in an Authorship Problem. Journal of the American Statistical Association, 58, 275--309.
• 15: Netanel, N. W. Impose noncommercial Use Levy to Allow Free P2P File-Swapping and Remixing. TPRC. 2002.
• 16: Peng, R. 1999. Statistical Aspects of Literary Style. Bachelor's Thesis, Yale University.
• 17: Peng, R. and Hengartner, N. 2002. Quantitative Analysis of Literary Styles. UCLA, preprints #338.
• 18: Gerard Salton , Amit Singhal , Chris Buckley , Mandar Mitra, Automatic text decomposition using text segments and text themes, Proceedings of the the seventh ACM conference on Hypertext, p.53-65, March 16-20, 1996, Bethesda, Maryland, United States [doi>10.1145/234828.234834]
• 19: Shivakumar, N. and Garcia-Molina, H. 1996a. SCAM: A Copy Detection Mechanism for Digital Documents. Proc. of the 2nd Intern'l Conference on Theory and Practice of Digital Libraries, 1996.
• 20: Narayanan Shivakumar , Hector Garcia-Molina, Building a scalable and accurate copy detection mechanism, Proceedings of the first ACM international conference on Digital libraries, p.160-168, March 20-23, 1996, Bethesda, Maryland, United States [doi>10.1145/226931.226961]
• 21: Thisted, R. and Efron, B. 1987. Did Shakespeare Write a Newly-discovered Poem? Biometrika, 74, 445--455.
• 22: Ian H. Witten , Eibe Frank, Data mining: practical machine learning tools and techniques with Java implementations, Morgan Kaufmann Publishers Inc., San Francisco, CA, 2000

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.1 Content Analysis and Indexing
          Subjects: Indexing methods

Additional Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.1 Content Analysis and Indexing
          Subjects: Linguistic processing

  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.7 Natural Language Processing
          Subjects: Text analysis

  K. Computing Milieux
  K.4 COMPUTERS AND SOCIETY
      K.4.1 Public Policy Issues
          Subjects: Intellectual property rights

Keywords:
content and expression based copy recognition, copy recognition, distribution volume tracking, part-of-speech tagged features, semantic features, surface parsing, syntactic parsing