Document similarity search is to find documents similar to a given query document and return a ranked list of similar documents to users, which is widely used in many text and web systems, such as digital library, search engine, etc. Traditional retrieval models, including the Okapi's BM25 model and the Smart's vector space model with length normalization, could handle this problem to some extent by taking the query document as a long query. In practice, the Cosine measure is considered as the best model for document similarity search because of its good ability to measure similarity between two documents. In this paper, the quantitative performances of the above models are compared using experiments. Because the Cosine measure is not able to reflect the structural similarity between documents, a new retrieval model based on TextTiling is proposed in the paper. The proposed model takes into account the subtopic structures of documents. It first splits the documents into text segments with TextTiling and calculates the similarities for different pairs of text segments in the documents. Lastly the overall similarity between the documents is returned by combining the similarities of different pairs of text segments with optimal matching method. Experiments are performed and results show: 1) the popular retrieval models (the Okapi's BM25 model and the Smart's vector space model with length normalization) do not perform well for document similarity search; 2) the proposed model based on TextTiling is effective and outperforms other models, including the Cosine measure; 3) the methods for the three components in the proposed model are validated to be appropriately employed.
This is a preview of subscription content, log in to check access.
Buy single article
Instant access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
Robertson S, Walker S. Some simple effective approximations to the 2-poisson model for probabilistic weighted retrieval. In Proc. the 17th International ACM/SIGIR Conference on Research and Development in Information Retrieval (SIGIR'2003), Dublin, Ireland, 1994, pp.232–241.
Salton G. The SMART Retrieval System: Experiments in Automatic Document Processing. Prentice-Hall, 1971.
Baeza-Yates R, Ribeiro-Neto B. Modern Information Retrieval. ACM Press and Addison Wesley, New York, 1999.
Jones W P, Furnas G W. Pictures of relevance: A geometric analysis of similarity measure. Journal of the American Society for Information Science, 1987, 38(6): 420–442.
Zobel J, Moffat A. Exploring the similarity space. ACM SIGIR FORUM, 1998, 32(1): 18–34.
Aslam J A, Frost M. An information-theoretic measure for document similarity. In Proc. the 26th International ACM/SIGIR Conference on Research and Development in Information Retrieval (SIGIR'2003), Toronto, Canada, 2003, pp.449–450.
Hammouda K M, Kamel M S. Document similarity using a phrase indexing graph model. Journal of Knowledge and Information Systems, 2004, 6(6): 710–717.
Hearst M A. Multi-paragraph segmentation of expository text. In Proc. the 32nd Meeting of the Association for Computational Linguistics (ACL'1994), Los Cruces, NM, 1994, pp.9–16.
van Rijsbergen C J. Information Retrieval. Butterworths, London, 1979.
Deerwester S C, Dumais S T, Landauer T K et al. Indexing by latent semantic analysis. Journal of the American Society of Information Science, 1990, 41(6): 211–240.
Buckley C, Sinhal A, Mitra M, Saltton G. New retrieval approaches using SMART: TREC 4. In Proc. the TREC 4 Conference, 1996, pp.25–48.
Singhal A, Buckley C, Mitra M. Pivoted document length normalization. In Proc. the 19th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, Zurich, Switzerland, 1996, pp.21–29.
Hearst M A, Plaunt C. Subtopic structuring for full-length document access. In Proc. the 16th Annual International ACM/SIGIR Conference, Pittsburgh, PA, 1993, pp.59–68.
Karlgren J. Stylistic variation in an information retrieval experiment. In Proc. the NeMLaP-2 Conference, Ankara, Turkey, 1996.
Mittal V, Kantrowitz M, Goldstein J, Carbonell J. Selecting text spans for document summaries: Heuristics and metrics. In Proc. the 16th Annual Conference on Artificial Intelligence (AAAI99), Orlando, FL, 1999, pp.467–473.
Lovasz L, and Plummer M D. Matching Theory, Amsterdam: North Holland, 1986.
Peng Y X, Ngo C W, Dong Q J et al. Video clip retrieval by maximal matching and optimal matching in graph theory. In Proc. 2003 IEEE Int. Conf. Multimedia & Expo (ICME'2003), Baltimore, Maryland, 2003, pp.317–320.
Choi F. JTextTile: A free platform independent text segmentation algorithm. http://www.cs.man.ac.uk/~choif
Allan J, Carbonell J, Doddington G, Yamron J P, Yang Y. Topic detection and tracking pilot study: Final report. In Proc. DARPA Broadcast News Transcription and Understanding Workshop, 1998, pp.194–218.
Porter M F. An algorithm for suffix stripping. Program, July 1980, 14(3): 130–137.
Aggarwal C C, Yu P S. On effective conceptual indexing and similarity search in text data. In Proc. the 2001 IEEE Int. Conf. Data Mining (ICDM2001), San Jose, California, 2001, pp.3–10.
Schrijver A. Combinatorial Optimization: Polyhedra and Efficiency. Volume A, Berlin: Springer, 2003.
In this paper, we use the abbreviation “Okapi” and “Smart” to represent the Okapi's BM25 model and the Smart's vector space model with length normalization respectively.
About this article
Cite this article
Wan, X., Peng, Y. A New Retrieval Model Based on TextTiling for Document Similarity Search. J Comput Sci Technol 20, 552–558 (2005). https://doi.org/10.1007/s11390-005-0552-9
- document similarity search
- retrieval model
- similarity measure
- optimal matching