Atatürk İlkelerinin Daha İyi Anlaşılması İçin Güvenilir ve Sürdürülebilir RAG destekli LLM Sistemi: AtamQAS

Öz

Büyük dil modelleri (LLM), günümüzün en popüler teknolojilerinden biri haline gelmiştir. Bu modellerin kullanıldığı alanlardan biri de eğitim alanıdır. Ancak, bu modellerin doğası gereği, halüsinasyon gibi bazı sorunlar ortaya çıkmaktadır. Bu sorunu ortadan kaldırmak veya en aza indirmek için bazı teknik çalışmalar yürütülmektedir. Bu çalışmada, modern Türkiye'nin kurucusu Mustafa Kemal Atatürk'ün devrimleri ve ilkelerinin eğitiminde kullanılmak üzere daha güvenilir bir yapı önerdik. Bu yapıda, Bilgi Geri Getirme ile Artırılmış Üretim (RAG) mimarisini ve özellikle bu alan için oluşturduğumuz benzersiz bir veri setini kullandık. AtamQAS adını verdiğimiz aracımızı, alanında uzman kişiler tarafından genel amaçlı yapay zeka araçlarıyla karşılaştırmayı sağladık. Sonuçlar, AtamQAS'ın genel amaçlı yapay zeka araçlarından daha güvenilir çıktılar ürettiğini göstermiştir.

Anahtar Kelimeler

• Mustafa Kemal Atatürk
• Üretken Yapay Zeka
• Büyük Dil Modelleri
• Bilgi Geri Getirme ile Artırılmış Üretim
• Soru Cevaplama Sistemleri
• İnsan-Bilgisayar Arayüzü

Etik Beyan

Bu makalede, bilimsel araştırma ve yayın etiği ilkeleri takip edilmiştir. Bu çalışma insan veya hayvan denekleri içermemiştir ve ek etik komite onayı gerektirmemiştir.

Towards a Reliable and Sustainable RAG-Supported LLM System for Better Understanding of Atatürk’s Principles: AtamQAS

Öz

Large language models (LLM) have become one of today’s current and popular technologies. One of the areas where these models are used is the education domain. However, due to the nature of these models, there are some problems such as hallucinations. Some technical studies are being carried out to eliminate or minimize this problem. In this study, we proposed a more reliable structure to be used in the education of the revolutions and principles of Mustafa Kemal Atatürk, the founder of modern Türkiye. In this structure, we used Retrieval Augmented Generation (RAG) architecture and a unique dataset we created specifically for the field. We enabled the comparison of our tool, which we named AtamQAS, with general-purpose artificial intelligence tools by experts in the field. The results showed that AtamQAS produced more reliable outputs than general-purpose artificial intelligence tools.

Anahtar Kelimeler

• Mustafa Kemal Atatürk
• Generative AI
• Large Language Models
• Retrieval Augmented Generation
• Question Answering Systems
• Human-computer interface

Etik Beyan

In this article, the principles of scientific research and publication ethics were followed. This study did not involve human or animal subjects and did not require additional ethics committee approval.

Kaynakça

1. Acun, C., & Acun, R. (2023). GAI-Enhanced Assignment Framework: A Case Study on Generative AI Powered History Education. In NeurIPS’23 Workshop on Generative AI for Education. https://gaied.org/neurips2023/files/48/48_paper.pdf
2. Acun, F. (2011). Tarihin kaynakları. In: Şimşek A (ed.) Tarih nasıl yazılır, 119-149.
3. Alan, A. Y., Karaarslan, E., & Aydin, Ö. (2025). A RAG-based question answering system proposal for understanding Islam: MufassirQAS LLM. arXiv preprint arXiv:2401.15378. https://doi.org/10.48550/arXiv.2401.15378
4. Alan, A. Y., Karaarslan, E., & Aydın, Ö. (2025). Improving LLM reliability with RAG in religious question-answering: MufassirQAS. Turkish Journal of Engineering, 9(3), 544-559. https://doi.org/10.31127/tuje.1624773
5. Albayrak, M. (2010). Kemalizmin düşünsel temelleri ve tarihsel oluşumu. Atatürk Araştırma Merkezi Dergisi, 26(77), 307-344. https://izlik.org/JA42RX52YS
6. Ali, I., Selim, A., & Skender, F. (2026). Leveraging OCR and AI tools: A comparative guide to enhancing data processing and decision-making efficiency in the digital age. Turkish Journal of Engineering, 10(1), 129-142. https://doi.org/10.31127/tuje.1796566
7. Alkan, A. (2025). Analysis of factors affecting academic success with machine learning: Data-driven inferences in education. Turkish Journal of Engineering, 10(1), 48-62. https://doi.org/10.31127/tuje.1779491
8. Aydın, Ö. (2021). Classification of documents extracted from images with optical character recognition methods. Computer Science, 6(2), 46-55. https://izlik.org/JA95DG25BF

