---
title: "Concept Based vs. Pseudo Relevance Feedback Performance"
description: "The paper asks which of two automatic query-expansion strategies better helps an information retrieval system for Arabic text. 24 references."
authors: ["Mohammed El Amine Abderrahim"]
year: 2013
venue: "arXiv:1403.4362 · International Journal of Computational Linguistics Research, ISSN: 0976-416X, Volume 4, Issue 4, December, 2013, Pages 149-158"
source: "https://arxiv.org/abs/1403.4362"
doi: "10.48550/arxiv.1403.4362"
arxiv: "1403.4362"
licence: "arXiv — non-exclusive license to distribute"
licence_url: "http://arxiv.org/licenses/nonexclusive-distrib/1.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/abderrahim-2014-concept-based-vs-pseudo-relevance/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/abderrahim-2014-concept-based-vs-pseudo-relevance/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2013](https://md.research.elamur.ai/free/en/papers/2013/page/2/) / Concept Based vs. Pseudo Relevance Feedback Performance Evaluation for Information Retrieval System

# Concept Based vs. Pseudo Relevance Feedback Performance Evaluation for Information Retrieval System

**Authors:** Mohammed El Amine Abderrahim · **Year:** 2013 · **Venue:** arXiv:1403.4362 · International Journal of Computational Linguistics Research, ISSN: 0976-416X, Volume 4, Issue 4, December, 2013, Pages 149-158 · **Licence:** [arXiv — non-exclusive license to distribute](http://arxiv.org/licenses/nonexclusive-distrib/1.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/1403.4362) · [PDF](https://arxiv.org/pdf/1403.4362) · [DOI 10.48550/arxiv.1403.4362](https://doi.org/10.48550/arxiv.1403.4362) · [arXiv 1403.4362](https://arxiv.org/abs/1403.4362) · [OpenAlex](https://openalex.org/W1486866821)

**Citations:** 3 citations (OpenAlex, 2026-10-14)

## Summary
The paper asks which of two automatic query-expansion strategies better helps an information retrieval system for Arabic text. The first is concept-based (CB) reformulation, which adds terms from an external lexical resource. The second is pseudo relevance feedback (PRF), which adds terms drawn from the top documents the system itself returned. For CB, query terms are matched to concepts in Arabic WordNet (AWN), and every synonym found is appended blindly; synonymy is the only relation used. For PRF, the top 15 retrieved documents are treated as relevant. A local term–term association matrix (local clustering) is built from them, and the 7 most strongly correlated terms are added. Both methods are compared with plain retrieval on 50 queries over an Arabic corpus of 22,824 documents indexed with Lucene. PRF comes out ahead: more queries actually improved, and its recall/precision behaviour is better than CB's. This matters for Arabic IR, where lexical resources like AWN are still incomplete. A cheap, collection-internal method may therefore be the more practical choice.

## Key points
- **Three retrieval runs on the same 50 queries:**
  - simple retrieval (SR) with no reformulation;
  - concept-based retrieval (CBR), which adds all AWN synonyms of each query term;
  - PRF retrieval (PRFR).
- **PRF settings:** D=15 feedback documents and T=7 expansion terms, values taken from the author's earlier analysis of Arabic text. Expansion terms are chosen from a local association matrix.
- **Corpus:**
  - 22,824 text files, about 180 MB;
  - about 17 million words and 193,736 distinct terms;
  - ten domains, including health, sports, politics, religion, law and tales.
- **Lucene index:** roughly 37 MB.
- **AWN resource:** 11,269 synsets, of which 7,961 are nominal, 2,538 verbal, 661 adjectival and 110 adverbial. It is built on the Princeton WordNet/EuroWordNet model and stored in MySQL.
- **Main finding:** PRF beat CB both in the number of queries it improved and in recall/precision.

