An Approach to Clustering Abstracts
Free access to full-text scientific papers in major digital libraries and other web repositories is limited to only their abstracts consisting of no more than several dozens of words. Current keyword-based techniques allow for clustering such type of short texts only when the data set is multi-category, e.g., some documents are devoted to sport, others to medicine, others to politics, etc. However, they fail on narrow domain-oriented libraries, e.g., those containing all documents only on physics, or all on geology, or all on computational linguistics, etc. Nevertheless, just such data sets are the most frequent and most interesting ones. We propose simple procedure to cluster abstracts, which consists in grouping keywords and using more adequate document similarity measure. We use Stein’s MajorClust method for clustering both keywords and documents. We illustrate our approach on the texts from the Proceedings of a narrow-topic conference. Limitations of our approach are also discussed. Our preliminary experiments show that abstracts cannot be clustered with the same quality as full texts, though the achieved quality is adequate for many applications; accordingly, we suggest Makagonov’s proposal that digital libraries should provide document images of full texts of the papers (and not only abstracts) for open access via Internet, in order to help in search, classification, clustering, selection, and proper referencing of the papers.
KeywordsDigital Library Near Neighbor Document Collection Document Image Short Document
Unable to display preview. Download preview PDF.
- 2.Alexandrov, M., Gelbukh, A., Rosso, P.: Clustering Very Short Documents based on Grouping Keywords. In: Abstracts of the 30-th Latin-American Conf. on Informatics, Univ. Edition, Peru, p. 133 (2004)Google Scholar
- 3.Baeza-Yates, R., Ribero-Neto, B.: Modern Information Retrieval. Addison-Wesley, Reading (1999)Google Scholar
- 6.Hardy, A., Andre, P.: An investigation of nine procedures for detecting the structure in a data set. In: Advances in data science and classification. Studies in Classification, Data Analysis and Knowledge Organization, pp. 29–36. Springer, Heidelberg (1998)Google Scholar
- 8.Hynek, J., Jezek, K., Rohlikm, O.: Short Document Categorization – Itemsets Method. In: PKDD-2000. LNCS, p. 6. Springer, Heidelberg (2000)Google Scholar
- 10.Makagonov, P., Alexandrov, M., Sboychakov, K.: Keyword-based technology for clustering short documents. In: Selected Papers. Computing Research, CIC-IPN, Mexico, pp. 105–114 (2000)Google Scholar
- 11.Makagonov, P., Alexandrov, M., Sboychakov, K.: A toolkit for development of the domainoriented dictionaries for structuring document flows. In: Data Analysis, Classification, and Related Methods, Studies in classification, data analysis, and knowledge organization, pp. 83–88. Springer, Heidelberg (2000)Google Scholar
- 14.Porter, M.: An algorithm for suffix stripping. Program 14, 130–137 (1980)Google Scholar
- 16.Solomon, G.: Data dependent methods of cluster analysis. In: Classification and Clustering, pp. 129–147. Academic Press, London (1977) (Russian version)Google Scholar
- 18.Stein, B., Eissen, S.M.: Document Categorization with MajorClust. In: Proc. 12th Workshop on Information Technology and Systems, Tech. Univ. of Barcelona, Spain, p. 6 (2002)Google Scholar
- 20.Stein, B., Eissen, S.M., Wissbrock, F.: On Cluster Validity and the Information Need of Users. In: Proc. 3-rd IASTED Intern. Conf. on Artificial Intelligence and Applications (AIA 2003), pp. 216–221. Acta Press (2003)Google Scholar
- 21.Strzalkowski, T. (ed.): Natural Language and Information Retrieval. Kluwer Academic Publishers, Dordrecht (1999)Google Scholar