---
title: "Adaptive Feedback Message Generation for Second Language"
description: "The paper describes a tutoring component for learners of Arabic as a second language. 18 references."
authors: ["Khaled Shaalan", "Marwa Magdy"]
year: 2011
venue: "Proceedings of the International Conference Recent Advances in Natural Language Processing 2011"
source: "https://aclanthology.org/R11-1110/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/shaalan-2011-adaptive-feedback-message-generation-second/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/shaalan-2011-adaptive-feedback-message-generation-second/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2011](https://md.research.elamur.ai/free/en/papers/2011/page/2/) / Adaptive Feedback Message Generation for Second Language Learners of Arabic

# Adaptive Feedback Message Generation for Second Language Learners of Arabic

**Authors:** [Khaled Shaalan](https://md.research.elamur.ai/free/en/authors/khaled-shaalan/) and Marwa Magdy · **Year:** 2011 · **Venue:** [Proceedings of the International Conference Recent Advances in Natural Language Processing 2011](https://md.research.elamur.ai/free/en/venues/ranlp/) ([RANLP 2011](https://md.research.elamur.ai/free/en/venues/ranlp/2011/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/R11-1110/) · [PDF](https://aclanthology.org/R11-1110.pdf) · [OpenAlex](https://openalex.org/W2250779906)

**Citations:** 2 citations (OpenAlex, 2026-10-14)

## Summary
The paper describes a tutoring component for learners of Arabic as a second language. It diagnoses mistakes in verbs and explains them at a level of detail matched to the individual learner. Answers are short, and each test question has a single correct answer. That answer is stored with feature structures giving root, pattern, affixes, tense, mood, agreement and similar properties. A word analyzer uses constraint relaxation and edit distance to propose prefix+stem+suffix segmentations of the faulty word. Heuristic rules prune these candidates. A shallow generator built on a Morphological Form Hierarchy then produces the stem that should have appeared, and a three-way match compares it with what the learner wrote. Errors fall into nine categories. A perturbation-style student model counts each learner's past mistakes per concept and places them at beginner, intermediate or advanced level. Advanced learners get only the broad error class, while beginners get the precise cause. On 116 real learner sentences containing 133 errors, 74.4% of errors were fully diagnosed. This suggests that methods tailored to learners suit morphologically rich languages better than spell-checker approaches.

## Key points
- Nine verb error categories are handled. They are CL (consonant letters), VL (vowel letters), VC (verb conjugation), CP (connected pronouns), VP (verb pattern), LC (lexical category), VT (verb tense), SVD (subject-verb disagreement) and VM (verb mood). These are grouped into word-formation, word-choice, and lexis–grammar interface errors.
- The pipeline has six main parts:
  - item bank, with dictation, word order, fill-in-blank and other question types, each annotated with per-word feature structures;
  - word analyzer;
  - lexical error checker, made up of analysis, classification and filtering;
  - student model;
  - tutoring module;
  - error priority queue, which ranks related errors, e.g. a tense error takes precedence over a conjugation error.
- The student model keeps a pair [S, T] for each concept/bug combination: S is how often the learner made that error, T is how often they met the concept. Every new learner starts at [0, 0].
- Evaluation used 116 sentences from learners with different first languages, containing 133 errors:
  - fully diagnosed: 99 (74.4%);
  - partially diagnosed: 25 (18.8%);
  - general indication only: 9 (6.8%).
- Recall by category:
  - CL and VM reached 100%;
  - VP was lowest at 57.1%, because the intended pattern is often ambiguous, e.g. whether the learner meant "knew" or "taught".

## Interesting ideas
- Instead of searching a lexicon for the closest word, the system generates the expected stem from target features and diffs it against the observed stem. The error then becomes an explicit, nameable feature mismatch.
- The three-way match splits two strings into a shared head, a shared tail and an error core, and only allows linguistically plausible edits: weak letters may be inserted or deleted, and only similar-sounding letters may be substituted. This makes edit distance much less noisy.
- The same diagnosis is presented at three depths: broad class, error type, and exact source. Which depth is shown depends on the learner's history, not on the diagnosis itself.
- Evaluation sorts results into fully diagnosed, partially diagnosed and general indication only, rather than a simple right/wrong. It also states what the user actually sees in each case.
- A priority queue based on error dependencies prevents one root error from producing a cascade of reports.