## Interesting ideas
- Blind synonym expansion pulls in off-topic senses. For example, the paper's "price" query picks up unit/thermal-unit/calorie senses, so expansion from a lexicon needs sense filtering, not just a lookup.
- Feedback from the document's own content often beats a general-purpose external resource, because co-occurring terms there reflect the actual topic (barrel, OPEC, production for oil prices).
- The paper treats feedback-set size and expansion size (D and T) as tunable, and fixes them from a prior language-specific study instead of guessing.
- The evaluation counts how many individual queries actually improved, alongside aggregate recall/precision. This exposes methods that help on average but hurt many cases.
- Comparing SR, CBR and PRFR on identical queries and an identical index is a clean ablation design. Only the reformulation step changes between runs.

## For Mova

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

- **I1** (world): when mapping words to WordNet-based ontology nodes, pick the sense that fits the frame and context before adding related concepts, rather than taking every synset.
- **I2** (world): build a text-local association layer, listing which entities and terms co-occur in this story, to rank candidate referents and concepts above generic lexicon links.
- **I3** (learn): when silver data is selected by agreement among witness parsers, tune and record how many witnesses and candidates are used per decision as explicit, studied parameters.
- **I4** (learn): report per-item win/loss counts against a baseline, not just mean scores, when comparing parser or repair-rule variants.
- **I5** (answer): evaluate reader variants (with and without knowledge-graph expansion) on the same fixed question sets, so differences are attributable to one component.

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

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,086 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,889 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 523 works

🧭 Topics (3): [📏 Evaluation & benchmarks](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · 685 works · [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 1,484 works · [🔍 Information retrieval & search](https://md.research.elamur.ai/free/en/topics/information-retrieval/) · 145 works

## Cite

Cite as: Mohammed El Amine Abderrahim. 2013. Concept Based vs. Pseudo Relevance Feedback Performance Evaluation for Information Retrieval System. arXiv:1403.4362 · International Journal of Computational Linguistics Research, ISSN: 0976-416X, Volume 4, Issue 4, December, 2013, Pages 149-158.

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

```bibtex
@misc{abderrahim2013concept,
  title = {{Concept Based vs. Pseudo Relevance Feedback Performance Evaluation for Information Retrieval System}},
  author = {Mohammed {El Amine Abderrahim}},
  year = {2013},
  doi = {10.48550/arxiv.1403.4362},
  eprint = {1403.4362},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.1403.4362}
}
```

- [← previous An Empirical Study of Differences between Conversion Schemes and Annotation Guidelines](https://md.research.elamur.ai/free/en/papers/sogaard-2013-empirical-study-differences-between-conversion/)
- [next → Fluid Construction Grammar for Historical and Evolutionary Linguistics](https://md.research.elamur.ai/free/en/papers/wellens-2013-fluid-construction-grammar-historical-evolutionary/)
- [↑ back to papers of 2013](https://md.research.elamur.ai/free/en/papers/2013/page/2/)

