Elamur Research⌕
☰
home / papers / 2014 / Authorship Attribution with Topic Models

Authorship Attribution with Topic Models

Paper · 2014 · Yanir Seroussi, Ingrid Zukerman and Fabian Bohnert

Summary

The paper addresses closed-set authorship attribution: assigning an anonymous text to one of several known candidate authors whose training texts are available. It focuses on difficult settings, such as short online texts (e-mails, blog posts, reviews) and large pools of candidate authors. The authors build author representations from topic models. They explore new ways of applying LDA and the Author-Topic (AT) model, and they propose the Disjoint Author-Document Topic (DADT) model. DADT keeps separate topic spaces for authors and for documents, which merges the advantages of the two earlier models. This extended version can also learn from multi-authored texts. Stopwords are mostly kept because they carry stylistic signal. The evaluation covers five data sets, ranging from formal texts by three authors to informal texts by about 20,000 users. A probabilistic classifier built on DADT (DADT-P) is usually the strongest, beating LDA, AT and a Token SVM baseline. The results suggest that topic models capture writing style as well as subject matter, and that the resulting author profiles are useful for other user-modelling tasks.

Key points

  • Three topic models are compared (LDA, AT, DADT). DADT-P reaches state-of-the-art accuracy in most scenarios, including the Blog and IMDb1M corpora. The informal data sets involve anywhere from 62 to roughly 20,000 possible authors.
  • A preliminary study used AT and DADT trained on reviewers' public, often multi-authored publications. It identified anonymous conference reviewers with moderate accuracy, mainly in small tracks and workshops.
  • Topic-based author representations also improved polarity inference and rating prediction. On IMDb1M in the Given1 setting, the RMSE values were:
    • AT: 2.668
    • DADT: 2.678
    • Non-personalized baseline: 2.691
    • Ratings-only model: 2.734
  • In rating prediction, DADT did not surpass AT. The authors suggest this is because its highly discriminative user profiles capture less of what users have in common.
  • The number of topics had the largest effect on performance. Proposed future work includes:
    • adding word order or character n-grams;
    • inferring the number of topics automatically;
    • extending the approach to open-set attribution and verification via probability thresholds.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Monash UniversityMelbourne · Australia60 works · 54,866 citations

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksSocial Sciences · 1,772 worksGeneral Social Sciences · 227 works

💡 Concepts 1Key ideas and methods explained on our concept pages.Topic modeling · 47 works

Cite

Cite as Yanir Seroussi, Ingrid Zukerman and Fabian Bohnert. 2014. Authorship Attribution with Topic Models. Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1162/coli_a_00173

@article{seroussi2014authorship,
  title = {{Authorship Attribution with Topic Models}},
  author = {Yanir Seroussi, Ingrid Zukerman and Fabian Bohnert},
  journal = {Computational Linguistics},
  year = {2014},
  doi = {10.1162/coli\_a\_00173},
  url = {https://doi.org/10.1162/coli_a_00173}
}

