Elamur Research⌕
☰
home / papers / 2017 / Developing a web-based workbook for English supporting the interaction of students and teachers

Developing a web-based workbook for English supporting the interaction of students and teachers

Paper · 2017 · Björn Rudzewitz, Ramon Ziai, Kordula De Kuthy et al.

Summary

The paper addresses a long-standing classroom bottleneck. Students need individual, prompt feedback on the English they produce, but one teacher with about 30 learners cannot give it in time, especially for homework. Intelligent tutors are common in fields like mathematics, yet rare in real language classrooms. Most language tutors remain research prototypes.

The authors built FeedBook together with the publisher Westermann Gruppe (Diesterweg). It is a browser-based version of the approved print workbook Camden Town Gymnasium 3, used in 7th-grade English in Baden-Württemberg. Students do exercises online and submit them. Teachers correct the work, helped by:

  • automatic highlighting,
  • a memory of feedback they gave earlier,
  • NLP-based suggestions for error annotation.

To decide which tasks to digitize, the authors classified every print activity, including how tightly its instructions limit what learners might write. FeedBook has run in pilot classes since October 2016, with positive reactions from users. The work matters because it puts NLP into everyday school practice and prepares the ground for automatic, step-by-step learner feedback.

Key points

  • FeedBook currently covers 55 of the workbook's 177 activities. The first supported types were short answers and fill-in-the-blanks, plus the matching and multiple-choice sub-tasks that prepare them.
  • Each activity was analysed along five dimensions:
    • activity type,
    • expected variation in learner language,
    • kind of input (check marks, single characters, words, sentences, texts),
    • whether it targets form or meaning,
    • which grammatical forms it targets.
  • Related deployed tutors discussed: Robo-Sensei (Japanese), E-Tutor (German), and TAGARELA (Portuguese, with a later UIMA-based architecture).
  • The interface is in German, works on any device (built with Bootstrap), and shows teacher ratings as traffic-light marks. Teachers also get aggregated views of class performance.
  • Planned next steps:
    • integrate the CoMiC short-answer assessment system,
    • move feedback from the teacher to the system,
    • run a randomized controlled field study comparing automatic feedback with teacher-only feedback.

Interesting ideas

  • I1. Before automating anything, sort tasks by how strongly the prompt restricts the possible answers, and automate only where a reference answer is realistic. This keeps effort on items where checking is reliable.
  • I2. Store the prompt, the target answer and the student answer together in one activity record, so the meaning check can compare all three directly.
  • I3. A feedback memory recalls a correction once given and offers it again when a similar error appears, so human effort accumulates instead of repeating.
  • I4. The planned evaluation pits the full automatic-feedback workbook against an otherwise identical version that passes on only teacher feedback, isolating the effect of one component.

For Mova

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

  • I1 (answer): School-course test items could be tagged by response constraint (gap, single word, sentence, free text), so the reader picks a matching strategy and reports when an item falls outside its reliable zone.
  • I2 (answer): QA on fables could keep question, gold answer and produced answer side by side as graphs, making each judgement traceable to a node-level comparison rather than surface overlap.
  • I3 (learn): Corrections from LLM teacher tiers or human review could be cached by error pattern and turned into candidate repair rules, cutting repeated teacher calls.
  • I4 (learn): The same paired design could serve as a negative control, running the pipeline with and without a given module on identical inputs to show what that module really contributes.

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

🎓University of TübingenTübingen · Germany75 works · 41,081 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksSocial Sciences · 1,772 worksInformation Systems · 1,437 worksEducation · 203 works

🧭 Topics 1Research subjects and languages the work is about.English · 2,263 works

Cite

Cite as Björn Rudzewitz, Ramon Ziai, Kordula De Kuthy and Detmar Meurers. 2017. Developing a web-based workbook for English supporting the interaction of students and teachers. Proceedings of the joint workshop on NLP for Computer Assisted Language Learning and NLP for Language Acquisition.