## References

24 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Abderrahim, M. E. A. and A, M.A. (2010). Using Arabic Wordnet for Query Expansion in Information Retrieval System. In IEEE The Third International Conference on Web and Information Technologies 16-19 June Marrakech Morocco. [find](https://scholar.google.com/scholar?q=Abderrahim%2C+M%2E+E%2E+A%2E+and+A%2C+M%2EA%2E+%282010%29%2E+Using+Arabic+Wordnet+for+Query+Expansion+in+Information+Retrieval+System%2E+In+IEEE+The+Third+International+Conference+on+Web+and+Information+Technologies+16%2D19+)
2. Abderrahim, M.E.A. and A, M.A. (2012). Réinjection automatique de la pertinence pour la recherche d’informations dans les textes arabes. In IEEE 4th International Conference on Arabic Language Processing May 23 Rabat Morocco, CITALA, pages 77–81. Number of queries CBR PRFR 7 (14%) 9 (18%) 29 (58%) 14 (28%) 14 (28%) 27 (54%) [find](https://scholar.google.com/scholar?q=Abderrahim%2C+M%2EE%2EA%2E+and+A%2C+M%2EA%2E+%282012%29%2E+R%C3%A9injection+automatique+de+la+pertinence+pour+la+recherche+d%E2%80%99informations+dans+les+textes+arabes%2E+In+IEEE+4th+International+Conference+on+Arabic+Language+Process)
3. Abderrahim, M.E.A., Benameur, S., and Abderrahim, M.A. (2013). The number of terms and documents for pseudo- relevant feedback for ad- hoc information retrieval. International Journal of Computer Science Issues, IJCSI, 10(1):661–667. [find](https://scholar.google.com/scholar?q=Abderrahim%2C+M%2EE%2EA%2E%2C+Benameur%2C+S%2E%2C+and+Abderrahim%2C+M%2EA%2E+%282013%29%2E+The+number+of+terms+and+documents+for+pseudo%2D+relevant+feedback+for+ad%2D+hoc+information+retrieval%2E+International+Journal+of+Computer+Scie)
4. L. Abouenour· K. Bouzoubaa· P. Rosso, (2013). On the evaluation and improvement of Arabic WordNet coverage and usability. Lang Resources & Evaluation (2013) 47:891-917, DOI 10.1007/s10579-013-9237-0. [↗](https://doi.org/10.1007/s10579-013-9237-0)
5. Al khalifa, M. (2006). Arabic Wordnet and Arabic NLP. In Journées d’Etudes sur le Traitement Automatique de la Langue Arabe JETALA 5-7 June Rabat. [find](https://scholar.google.com/scholar?q=Al+khalifa%2C+M%2E+%282006%29%2E+Arabic+Wordnet+and+Arabic+NLP%2E+In+Journ%C3%A9es+d%E2%80%99Etudes+sur+le+Traitement+Automatique+de+la+Langue+Arabe+JETALA+5%2D7+June+Rabat%2E)
6. Al Khalifa, M., & Rodrı´guez, H. (2009). Automatically extending NE coverage of Arabic WordNet using Wikipedia. InProceedings of the 3rd international conference on Arabic language processing CITALA’09, May, Rabat, Morocco. [find](https://scholar.google.com/scholar?q=Al+Khalifa%2C+M%2E%2C+%26+Rodr%C4%B1%C2%B4guez%2C+H%2E+%282009%29%2E+Automatically+extending+NE+coverage+of+Arabic+WordNet+using+Wikipedia%2E+InProceedings+of+the+3rd+international+conference+on+Arabic+language+processing+CITALA%E2%80%990)
7. Baeza-Yates, R. and Berthier, R.-N. (1999). Modern Information Retrieval. Addison Wesley New York City NY ACM Press. [find](https://scholar.google.com/scholar?q=Baeza%2DYates%2C+R%2E+and+Berthier%2C+R%2E%2DN%2E+%281999%29%2E+Modern+Information+Retrieval%2E+Addison+Wesley+New+York+City+NY+ACM+Press%2E)
8. Baziz, M. (2005). Indexation conceptuelle guidée par ontologie pour la recherche d’information. PhD thesis, Université de Toulouse, Université Toulouse IIIPaul Sabatier. [find](https://scholar.google.com/scholar?q=Baziz%2C+M%2E+%282005%29%2E+Indexation+conceptuelle+guid%C3%A9e+par+ontologie+pour+la+recherche+d%E2%80%99information%2E+PhD+thesis%2C+Universit%C3%A9+de+Toulouse%2C+Universit%C3%A9+Toulouse+IIIPaul+Sabatier%2E)