## For Mova

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

- **I1** (tokens): Morphology in crate en could check a proposed lemma+features by regenerating the surface form and logging the exact character diff as an explainable trace.
- **I2** (tokens): Normalization of misspelled or nonstandard English tokens could restrict edits to plausible classes, such as doubled consonants or vowel swaps, before lemma lookup.
- **I3** (answer): School-test readers could reveal their reasoning chain at adjustable granularity, giving a hint first and the full frame-level justification on request.
- **I4** (learn): Evaluations could score whether a parse or answer was correct at the root-cause level, the category level, or only flagged, exposing partial successes that a binary metric hides.
- **I5** (learn): Self-induced repair rules could be ordered so that a fix to an upstream cause, e.g. a wrong POS tag, takes precedence over the downstream tree errors it triggers.

*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

- [Cairo University](https://md.research.elamur.ai/free/en/universities/cairo-university/) — University, Giza, Egypt · 43 works · 1,352 citations
- [British University in Dubai](https://md.research.elamur.ai/free/en/universities/british-university-in-dubai/) — University, Dubai, United Arab Emirates · 11 works · 589 citations

📚 Shelves (6): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,954 works · [🎓 Education](https://md.research.elamur.ai/free/en/shelves/social-sciences/education/) · 212 works

🧭 Topics (2): [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/) · 3,066 works · [🖍️ Grammatical error correction & learner language](https://md.research.elamur.ai/free/en/topics/grammatical-error-correction/) · 162 works

## Cite

Cite as: Khaled Shaalan and Marwa Magdy. 2011. Adaptive Feedback Message Generation for Second Language Learners of Arabic. Proceedings of the International Conference Recent Advances in Natural Language Processing 2011.

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

```bibtex
@inproceedings{shaalan2011adaptive,
  title = {{Adaptive Feedback Message Generation for Second Language Learners of Arabic}},
  author = {Khaled Shaalan and Marwa Magdy},
  booktitle = {Proceedings of the International Conference Recent Advances in Natural Language Processing 2011},
  year = {2011},
  url = {https://aclanthology.org/R11-1110/}
}
```

- [← previous A New Representation Model for the Automatic Recognition and Translation of Arabic Named Entities with NooJ](https://md.research.elamur.ai/free/en/papers/fehri-2011-new-representation-model-automatic-recognition/)
- [next → Domain-Dependent Detection of Light Verb Constructions](https://md.research.elamur.ai/free/en/papers/nagy-2011-domain-dependent-detection-light-verb/)
- [↑ back to papers of 2011](https://md.research.elamur.ai/free/en/papers/2011/page/2/)