References

62 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Argamon, Shlomo and Patrick Juola. 2011. Overview of the international authorship identification competition at PAN-2011. In CLEF 2011: Proceedings of the 2011 Conference on Multilingual and Multimodal Information Access Evaluation (Lab and Workshop Notebook Papers), Amsterdam. find
  2. Argamon, Shlomo, Moshe Koppel, James W. Pennebaker, and Jonathan Schler. 2009. Automatically profiling the author of an anonymous text. Communications of the ACM, 52(2):119–123. find
  3. Blei, David M. 2012. Probabilistic topic models. Communications of the ACM, 55(4):77–84. find
  4. Blei, David M. and Jon D. McAuliffe. 2007. Supervised topic models. In NIPS 2007: Proceedings of the 21st Annual Conference on Neural Information Processing Systems, pages 121–128, Vancouver. find
  5. Blei, David M., Andrew Y. Ng, and Michael I. Jordan. 2003. Latent Dirichlet allocation. Journal of Machine Learning Research, 3:993–1022.
  6. Breese, John S., David Heckerman, and Carl Kadie. 1998. Empirical analysis of predictive algorithms for collaborative filtering. In UAI 1998: Proceedings of the 14th Conference on Uncertainty in Artificial Intelligence, pages 43–52, Madison, WI. find
  7. Brennan, Michael and Rachel Greenstadt. 2009. Practical attacks against authorship recognition techniques. In IAAI 2009: Proceedings of the 21st Conference on Innovative Applications of Artificial Intelligence, pages 60–65, Pasadena, CA. find
  8. Chaski, Carole E. 2005. Who’s at the keyboard? Authorship attribution in digital evidence investigations. International Journal of Digital Evidence, 4(1). find
  9. Fan, Rong-En, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang, and Chih-Jen Lin. 2008. LIBLINEAR: A library for large linear classification. Journal of Machine Learning Research, 9:1871–1874. find
  10. Fog, Agner. 2008. Calculation methods for Wallenius’ noncentral hypergeometric distribution. Communications in Statistics, Simulation and Computation, 37(2):258–273. find
  11. Griffiths, Thomas L. and Mark Steyvers. 2004. Finding scientific topics. Proceedings of the National Academy of Sciences, 101(Suppl. 1):5228–5235. find
  12. Griffiths, Thomas L., Mark Steyvers, David M. Blei, and Joshua B. Tenenbaum. 2004. Integrating topics and syntax. In NIPS 2004: Proceedings of the 18th Annual Conference on Neural Information Processing Systems, pages 537–544, Vancouver. find
  13. Groves, Trish. 2010. Is open peer review the fairest system? Yes. BMJ, 341:c6424. find
  14. Herlocker, Jonathan L., Joseph A. Konstan, Al Borchers, and John Riedl. 1999. An algorithmic framework for performing collaborative filtering. In SIGIR 1999: Proceedings of the 22nd International ACM SIGIR Conference on Research and Development in Information Retrieval, pages 230–237, Berkeley, CA. find
  15. Juola, Patrick. 2004. Ad-hoc authorship attribution competition. In ALLC-ACH 2004: Proceedings of the 2004 Joint International Conference of the Association for Literary and Linguistic Computing and the Association for Computers and the Humanities, pages 175–176, Göteborg. find
  16. Juola, Patrick. 2006. Authorship attribution. Foundations and Trends in Information Retrieval, 1(3):233–334. find
  17. Kacmarcik, Gary and Michael Gamon. 2006. Obfuscating document stylometry to preserve author anonymity. In COLING-ACL 2006: Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics (Main Conference Poster Se… find
  18. Kern, Roman, Christin Seifert, Mario Zechner, and Michael Granitzer. 2011. Vote/veto meta-classifier for authorship identification. In CLEF 2011: Proceedings of the 2011 Conference on Multilingual and Multimodal Information Access Evaluation (Lab and Workshop Notebook Papers), Amsterdam. find
  19. Koppel, Moshe and Jonathan Schler. 2004. Authorship verification as a one-class classification problem. In ICML 2004: Proceedings of the 21st International Conference on Machine Learning, pages 62–68, Banff. find
  20. Koppel, Moshe, Jonathan Schler, and Shlomo Argamon. 2009. Computational methods in authorship attribution. Journal of the American Society for Information Science and Technology, 60(1):9–26. find
  21. Koppel, Moshe, Jonathan Schler, and Shlomo Argamon. 2011. Authorship attribution in the wild. Language Resources and Evaluation, 45(1):83–94. find