9. Black, W., Elkateb, S., Rodriguez, H., Alkhalifa, M., Vossen, P., Pease, A., and Fellbaum, C. (2006). Introducing the Arabic WordNet Project. In Proceedings of the Third International WordNet Conference, pages 295–300. [find](https://scholar.google.com/scholar?q=Black%2C+W%2E%2C+Elkateb%2C+S%2E%2C+Rodriguez%2C+H%2E%2C+Alkhalifa%2C+M%2E%2C+Vossen%2C+P%2E%2C+Pease%2C+A%2E%2C+and+Fellbaum%2C+C%2E+%282006%29%2E+Introducing+the+Arabic+WordNet+Project%2E+In+Proceedings+of+the+Third+International+WordNet+Conferen)
10. Violetta Cavalli-Sforza, H. Saddiki, K. Bouzoubaa, L. Abouenour, M. Maamouri, E. Goshey Bootstrapping a WordNet for an Arabic Dialect from Other WordNets and Dictionary Resources, 10 th ACS/IEEE International Conference on Computer Systems and Applications (AICCSA 2013), May 27-30, 2013, Fes/Ifrane, Morocco. [find](https://scholar.google.com/scholar?q=Violetta+Cavalli%2DSforza%2C+H%2E+Saddiki%2C+K%2E+Bouzoubaa%2C+L%2E+Abouenour%2C+M%2E+Maamouri%2C+E%2E+Goshey+Bootstrapping+a+WordNet+for+an+Arabic+Dialect+from+Other+WordNets+and+Dictionary+Resources%2C+10+th+ACS%2FIEEE+Inter)
11. Carpineto, C. and Romano, G. (2012). A survey of Automatic Query Expansion in Information Retrieval. ACM Computing Surveys (CSUR), 44(1):1. [↗](https://doi.org/10.1145/2071389.2071390)
12. Efthimiadis, E. N. (1996). Query Expansion. Annual review of information science and technology, 31:121–187. [find](https://scholar.google.com/scholar?q=Efthimiadis%2C+E%2E+N%2E+%281996%29%2E+Query+Expansion%2E+Annual+review+of+information+science+and+technology%2C+31%3A121%E2%80%93187%2E)
13. Elkateb, S., Black, W., Vossen, P., Farwell, D., Rodríguez, H., Pease, A., and Alkhalifa, M. (2006). Arabic WordNet and the Challenges of Arabic. In Proceedings of Arabic NLP/MT Conference, London, UK. Citeseer. [find](https://scholar.google.com/scholar?q=Elkateb%2C+S%2E%2C+Black%2C+W%2E%2C+Vossen%2C+P%2E%2C+Farwell%2C+D%2E%2C+Rodr%C3%ADguez%2C+H%2E%2C+Pease%2C+A%2E%2C+and+Alkhalifa%2C+M%2E+%282006%29%2E+Arabic+WordNet+and+the+Challenges+of+Arabic%2E+In+Proceedings+of+Arabic+NLP%2FMT+Conference%2C+London%2C+UK)
14. Fellbaum, C. (ed.) (1998) WordNet: An Electronic Lexical Database. Cambridge, MA: MIT Press. [find](https://scholar.google.com/scholar?q=Fellbaum%2C+C%2E+%28ed%2E%29+%281998%29+WordNet%3A+An+Electronic+Lexical+Database%2E+Cambridge%2C+MA%3A+MIT+Press%2E)
15. Harb, H. M., Fouad, K. M., and Nagdy, N. M. (2011). Semantic Retrieval Approach for Web Documents. IJACSA) International Journal of Advanced Computer Science and Applications, 2(9):11–75. [↗](https://doi.org/10.14569/ijacsa.2011.020912)
16. Kanaan, G., AlShalabi, R., AbuAlrub, M., and Rawashdeh, M. (2005). Relevance Feedback: Experimenting with a Simple Arabic Information Retrieval System with Evaluation. International Journal of Applied Science and Computations Vol 12 No 2 USA. [find](https://scholar.google.com/scholar?q=Kanaan%2C+G%2E%2C+AlShalabi%2C+R%2E%2C+AbuAlrub%2C+M%2E%2C+and+Rawashdeh%2C+M%2E+%282005%29%2E+Relevance+Feedback%3A+Experimenting+with+a+Simple+Arabic+Information+Retrieval+System+with+Evaluation%2E+International+Journal+of+Applied)
17. Nathalie, H., Gilles, H., Josiane, M., and Bachelin, R. (2008). Ri et ontologies etat de lart. Technical report, INP Toulouse, Université Paul Sabatier Toulouse III. [find](https://scholar.google.com/scholar?q=Nathalie%2C+H%2E%2C+Gilles%2C+H%2E%2C+Josiane%2C+M%2E%2C+and+Bachelin%2C+R%2E+%282008%29%2E+Ri+et+ontologies+etat+de+lart%2E+Technical+report%2C+INP+Toulouse%2C+Universit%C3%A9+Paul+Sabatier+Toulouse+III%2E)
