---
title: "Deep Learning Based Amharic Chatbot for FAQs in — Hailu 2024"
description: "The paper tackles a practical gap: Ethiopian university students repeatedly ask staff the same questions, and no chatbot existed for answering such FAQs…"
authors: ["Goitom Ybrah Hailu", "Hadush Hailu", "Shishay Welay"]
year: 2024
venue: "arXiv:2402.01720 · Machine Learning (cs.LG), V1, 2024"
source: "https://arxiv.org/abs/2402.01720"
doi: "10.48550/arxiv.2402.01720"
arxiv: "2402.01720"
licence: "CC BY 4.0 (arXiv)"
licence_url: "http://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/hailu-2024-deep-learning-based-amharic-chatbot/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/hailu-2024-deep-learning-based-amharic-chatbot/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2024](https://md.research.elamur.ai/free/en/papers/2024/page/2/) / Deep Learning Based Amharic Chatbot for FAQs in Universities

# Deep Learning Based Amharic Chatbot for FAQs in Universities

**Authors:** Goitom Ybrah Hailu, Hadush Hailu, Shishay Welay · **Year:** 2024 · **Venue:** arXiv:2402.01720 · Machine Learning (cs.LG), V1, 2024 · **Licence:** [CC BY 4.0 (arXiv)](http://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://arxiv.org/abs/2402.01720) · [PDF](https://arxiv.org/pdf/2402.01720) · [DOI 10.48550/arxiv.2402.01720](https://doi.org/10.48550/arxiv.2402.01720) · [arXiv 2402.01720](https://arxiv.org/abs/2402.01720) · [OpenAlex](https://openalex.org/W4391670495)

**Citations:** 1 citations (openalex-snapshot · 2026-10-14) · 1 in our library, of the 140884 papers whose bibliographies we read

## Summary
The paper tackles a practical gap: Ethiopian university students repeatedly ask staff the same questions, and no chatbot existed for answering such FAQs in Amharic, a morphologically rich, low-resource language. The authors framed the work as Design Science Research, following the six-step process of Peffers et al. They gathered real student questions through a questionnaire at Mekelle University and Aksum University, translated and corrected them into Amharic with Google Translator and native experts, and organized them into 60 intent topics in a JSON file. Input text is tokenized, normalized to handle Amharic Fidel character variants, cleaned of stop words, and stemmed. Three intent classifiers were compared: SVM, Multinomial Naïve Bayes, and a deep neural network built with TensorFlow, Keras and NLTK. The neural model performed best, at 91.55% accuracy. The bot was connected to Facebook Messenger and hosted on Heroku for round-the-clock use. The work matters as an early end-to-end, openly released Amharic FAQ assistant and a template for other under-resourced languages.

## Key points
- The dataset came from questionnaires filled in by 80 engineering students at two universities (Mekelle and Aksum), chosen by purposive sampling, and was reduced to 60 topics.
- Each intent in the JSON file stores a tag, sample patterns, candidate responses (one is picked at random), a context set and a context filter.
- The best model, a deep neural network trained with the Adam optimizer and a SoftMax output, reached 91.55% accuracy and 0.3548 validation loss, beating the SVM and Multinomial Naïve Bayes baselines.
- Preprocessing targets Amharic-specific issues: Fidel (script) variation, morphological variation and lexical gaps.
- The bot was deployed via Facebook Messenger on Heroku. The dataset, model and code are released on GitHub, and integrating an Amharic WordNet is proposed as future work.

## Interesting ideas
- Intents carry a "context set" that changes the dialogue state and a "context filter" that limits which intents can match in that state, giving a cheap, inspectable way to model conversation flow. — For Mova (frames): A comparable explicit state flag could restrict which speech-act or frame readings are allowed given the preceding turn, keeping the disambiguation choice traceable.
- Script normalization collapses interchangeable Fidel characters before classification, so surface spelling variants do not split the vocabulary. — For Mova (tokens): A transparent normalization table for English spelling variants (British vs. American, hyphenation) placed before lemmatization would help lemmas and lookups line up, with each mapping logged for explanation.
- The authors gathered evaluation material by asking real users what they actually ask, rather than inventing questions. — For Mova (answer): Test sets for the QA readers could be seeded from questions collected from real learners, giving a more realistic difficulty profile than synthetic items.
- Classic linear classifiers (SVM, Naïve Bayes) were kept as baselines next to the neural model on the same data. — For Mova (learn): Every new learned component could be reported against a simple, cheap baseline on identical splits, which shows how much the added complexity really buys.
- The authors see a WordNet as the remedy for lexical gaps, where users phrase a question with synonyms absent from the training patterns. — For Mova (world): WordNet synonym and hypernym links in the compiled ontology could map unseen wording onto known concepts, so matching relies on meaning rather than surface overlap.