  22. Koren, Yehuda, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. IEEE Computer, 42(8):30–37. find
  23. Kourtis, Ioannis and Efstathios Stamatatos. 2011. Author identification using semi-supervised learning. In CLEF 2011: Proceedings of the 2011 Conference on Multilingual and Multimodal Information Access Evaluation (Lab and Workshop Notebook Papers), Amsterdam. find
  24. Lacoste-Julien, Simon, Fei Sha, and Michael I. Jordan. 2008. DiscLDA: Discriminative learning for dimensionality reduction and classification. In NIPS 2008: Proceedings of the 22nd Annual Conference on Neural Information Processing Systems, pages 897–904, Vancouver. find
  25. Liu, Bing and Lei Zhang. 2012. A survey of opinion mining and sentiment analysis. In Charu C. Aggarwal and ChengXiang Zhai, editors, Mining Text Data. Springer US, pages 415–463. find
  26. Luyckx, Kim and Walter Daelemans. 2011. The effect of author set size and data size in authorship attribution. Literary and Linguistic Computing, 26(1):35–55. find
  27. Mendenhall, Thomas C. 1887. The characteristic curves of composition. Science, 9(214S):237–246. find
  28. Mimno, David and Andrew McCallum. 2008. Topic models conditioned on arbitrary features with Dirichlet-multinomial regression. In UAI 2008: Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, pages 411–418, Helsinki. find
  29. Mosteller, Frederick and David L. Wallace. 1964. Inference and Disputed Authorship: The Federalist. Addison-Wesley. find
  30. Nanavati, Mihir, Nathan Taylor, William Aiello, and Andrew Warfield. 2011. Herbert West—deanonymizer. In HotSec’11: Proceedings of the 6th USENIX Workshop on Hot Topics in Security, San Francisco, CA. find
  31. Ng, Andrew Y. and Michael I. Jordan. 2001. On discriminative vs. generative classifiers: A comparison of logistic regression and naive Bayes. In NIPS 2001: Proceedings of the 15th Annual Conference on Neural Information Processing Systems, pages 841–848, Vancouver. find
  32. Pang, Bo and Lillian Lee. 2005. Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales. In ACL 2005: Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics, pages 115–124, Ann Arbor, MI.
  33. Pang, Bo and Lillian Lee. 2008. Opinion mining and sentiment analysis. Foundations and Trends in Information Retrieval, 2(1–2):1–135. find
  34. Pearl, Lisa and Mark Steyvers. 2012. Detecting authorship deception: A supervised machine learning approach using author writeprints. Literary and Linguistic Computing, 27(2):183–196. find
  35. Rajkumar, Arun, Saradha Ravi, Venkatasubramanian Suresh, M. Narasimha Murthy, and C. E. Veni Madhavan. 2009. Stopwords and stylometry: A latent Dirichlet allocation approach. In Proceedings of the NIPS 2009 Workshop on Applications for Topic Models: Text and Beyond (Poster Session), Whistler. find
  36. Ramage, Daniel, David Hall, Ramesh Nallapati, and Christopher D. Manning. 2009. Labeled LDA: A supervised topic model for credit attribution in multi-labeled corpora. In EMNLP 2009: Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 248–256, Singapore.
  37. Resnick, Paul and Hal R. Varian. 1997. Recommender systems. Communications of the ACM, 40(3):56–58. find
  38. Rifkin, Ryan and Aldebaro Klautau. 2004. In defense of one-vs-all classification. Journal of Machine Learning Research, 5(Jan):101–141. find
  39. Rosen-Zvi, Michal, Chaitanya Chemudugunta, Thomas Griffiths, Padhraic Smyth, and Mark Steyvers. 2010. Learning author-topic models from text corpora. ACM Transactions on Information Systems, 28(1):1–38. find
  40. Rosen-Zvi, Michal, Thomas Griffiths, Mark Steyvers, and Padhraic Smyth. 2004. The author-topic model for authors and documents. In UAI 2004: Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence, pages 487–494, Banff. find
  41. Salton, Gerard. 1971. The SMART Retrieval System—Experiments in Automatic Document Processing. Prentice-Hall, Inc., Upper Saddle River, NJ. find
  42. Salton, Gerard. 1981. A blueprint for automatic indexing. SIGIR Forum, 16(2):22–38. find
  43. Sanderson, Conrad and Simon Guenter. 2006. Short text authorship attribution via sequence kernels, Markov chains and author unmasking: An investigation. In EMNLP 2006: Proceedings of the 2006 Conference on Empirical Methods in Natural Language Processing, pages 482–491, Sydney. find