18. Rodríguez, R., Farwell, D., Farreres, J., Bertran, M., Alkhalifa, M., Martí, M.A., Black, W., Elkateb, S., Kirk, J., Pease, A., Vossen, P., and Fellbaum, C., (2008). Arabic WordNet: Current State and Future Extensions. Proceedings of The Fourth Global WordNet Conference, Szeged, Hungary. January 22-25, 2008. [find](https://scholar.google.com/scholar?q=Rodr%C3%ADguez%2C+R%2E%2C+Farwell%2C+D%2E%2C+Farreres%2C+J%2E%2C+Bertran%2C+M%2E%2C+Alkhalifa%2C+M%2E%2C+Mart%C3%AD%2C+M%2EA%2E%2C+Black%2C+W%2E%2C+Elkateb%2C+S%2E%2C+Kirk%2C+J%2E%2C+Pease%2C+A%2E%2C+Vossen%2C+P%2E%2C+and+Fellbaum%2C+C%2E%2C+%282008%29%2E+Arabic+WordNet%3A+Current+State+and+)
19. Rodríguez, H. Farwell, D. Farreres, J. Bertran, M. Alkhalifa, M. Martí M.A (2008) Arabic WordNet: Semi-automatic Extensions using Bayesian Inference. In Proceedings of the the 6th Conference on Language Resources and Evaluation LREC2008. Marrakech (Morocco), May 2008. [find](https://scholar.google.com/scholar?q=Rodr%C3%ADguez%2C+H%2E+Farwell%2C+D%2E+Farreres%2C+J%2E+Bertran%2C+M%2E+Alkhalifa%2C+M%2E+Mart%C3%AD+M%2EA+%282008%29+Arabic+WordNet%3A+Semi%2Dautomatic+Extensions+using+Bayesian+Inference%2E+In+Proceedings+of+the+the+6th+Conference+on+Langua)
20. Salton, G. and Buckley, C. (1990). Improving Retrieval Performance by Relevance Feedback. Journal of the American Society for Information Science, 41(4):288–97. [↗](https://doi.org/10.1002/(sici)1097-4571(199006)41:4<288::aid-asi8>3.0.co;2-h)
21. Vossen P. (2004) EuroWordNet: a multilingual database of autonomous and language specific wordnets connected via an Inter-Lingual-Index. International Journal of Lexicography, Vol.17 No. 2, OUP, 161-173. [find](https://scholar.google.com/scholar?q=Vossen+P%2E+%282004%29+EuroWordNet%3A+a+multilingual+database+of+autonomous+and+language+specific+wordnets+connected+via+an+Inter%2DLingual%2DIndex%2E+International+Journal+of+Lexicography%2C+Vol%2E17+No%2E+2%2C+OUP%2C+161%2D1)
22. Wedyan, M., Alhadidi, B., and Alrabea, A. (2012). The effect of using a thesaurus in arabic information retrieval system. International Journal of Computer Science Issues, IJCSI, 9(1):431–435. [find](https://scholar.google.com/scholar?q=Wedyan%2C+M%2E%2C+Alhadidi%2C+B%2E%2C+and+Alrabea%2C+A%2E+%282012%29%2E+The+effect+of+using+a+thesaurus+in+arabic+information+retrieval+system%2E+International+Journal+of+Computer+Science+Issues%2C+IJCSI%2C+9%281%29%3A431%E2%80%93435%2E)
23. Xu, J. and Croft, W. B. (2000). Improving the Effectiveness of Information Retrieval with Local Context Analysis. ACM Transactions on Information Systems (TOIS), 18(1):79–112. [↗](https://doi.org/10.1145/333135.333138)
24. Xu, J., Fraser, A., and Weischedel, R. (2002). Empirical Studies in Strategies for Arabic Retrieval. In Annual ACM Conference on Research and Development in Information Retrieval: Proceedings of the 25 th annual international ACM SIGIR conference on Research and development in information retrieval, volume 11, pages 269–274. [find](https://scholar.google.com/scholar?q=Xu%2C+J%2E%2C+Fraser%2C+A%2E%2C+and+Weischedel%2C+R%2E+%282002%29%2E+Empirical+Studies+in+Strategies+for+Arabic+Retrieval%2E+In+Annual+ACM+Conference+on+Research+and+Development+in+Information+Retrieval%3A+Proceedings+of+the+2)

*Updated 24 October 2026*