## References

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

1. Abd Alghaniy, K. E. 1998. Arabic and Malaysian Languages from Phonological and Morphological Perspective: A Contrastive Analysis Approach. [find](https://scholar.google.com/scholar?q=Abd+Alghaniy%2C+K%2E+E%2E+1998%2E+Arabic+and+Malaysian+Languages+from+Phonological+and+Morphological+Perspective%3A+A+Contrastive+Analysis+Approach%2E)
2. Master Thesis, Cairo University, Egypt, 1998. [find](https://scholar.google.com/scholar?q=Master+Thesis%2C+Cairo+University%2C+Egypt%2C+1998%2E)
3. Ali, M. B. 1998. Linguistic Analysis of Mistakes by Students at the University of Malaya: An Error Analysis Approach. [find](https://scholar.google.com/scholar?q=Ali%2C+M%2E+B%2E+1998%2E+Linguistic+Analysis+of+Mistakes+by+Students+at+the+University+of+Malaya%3A+An+Error+Analysis+Approach%2E)
4. Master Thesis, Cairo University, Egypt, 1998. [find](https://scholar.google.com/scholar?q=Master+Thesis%2C+Cairo+University%2C+Egypt%2C+1998%2E)
5. Cavalli-Sforza, V., Soudi, A. and Mitamura, T. 2000. [Arabic Morphology Generation Using a Concatenative Strategy](https://md.research.elamur.ai/free/en/papers/sforza-2000-arabic-morphology-generation-concatenative-strategy/). In Proceedings of the 1st Conference, NAACL 00. Seattle, Washington, pp: 86-93.
6. Elmi, M. A. and Evens, M. 1998. Spelling Correction Using Context. In Proceedings of 36th ACL 98, Montreal, Canada, pp: 360-364. [find](https://scholar.google.com/scholar?q=Elmi%2C+M%2E+A%2E+and+Evens%2C+M%2E+1998%2E+Spelling+Correction+Using+Context%2E+In+Proceedings+of+36th+ACL+98%2C+Montreal%2C+Canada%2C+pp%3A+360%2D364%2E)
7. Faltin, A. V., L’haire, S. and Ndiaye, M. 2005. A Spell Checker for Language Learners of French and a Learner Corpus. In Proceedings of EURO-CALL 05. Cracow, Poland. [find](https://scholar.google.com/scholar?q=Faltin%2C+A%2E+V%2E%2C+L%E2%80%99haire%2C+S%2E+and+Ndiaye%2C+M%2E+2005%2E+A+Spell+Checker+for+Language+Learners+of+French+and+a+Learner+Corpus%2E+In+Proceedings+of+EURO%2DCALL+05%2E+Cracow%2C+Poland%2E)
8. Faltin, A. V. 2003. Syntactic Error Diagnosis in the Context of Computer Assisted Language Learning. [find](https://scholar.google.com/scholar?q=Faltin%2C+A%2E+V%2E+2003%2E+Syntactic+Error+Diagnosis+in+the+Context+of+Computer+Assisted+Language+Learning%2E)
9. PhD Thesis, University of Geneva, Switzerland, 2003. [find](https://scholar.google.com/scholar?q=PhD+Thesis%2C+University+of+Geneva%2C+Switzerland%2C+2003%2E)
10. Habash, N., Soudi, A., and Buckwalter, T. 2007. On Arabic Transliteration. In Arabic Computational Morphology: Knowledge-based and Empirical Methods. [find](https://scholar.google.com/scholar?q=Habash%2C+N%2E%2C+Soudi%2C+A%2E%2C+and+Buckwalter%2C+T%2E+2007%2E+On+Arabic+Transliteration%2E+In+Arabic+Computational+Morphology%3A+Knowledge%2Dbased+and+Empirical+Methods%2E)
11. Soudi, Abdelhadi; van den Bosch, Antal; Neumann, Günter (Eds.), 2007. ISBN: 978-1-4020-6045-8 [find](https://scholar.google.com/scholar?q=Soudi%2C+Abdelhadi%3B+van+den+Bosch%2C+Antal%3B+Neumann%2C+G%C3%BCnter+%28Eds%2E%29%2C+2007%2E+ISBN%3A+978%2D1%2D4020%2D6045%2D8)