  44. Schler, Jonathan, Moshe Koppel, Shlomo Argamon, and James W. Pennebaker. 2006. Effects of age and gender on blogging. In Proceedings of AAAI Spring Symposium on Computational Approaches for Analyzing Weblogs, pages 199–205, Stanford, CA. find
  45. Seroussi, Yanir. 2012. Text Mining and Rating Prediction with Topical User Models. Ph.D. thesis, Faculty of Information Technology, Monash University, Clayton, Victoria, Australia. find
  46. Seroussi, Yanir, Fabian Bohnert, and Ingrid Zukerman. 2011. Personalized rating prediction for new users using latent factor models. In HT 2011: Proceedings of the 22nd International ACM Conference on Hypertext and Hypermedia, pages 47–56, Eindhoven. find
  47. Seroussi, Yanir, Fabian Bohnert, and Ingrid Zukerman. 2012. Authorship attribution with author-aware topic models. In ACL 2012: Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 264–269, Jeju Island. find
  48. Seroussi, Yanir, Russell Smyth, and Ingrid Zukerman. 2011. Ghosts from the High Court’s past: Evidence from computational linguistics for Dixon ghosting for McTiernan and Rich. University of New South Wales Law Journal, 34(3):984–1005. find
  49. Seroussi, Yanir, Ingrid Zukerman, and Fabian Bohnert. 2010. Collaborative inference of sentiments from texts. In UMAP 2010: Proceedings of the 18th International Conference on User Modeling, Adaptation and Personalization, pages 195–206, Waikoloa, HI. find
  50. Seroussi, Yanir, Ingrid Zukerman, and Fabian Bohnert. 2011. Authorship attribution with latent Dirichlet allocation. In CoNLL 2011: Proceedings of the 15th International Conference on Computational Natural Language Learning, pages 181–189, Portland, OR. find
  51. Stamatatos, Efstathios. 2009. A survey of modern authorship attribution methods. Journal of the American Society for Information Science and Technology, 60(3):538–556. find
  52. Steyvers, Mark and Tom Griffiths. 2007. Probabilistic topic models. In Thomas K. Landauer, Danielle S. McNamara, Simon Dennis, and Walter Kintsch, editors, Handbook of Latent Semantic Analysis. Lawrence Erlbaum Associates, pages 427–448. find
  53. Tanguy, Ludovic, Assaf Urieli, Basilio Calderone, Nabil Hathout, and Franck Sajous. 2011. A multitude of linguistically-rich features for authorship attribution. In CLEF 2011: Proceedings of the 2011 Conference on Multilingual and Multimodal Information Access Evaluation (Lab and Workshop Notebook P… find
  54. Teh, Yee Whye, Michael I. Jordan, Matthew J. Beal, and David M. Blei. 2006. find
  55. Hierarchical Dirichlet processes. Journal of the American Statistical Association, 101(476):1566–1581. find
  56. Wallach, Hanna M. 2006. Topic modeling: Beyond bag-of-words. In ICML 2006: Proceedings of the 23rd International Conference on Machine Learning, pages 977–984, Pittsburgh, PA. find
  57. Wallach, Hanna M., David Mimno, and Andrew McCallum. 2009. Rethinking LDA: Why priors matter. In NIPS 2009: Proceedings of the 23rd Annual Conference on Neural Information Processing Systems, pages 1,973–1,981, Vancouver. find
  58. Wang, Xuerui, Andrew McCallum, and Xing Wei. 2007. Topical N-grams: Phrase and topic discovery, with an application to information retrieval. In ICDM 2007: Proceedings of the 7th IEEE International Conference on Data Mining, pages 697–702, Omaha, NE. find
  59. Webb, Geoffrey I., Janice R. Boughton, and Zhihai Wang. 2005. Not so naive Bayes: Aggregating one-dependence estimators. Machine Learning, 58(1):5–24. find
  60. Wong, Sze-Meng Jojo, Mark Dras, and Mark Johnson. 2011. Topic modeling for native language identification. In ALTA 2011: Proceedings of the Australasian Language Technology Association Workshop, pages 115–124, Canberra. find
  61. Zhu, Jun, Amr Ahmed, and Eric P. Xing. 2009. MedLDA: Maximum margin supervised topic models for regression and classification. In ICML 2009: Proceedings of the 26th International Conference on Machine Learning, pages 1,257–1,264, Montreal. find
  62. Zhu, Jun and Eric P. Xing. 2010. Conditional topic random fields. In ICML 2010: Proceedings of the 27th International Conference on Machine Learning, pages 1,239–1,246, Haifa. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/seroussi-2014-authorship-attribution-topic-models/