Elamur Research⌕
☰
home / papers / 2013 / Arabic text summarization based on latent semantic analysis to enhance arabic documents clustering

Arabic text summarization based on latent semantic analysis to enhance arabic documents clustering

Paper · 2013 · Hanane Froud, Abdelmonaime Lachkar and Said Alaoui Ouatik

Summary

The paper asks whether Arabic documents can be clustered better if each document is first reduced to a short extractive summary. The motivation is that full texts carry noise and differ widely in length, and both hurt vector-space clustering. For every document the authors build a term-by-sentence matrix. Each cell is the term's count in the sentence multiplied by a log inverse sentence-frequency computed inside that document. They apply SVD and follow the Gong & Liu procedure: for each of the top right singular vectors, the sentence with the largest value is taken into the summary. Arabic punctuation is unreliable, so sentence boundaries come from lists of morphosyntactic markers and function words. The summaries are then represented in the Vector Space Model and clustered with five similarity/distance measures, once without stemming and once with stemming (root-based Khoja, light Larkey). The authors report that summarization-based representations clearly beat full-text ones. They present this as evidence that a summary can act as a noise filter before IR-oriented clustering. Specific scores and dataset details fall in a part of the text we did not have.

Key points

  • Sentence weighting: local weight is the raw term frequency in a sentence. Global weight is log(N / n), where N is the number of sentences in the document and n is the number of sentences containing the term.
  • Summarizer: LSA via SVD (A = UΣVᵀ), following Gong & Liu (SIGIR 2001). The k-th summary sentence is the one with the highest entry in the k-th row of Vᵀ, repeated until a preset summary length is reached.
  • Sentence segmentation relies on Arabic morphosyntactic markers and functional words (roughly: and, or, then, but, when, also, after, although). The authors note that these words can also serve other roles.
  • Five measures are compared, each with and without stemming:
    • Euclidean (straight-line) distance
    • Cosine of the angle between vectors
    • Jaccard's overlap coefficient
    • Pearson's correlation coefficient
    • Averaged Kullback-Leibler Divergence
  • Stemming conditions contrast a root-based approach (Khoja) with a stem-based light stemmer (Larkey).

Interesting ideas

  • I1. When punctuation cannot be trusted, split sentences at closed-class connective words, while accepting that these words are ambiguous.
  • I2. Choose one representative sentence per latent topic, ranked by singular value, so the summary covers distinct themes instead of repeating the strongest one.
  • I3. Inverse frequency computed over sentences within a single document estimates how distinctive a unit is locally, without any external corpus.
  • I4. Judge a compression step by whether a downstream task improves on the compressed input, not only by intrinsic summary quality.
  • I5. Run every condition as a full grid (several similarity measures × stemming on/off) so preprocessing effects are not tied to one metric.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

  • I1 (tokens): a fallback segmenter could break unpunctuated or run-on English (chat, transcripts) at coordinators and subordinators, with the parser later confirming or rejecting each break.
  • I2 (back): the text → graph → text compressor could choose events or frames for output by ensuring each major cluster of related entities or actions is covered once, not by raw frequency.
  • I3 (world): entity or predicate salience in a text's world could be scored by how few sentences mention it versus how often, which helps decide which entities matter for answers.
  • I4 (learn): the compressor could be evaluated by checking that QA readers still answer correctly from the regenerated text, with the full text as a control.
  • I5 (tokens): when choosing lemma granularity (aggressive vs light normalization), compare it across several downstream metrics instead of one, so a gain is not an artefact of a single measure.

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

🎓Sidi Mohamed Ben Abdellah UniversityFes · Morocco8 works · 156 citations

 Fields, topics and concepts

📚 Shelves 4The 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 works

🧭 Topics 2Research subjects and languages the work is about.Semitic languages · 2,908 worksSummarization · 232 works

Cite

Cite as Hanane Froud, Abdelmonaime Lachkar and Said Alaoui Ouatik. 2013. Arabic text summarization based on latent semantic analysis to enhance arabic documents clustering. arXiv:1302.1612 · International Journal of Data Mining & Knowledge Management Process (IJDKP)- 2013.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.1302.1612 · arXiv 1302.1612

@misc{froud2013arabic,
  title = {{Arabic text summarization based on latent semantic analysis to enhance arabic documents clustering}},
  author = {Hanane Froud, Abdelmonaime Lachkar and Said Alaoui Ouatik},
  year = {2013},
  doi = {10.48550/arxiv.1302.1612},
  eprint = {1302.1612},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.1302.1612}
}