12. Hsieh, C.-C., Tsai, T.-H., Wible, D. and Hsu, W.-L. 2002. Exploiting Knowledge Representation in an Intelligent Tutoring System for English Lexical Errors. In Proceedings of ICCE 2002, Auckland, New Zealand, pp: 115-116. [find](https://scholar.google.com/scholar?q=Hsieh%2C+C%2E%2DC%2E%2C+Tsai%2C+T%2E%2DH%2E%2C+Wible%2C+D%2E+and+Hsu%2C+W%2E%2DL%2E+2002%2E+Exploiting+Knowledge+Representation+in+an+Intelligent+Tutoring+System+for+English+Lexical+Errors%2E+In+Proceedings+of+ICCE+2002%2C+Auckland%2C+New+Z)
13. Jassem, J. A. 2000. Study on Second Language Learners of Arabic: An Error Analysis Approach. Kuala Lumpur (Malaysia): A.S. Noordeen. ISBN 983-065-093-6. [find](https://scholar.google.com/scholar?q=Jassem%2C+J%2E+A%2E+2000%2E+Study+on+Second+Language+Learners+of+Arabic%3A+An+Error+Analysis+Approach%2E+Kuala+Lumpur+%28Malaysia%29%3A+A%2ES%2E+Noordeen%2E+ISBN+983%2D065%2D093%2D6%2E)
14. Rimrott, A. 2003. SANTY: A Spell Checking Algorithm for Treating Predictable Verb Inflection Mistakes Made by Non-Native Writers of German. Term Paper for LING 807 – Computational Linguistics at Simon Fraser University (Burnaby, Canada). [find](https://scholar.google.com/scholar?q=Rimrott%2C+A%2E+2003%2E+SANTY%3A+A+Spell+Checking+Algorithm+for+Treating+Predictable+Verb+Inflection+Mistakes+Made+by+Non%2DNative+Writers+of+German%2E+Term+Paper+for+LING+807+%E2%80%93+Computational+Linguistics+at+Simon)
15. Shaalan, K., Magdy, M., Fahmy, A. 2011. Morphological analysis of ill formed Arabic verbs for second language learners. In McCarthy, P.M & Boonthum, C. (ed.), Applied Natural Language Processing and content analysis: Identification, Investigation, and Resolution (In Press). [find](https://scholar.google.com/scholar?q=Shaalan%2C+K%2E%2C+Magdy%2C+M%2E%2C+Fahmy%2C+A%2E+2011%2E+Morphological+analysis+of+ill+formed+Arabic+verbs+for+second+language+learners%2E+In+McCarthy%2C+P%2EM+%26+Boonthum%2C+C%2E+%28ed%2E%29%2C+Applied+Natural+Language+Processing+and+c)
16. Shaalan, K., Magdy, M., Fahmy, A. 2010a. Morphological Analysis of Ill-formed Arabic Verbs in Intelligent Language Tutoring Framework. In the Proceedings of FLAIRS-23, Applied Natural Language Processing Track, Florida, USA, 2010. [find](https://scholar.google.com/scholar?q=Shaalan%2C+K%2E%2C+Magdy%2C+M%2E%2C+Fahmy%2C+A%2E+2010a%2E+Morphological+Analysis+of+Ill%2Dformed+Arabic+Verbs+in+Intelligent+Language+Tutoring+Framework%2E+In+the+Proceedings+of+FLAIRS%2D23%2C+Applied+Natural+Language+Process)
17. Shaalan, K., Samy, D. and Magdy, M. 2010b. Towards Resolving Morphological Ambiguity in Arabic Intelligent Language Tutoring Framework. In Proceedings of International Workshop on Supporting e-Learning with Language Resources and Semantic Data (LREC 2010), Valletta, Malta Shaalan K. 2005a. An Intelligent Computer Assisted Language Learning System for Arabic Learners, in Computer Assisted Language Learning: An International Journal, Taylor & Francis Group Ltd., 18(1 & 2): 81-108. [find](https://scholar.google.com/scholar?q=Shaalan%2C+K%2E%2C+Samy%2C+D%2E+and+Magdy%2C+M%2E+2010b%2E+Towards+Resolving+Morphological+Ambiguity+in+Arabic+Intelligent+Language+Tutoring+Framework%2E+In+Proceedings+of+International+Workshop+on+Supporting+e%2DLearnin)
18. Shaalan K. 2005b. Arabic GramCheck: A Grammar Checker for Arabic, Software Practice and Experience, John Wiley & sons Ltd., UK, 35(7):643-665. [find](https://scholar.google.com/scholar?q=Shaalan+K%2E+2005b%2E+Arabic+GramCheck%3A+A+Grammar+Checker+for+Arabic%2C+Software+Practice+and+Experience%2C+John+Wiley+%26+sons+Ltd%2E%2C+UK%2C+35%287%29%3A643%2D665%2E)

*Updated 2 November 2026*