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

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/)

## Cite

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

```bibtex
@misc{hailu2024deep,
  title = {{Deep Learning Based Amharic Chatbot for FAQs in Universities}},
  author = {Hailu, Goitom Ybrah and Hailu, Hadush and Welay, Shishay},
  year = {2024},
  doi = {10.48550/arxiv.2402.01720},
  eprint = {2402.01720},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2402.01720}
}
```

- [← previous Dating ancient manuscripts using radiocarbon and AI-based writing style analysis](https://md.research.elamur.ai/free/en/papers/langstream-2024-popovic-dating-ancient-manuscripts-using-radiocarbon/)
- [next → DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence](https://md.research.elamur.ai/free/en/papers/deepseek-2024-ai-deepseek-coder-v2-breaking-barrier/)
- [↑ back to papers of 2024](https://md.research.elamur.ai/free/en/papers/2024/page/2/)

## References

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

1. A. Ali, Z. Memon, A. H. Jalbani, and M. Shaikh, “Multi-agent communication system with chatbots,” Mehran University Research Journal of Engineering and Technology, vol. 37, no. 3, pp. 663–672, 2018. [find](https://scholar.google.com/scholar?q=A%2E+Ali%2C+Z%2E+Memon%2C+A%2E+H%2E+Jalbani%2C+and+M%2E+Shaikh%2C+%E2%80%9CMulti%2Dagent+communication+system+with+chatbots%2C%E2%80%9D+Mehran+University+Research+Journal+of+Engineering+and+Technology%2C+vol%2E+37%2C+no%2E+3%2C+pp%2E+663%E2%80%93672%2C+2018%2E)
2. A. Khanna, B. Pandey, K. Vashishta, K. Kalia, B. Pradeepkumar, and T. Das, “A study of today’s A.I. through chatbots and rediscovery of machine intelligence,” International Journal of u- and e-Service, Science and Technology, vol. 8, no. 7, pp. 277–284, 2015. [find](https://scholar.google.com/scholar?q=A%2E+Khanna%2C+B%2E+Pandey%2C+K%2E+Vashishta%2C+K%2E+Kalia%2C+B%2E+Pradeepkumar%2C+and+T%2E+Das%2C+%E2%80%9CA+study+of+today%E2%80%99s+A%2EI%2E+through+chatbots+and+rediscovery+of+machine+intelligence%2C%E2%80%9D+International+Journal+of+u%2D+and+e%2DService)
3. D. Khurana, A. Koli, K. Khatter, and S. Singh, “Natural language processing: State of the art, current trends and challenges,” Manav Rachna International University, 2017. [find](https://scholar.google.com/scholar?q=D%2E+Khurana%2C+A%2E+Koli%2C+K%2E+Khatter%2C+and+S%2E+Singh%2C+%E2%80%9CNatural+language+processing%3A+State+of+the+art%2C+current+trends+and+challenges%2C%E2%80%9D+Manav+Rachna+International+University%2C+2017%2E)
4. B. A. Shawar and E. Atwell, “Chatbots: Are they really useful?” ResearchGate, 2007. [find](https://scholar.google.com/scholar?q=B%2E+A%2E+Shawar+and+E%2E+Atwell%2C+%E2%80%9CChatbots%3A+Are+they+really+useful%3F%E2%80%9D+ResearchGate%2C+2007%2E)
5. B. A. Shawar, E. Atwell, and A. Roberts, “FAQchat as an information retrieval system,” in Proceedings of the Human Language Technologies Conference, 2005. [find](https://scholar.google.com/scholar?q=B%2E+A%2E+Shawar%2C+E%2E+Atwell%2C+and+A%2E+Roberts%2C+%E2%80%9CFAQchat+as+an+information+retrieval+system%2C%E2%80%9D+in+Proceedings+of+the+Human+Language+Technologies+Conference%2C+2005%2E)
6. B. R. Ranoliya, N. Raghuwanshi, and S. Singh, “Chatbot for university related FAQs,” in Proceedings of the IEEE International Conference on Advances in Computing, Communications and Informatics, 2017, pp. 1525–1530. [find](https://scholar.google.com/scholar?q=B%2E+R%2E+Ranoliya%2C+N%2E+Raghuwanshi%2C+and+S%2E+Singh%2C+%E2%80%9CChatbot+for+university+related+FAQs%2C%E2%80%9D+in+Proceedings+of+the+IEEE+International+Conference+on+Advances+in+Computing%2C+Communications+and+Informatics%2C+2017%2C)
7. S. Nair, S. AD, S. SP, and T. Sinha, “Implementation of FAQ pages using chatbot,” International Journal of Computer Science and Information Security, vol. 16, no. 6, pp. 187–194, 2018. [find](https://scholar.google.com/scholar?q=S%2E+Nair%2C+S%2E+AD%2C+S%2E+SP%2C+and+T%2E+Sinha%2C+%E2%80%9CImplementation+of+FAQ+pages+using+chatbot%2C%E2%80%9D+International+Journal+of+Computer+Science+and+Information+Security%2C+vol%2E+16%2C+no%2E+6%2C+pp%2E+187%E2%80%93194%2C+2018%2E)