References

25 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. P. Bellot and M. El-Bèze, “Clustering by means of Unsupervised Decision Trees or Hierarchical and K-means-like Algorithm”, in Proc. of RIAO 2000, pp. 344-363. find
  2. Sameh H. Ghwanmeh, “Applying Clustering of Hierarchical K-means-like Algorithm on Arabic Language” International Journal of Information Technology IJIT Volume 3 Number 3 2007, p 168-172. ↗
  3. A.Huang, “Similarity Measures for Text Document Clustering”, NZCSRSC 2008, April 2008, Christchurch, New Zealand. find
  4. Khoja, S. and Garside, R. “Stemming Arabic Text”. Computing Department, Lancaster University, Lancaster, 1999. http://www.comp.lancs.ac.uk/computing/users/khoja/stemmer.ps ↗
  5. Larkey, Leah S., Ballesteros, Lisa, and Connell, Margaret. (2002) “Improving Stemming for Arabic Information Retrieval: Light Stemming and Co-occurrence Analysis”. In Proceedings of the 25th Annual International Conference on Research and Development in Information Retrieval (SIGIR 2002), Tampere, Finland, August 11-15, 2002, pp. 275-282. find
  6. R. B. Yates and B. R. Neto.”Modern Information Retrieval”. ADDISON-WESLEY, New York, 1999. find
  7. B. Larsen and C. Aone.” Fast and Effective Text Mining using Linear-time Document Clustering”. In Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1999. find
  8. N. Z. Tishby, F. Pereira, and W. Bialek. “The Information Bottleneck Method”. In Proceedings of the 37th Allerton Conference on Communication, Control and Computing, 1999. find
  9. Y. Zhao and G. Karypis.”Evaluation of Hierarchical Clustering Algorithms for Document Datasets”. In Proceedings of the International Conference on Information and Knowledge Management, 2002. find
  10. Y. Zhao and G. Karypis.”Empirical and Theoretical Comparisons of Selected Criterion Functions for Document Clustering”. Machine Learning, 55(3), 2004. find
  11. Xuanhui Wang, Dou Shen,Hua-Jun Zeng, Zheng Chen, and Wei-Ying Ma. “Web page clustering enhanced by summarization”. Proceedings of the 2004 ACM CIKM International Conference on Information and Knowledge Management, Washington, DC, USA, November 8-13, 2004. find
  12. H.Froud, A.Lachkar, S. Ouatik, and R.Benslimane (2010). Stemming and Similarity Measures for Arabic Documents Clustering. 5th International Symposium on I/V Communications and Mobile Networks ISIVC, IEEE Xplore. find
  13. S. Deerwester, S. Dumais, G. Furnas, T. Landauer, and R. Harshman, “Indexing by latent semantic analysis,” Journal of the American Society for Information Science, vol. 41, pp.391-407, 1990. find
  14. H. Froud, A. Lachkar, S. Alaoui Ouatik, “Stemming Versus Light Stemming for Measuring the Simitilarity between Arabic Words with Latent Semantic Analysis Model”, 2 nd International Colloquium in Information Science and Technology (CIST) 22-24 October, 2012. IEEE Xplore. find
  15. Y.H. Gong and X. Liu, “Generic text summarization using relevance measure and latent semantic analysis,” Proc. The 24th annual international ACM SIGIR, pp. 19 - 25, 2001. find
  16. Steinberger J., Jezek K. Using Latent Semantic Analysis in text summarization and summary evaluation. In: Proceedings of ISIM `04 2004: 93-100. find
  17. W. Press and et al., Numerical Recipes in C: The Art of Scientific Computing. Cambridge, England: Cambridge University Press, 2 ed., 1992. find
  18. M. W. Berry, S. T. Dumais, G. W O’Brien: Using Linear Algebra for Intelligent Information Retrieval. SIAM Review 1995. find
  19. Ouersighni R. 2001. A major offshoot of the DIINAR-MBC project: AraParse, a morphosyntactic analyzer for unvowelled Arabic texts, ACL/EACL 2001 Workshop on Arabic Language Processing, Toulouse July 2001, pp. 9-16. find
  20. Abu-Hamdiyyah, Mohammad.2000. “The Qur'An: An Introduction” find
  21. Al-Fedaghi S. and F. Al-Anzi. “A new algorithm to generate Arabic root-pattern forms”. In proceedings of the 11th national Computer Conference and Exhibition. PP 391-400. March 1989. find
  22. Al-Shalabi R. and M. Evens. “A computational morphology system for Arabic”. In Workshop on Computational Approaches to Semitic Languages, COLING-ACL98. August 1998.
  23. Aljlayl M. and O. Frieder. “On Arabic search: improving the retrieval effectiveness via a light temming approach”. In ACM CIKM 2002 International Conference on Information and Knowledge Management, McLean, VA, USA. PP 340-347. 2002. find
  24. Larkey L., and M. E. Connell. “Arabic information retrieval at UMass in TREC-10”. Proceedings of TREC 2001, Gaithersburg: NIST. 2001. find
  25. Chen A. and F. Gey. “Building an Arabic Stemmer for Information Retrieval”. In Proceedings of the 11th Text Retrieval Conference (TREC 2002), National Institute of Standards and Technology. 2002. +،ة S ا ا !3 T ا تاودأ ،! ارU ا !#% ن ر و / ' . ه V .أ . د Authors Miss. Hanane Froud Phd Student in Laboratory of Information Science and Systems, ECOLE NATIONALE DES SCIENCES APPLIQUÉES, University Sidi Mohamed Ben Abdellah (USMBA), Fez, Morocco. She has also presented different papers at different National and International conferences. Pr. Abdelmonaime LACHKAR received his PhD degree from the USMBA, Morocco in 2004 in computer science; He is working as a Professor and Head of Computer Science and Engineering (E.N.S.A), in University Sidi Mohamed Ben Abdellah (USMBA), Fez, Morocco. His current research interests include Arabic Text Mining Applications: Arabic Web Document Clustering and Categorization. Arabic Information and Retrieval Systems, Arabic Text Summarization, etc …, Image Indexing and Retrieval, 3D Shape Indexing and Retrieval in large 3D Objects Databases, Color Image Segmentation, Unsupervised clustering, Cluster Validity Index, on-line and off-line Arabic and Latin handwritten recognition, and Medical Image Applications. Pr. Said Alaoui Ouatik is working as a Professor in Department of Computer Science, Faculty of Science Dhar EL Mahraz (FSDM), Fez, Morocco. His research interests include high-dimensional indexing and content-based retrieval, Arabic Document Categorization. 2D/3D Shapes Indexing and Retrieval in large 3D Objects Database. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/froud-2013-arabic-text-summarization-based-latent/