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

@inproceedings{rudzewitz2017developing,
  title = {{Developing a web-based workbook for English supporting the interaction of students and teachers}},
  author = {Björn Rudzewitz, Ramon Ziai, Kordula {De Kuthy} and Detmar Meurers},
  booktitle = {Proceedings of the joint workshop on NLP for Computer Assisted Language Learning and NLP for Language Acquisition},
  year = {2017},
  doi = {10.15496/publikation-61697},
  url = {https://doi.org/10.15496/publikation-61697}
}

References

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

  1. Daniel Karp, Yves Schabes, Martin Zaidel, and Dania Egedi. 1992. A freely available wide coverage morphological analyzer for English. In Proceedings of the 14th Conference on Computational Linguistics, COLING ’92, pages 950–955, Stroudsburg, PA. find
  2. Luiz Amaral and Detmar Meurers. 2011. On using intelligent computer-assisted language learning in real-life foreign language teaching and learning. ReCALL, 23(1):4–24. ↗
  3. Alison Mackey. 2006. Feedback, noticing and instructed second language learning. Applied Linguistics, 27(3):405–430. ↗
  4. Luiz Amaral, Detmar Meurers, and Ramon Ziai. 2011. Analyzing learner language: Towards a flexible NLP architecture for intelligent language tutors. Computer-Assisted Language Learning, 24(1):1–16. find
  5. Detmar Meurers and Markus Dickinson. 2017. Evidence and interpretation in language learning research: Opportunities for collaboration with computational linguistics. Language Learning, 67(2). http://dx.doi.org/10.1111/lang.12233. ↗
  6. Fred J. Damerau. 1964. A technique for computer detection and correction of errors. Communications of the ACM, 7:171–176. find
  7. Richard Eckart de Castilho and Iryna Gurevych. 2014. A broad-coverage collection of portable NLP components for building shareable analysis pipelines. In Proceedings of the Workshop on Open Infrastructures and Analysis Frameworks for HLT (OIAF4HLT) at COLING 2014, pages 1–11, Dublin, Ireland. ACL and Dublin City University.
  8. Detmar Meurers, Ramon Ziai, Niels Ott, and Stacey Bailey. 2011a. Integrating parallel analysis modules to evaluate the meaning of answers to reading comprehension questions. IJCEELL. Special Issue on Automatic Free-text Evaluation, 21(4):355–369. ↗
  9. Detmar Meurers, Ramon Ziai, Niels Ott, and Janina Kopp. 2011b. Evaluating answers to reading comprehension questions in context: Results for German and the role of information structure. In Proceedings of the TextInfer 2011 Workshop on Textual Entailment, pages 1–9, Edinburgh. ↗
  10. William DeSmedt. 1995. Herr Kommissar: An ICALL conversation simulator for intermediate German. In V. Melissa Holland, Jonathan Kaplan, and Michelle Sams, editors, Intelligent Language Tutors: Theory Guido Minnen, John Carroll, and Darren Pearce. 2001. Applied morphological processing of English. Natural Language Engineering, 7(3):207–233. find
  11. Noriko Nagata. 1993. Intelligent computer feedback for second language instruction. The Modern Language Journal, 77(3):330–339. ↗
  12. Noriko Nagata. 1996. Computer vs. workbook instruction in second language acquistion. CALICO Journal, 14(1):53–75. find
  13. Noriko Nagata. 1997. The effectiveness of computer-assisted metalinguistic instruction: A case study in Japanese. Foreign Language Annals, 30(2):187– 200. ↗
  14. Noriko Nagata. 2002. BANZAI: An application of natural language processing to web-based language learning. CALICO Journal, 19(3):583–599. find
  15. Noriko Nagata. 2009. Robo-Sensei’s NLP-based error detection and feedback generation. CALICO Journal, 26(3):562–579. ↗
  16. Joakim Nivre, Jens Nilsson, Johan Hall, Atanas Chanev, Gülsen Eryigit, Sandra Kübler, Svetoslav Marinov, and Erwin Marsi. 2007. MaltParser: A language-independent system for data-driven dependency parsing. Natural Language Engineering, 13(1):1–41. ↗
  17. Niels Ott, Ramon Ziai, Michael Hahn, and Detmar Meurers. 2013. CoMeT: Integrating different levels of linguistic modeling for meaning assessment. In Proceedings of the 7th International Workshop on Semantic Evaluation (SemEval), pages 608–616, Atlanta, GA. ACL. find
  18. Martı́ Quixal and Detmar Meurers. 2016. How can writing tasks be characterized in a way serving pedagogical goals and automatic analysis needs? CAL-ICO Journal, 33:19–48. ↗
  19. Kent E. Sabo, Robert K. Atkinson, Angela L. Barrus, Stacey S. Joseph, and Ray S. Perez. 2013. Searching for the two sigma advantage: Evaluating algebra intelligent tutors. Computers in Human Behavior, 29(4):1833–1840. ↗
  20. Helmut Schmid. 2005. A programming language for finite state transducers. In Proceedings of the 5th International Workshop on Finite State Methods in Natural Language Processing (FSMNLP), pages 308–309. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/rudzewitz-2017-developing-web-based-workbook-english/