8. R. Singh, M. Paste, N. Shinde, H. Patel, and N. Mishra, “Chatbot using TensorFlow for small businesses,” in Second International Conference on Inventive Communication and Computational Technologies (ICICCT), 2018, pp. 1614–1619. [find](https://scholar.google.com/scholar?q=R%2E+Singh%2C+M%2E+Paste%2C+N%2E+Shinde%2C+H%2E+Patel%2C+and+N%2E+Mishra%2C+%E2%80%9CChatbot+using+TensorFlow+for+small+businesses%2C%E2%80%9D+in+Second+International+Conference+on+Inventive+Communication+and+Computational+Technologies+%28I)
9. B. A. Shawar and E. Atwell, “Arabic question-answering via instance based learning from an FAQ corpus,” in CL2009 International Conference on Corpus Linguistics, 2009. [find](https://scholar.google.com/scholar?q=B%2E+A%2E+Shawar+and+E%2E+Atwell%2C+%E2%80%9CArabic+question%2Danswering+via+instance+based+learning+from+an+FAQ+corpus%2C%E2%80%9D+in+CL2009+International+Conference+on+Corpus+Linguistics%2C+2009%2E)
10. D. A. Ali and N. Habash, “Botta: An Arabic dialect chatbot,” in International Conference on Computational Linguistics, 2016, pp. 208– 212. [find](https://scholar.google.com/scholar?q=D%2E+A%2E+Ali+and+N%2E+Habash%2C+%E2%80%9CBotta%3A+An+Arabic+dialect+chatbot%2C%E2%80%9D+in+International+Conference+on+Computational+Linguistics%2C+2016%2C+pp%2E+208%E2%80%93+212%2E)
11. S. M. Yimam, “TETEYEQ: Amharic question answering system for factoid questions,” Master’s thesis, Addis Ababa University, 2009. [find](https://scholar.google.com/scholar?q=S%2E+M%2E+Yimam%2C+%E2%80%9CTETEYEQ%3A+Amharic+question+answering+system+for+factoid+questions%2C%E2%80%9D+Master%E2%80%99s+thesis%2C+Addis+Ababa+University%2C+2009%2E)
12. D. A. Zeleke, “LETEYEQ: A web based Amharic question answering system for factoid questions using machine learning approach,” Master’s thesis, Addis Ababa University, 2013. [find](https://scholar.google.com/scholar?q=D%2E+A%2E+Zeleke%2C+%E2%80%9CLETEYEQ%3A+A+web+based+Amharic+question+answering+system+for+factoid+questions+using+machine+learning+approach%2C%E2%80%9D+Master%E2%80%99s+thesis%2C+Addis+Ababa+University%2C+2013%2E)
13. T. Abedissa, “Amharic question answering for definitional, biographical and description questions,” Master’s thesis, Addis Ababa University, 2013. [find](https://scholar.google.com/scholar?q=T%2E+Abedissa%2C+%E2%80%9CAmharic+question+answering+for+definitional%2C+biographical+and+description+questions%2C%E2%80%9D+Master%E2%80%99s+thesis%2C+Addis+Ababa+University%2C+2013%2E)
14. A. R. Hevner, S. Ram, S. T. March, and J. Park, “Design science in information systems research,” MIS Quarterly, vol. 10, no. 2, pp. 199– 217, 2004. [find](https://scholar.google.com/scholar?q=A%2E+R%2E+Hevner%2C+S%2E+Ram%2C+S%2E+T%2E+March%2C+and+J%2E+Park%2C+%E2%80%9CDesign+science+in+information+systems+research%2C%E2%80%9D+MIS+Quarterly%2C+vol%2E+10%2C+no%2E+2%2C+pp%2E+199%E2%80%93+217%2C+2004%2E)
15. K. Peffers, T. Tuure, and C. Samir, “A design science research methodology for information systems research,” Journal of Management Information Systems, vol. 24, no. 3, pp. 45–78, 2007. [find](https://scholar.google.com/scholar?q=K%2E+Peffers%2C+T%2E+Tuure%2C+and+C%2E+Samir%2C+%E2%80%9CA+design+science+research+methodology+for+information+systems+research%2C%E2%80%9D+Journal+of+Management+Information+Systems%2C+vol%2E+24%2C+no%2E+3%2C+pp%2E+45%E2%80%9378%2C+2007%2E)
16. D. P. Kingma and J. L. Ba, “ADAM: A method for stochastic optimization,” in International Conference on Learning Representations (ICLR), 2015, pp. 1–15. V. C ONTRIBUTION AND F UTURE W ORK [find](https://scholar.google.com/scholar?q=D%2E+P%2E+Kingma+and+J%2E+L%2E+Ba%2C+%E2%80%9CADAM%3A+A+method+for+stochastic+optimization%2C%E2%80%9D+in+International+Conference+on+Learning+Representations+%28ICLR%29%2C+2015%2C+pp%2E+1%E2%80%9315%2E+V%2E+C+ONTRIBUTION+AND+F+UTURE+W+ORK)