9. Aydın, Ö. (2023). Google Bard generated literature review: Metaverse. Journal of AI, 7(1), 1-14. https://doi.org/10.61969/jai.1311271
10. Aydın, Ö., & Karaarslan, E. (2022). OpenAI OpenAI ChatGPT Generated Literature Review: Digital Twin in Healthcare. In Ö. Aydın (Ed.), Emerging Computer Technologies 2, 22-31. https://doi.org/10.2139/ssrn.4308687
11. Aydın, Ö., & Karaarslan, E. (2023). Is ChatGPT leading generative AI? What is beyond expectations?. Academic Platform Journal of Engineering and Smart Systems, 11(3), 118-134. https://doi.org/10.21541/apjess.1293702
12. Aydın, Ö., Karaarslan, E., Erenay, F. S., & Bacanin, N. (2025). Generative AI in academic writing: A comparison of DeepSeek, Qwen, ChatGPT, Gemini, Llama, Mistral, and Gemma. Turkish Journal of Engineering, 10(2), 592-607. https://doi.org/10.31127/tuje.1838765
13. Babits, C. (2025). Where are the citations? ChatGPT discussions in the History of Sexuality. Teaching History: A Journal of Methods, 49(1), 5-9. https://doi.org/10.33043/a2az53dz2
14. Baruwa, I. (2024). A Conversation with ChatGPT: Philosophy, Critical Thinking, and Higher Education. Journal of AI, 9(1), 1-12. https://doi.org/10.61969/jai.1498257
15. Bayley, E. (2023). Teaching historical empathy using AI. Teaching History, 57(3), 48-49. https://search.informit.org/doi/10.3316/informit.323270697575953
16. Béchard, P., & Ayala, O. M. (2024, June). Reducing hallucination in structured outputs via Retrieval-Augmented Generation. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 6: Industry Track) (pp. 228-238). https://doi.org/10.18653/v1/2024.naacl-industry.19
17. Black, A. (2025). The ChatGPT exam: Critiquing generative AI to assess learning. Teaching History: A Journal of Methods, 49(1), 34-41. https://doi.org/10.33043/gg58bfzzg
18. Brandtzaeg, P.B., Følstad, A. (2017). Why People Use Chatbots. In: Kompatsiaris, I., et al. Internet Science. INSCI 2017. Lecture Notes in Computer Science(pp. 377-392), vol 10673. Springer, Cham. https://doi.org/10.1007/978-3-319-70284-1_30
19. Gossard, J. (2024). Introduction: Teaching history in the time of generative AI. Teaching History: A Journal of Methods, 49(1), 3-4. https://doi.org/10.33043/5zbfza2bc
20. Humphries, M. (2023). History and generative AI. Teaching History, 57(3), 4-9. https://search.informit.org/doi/10.3316/informit.322748974380724
21. Jo, H. (2024). From concerns to benefits: a comprehensive study of ChatGPT usage in education. International Journal of Educational Technology in Higher Education, 21(1), 35. https://doi.org/10.1186/s41239-024-00471-4
22. Kirchenbauer, J., & Barns, C. (2024). Hallucination reduction in large language models with retrieval-augmented generation using wikipedia knowledge. ArXiv Preprint. https://doi.org/10.31219/osf.io/pv7r5
23. Liu, N. F., Lin, K., Hewitt, J., Paranjape, A., Bevilacqua, M., Petroni, F., & Liang, P. (2024). Lost in the middle: How language models use long contexts. Transactions of the association for computational linguistics, 12, 157-173. https://doi.org/10.1162/tacl_a_00638
24. Molteno, M. I. (1972). History: Science or art?. Kleio, 4(2), 33-45. https://doi.org/10.1080/00232087285310091
25. Mottl, J., & Musílek, M. (2024, May). Information technology as a tool for teaching history with a focus on Artificial Intelligence and audiovisual material. In 2024 47th MIPRO ICT and Electronics Convention (MIPRO) (pp. 278-282). IEEE. https://doi.org/10.1109/MIPRO60963.2024.10569216
26. Özerdim, S. N. (1981). Nutuk'ta altı çizilmiş satırlar. Belleten, 45(177), 57-78. https://doi.org/10.37879/belleten.1981.57
27. Park, K., Song, H., Seo, S., Kim, J., & Suh, B. (2025, April). " Ask Sir Oliver Ingham": LLM-based Social Simulations for History Education. In Proceedings of the Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (pp. 1-13). https://doi.org/10.1145/3706599.3719728
28. Pittard, K. (2023). Navigating the future of history education: Harnessing the benefits of generative AI. Teaching History, 57(3), 27-31. https://search.informit.org/doi/10.3316/informit.323047101920855
29. Pope, A., & Ma, R. (2024). Exploring historians' critical use of generative AI technologies for history education. Proceedings of the Association for Information Science and Technology, 61(1), 1071-1073. https://doi.org/10.1002/pra2.1188
30. Rakin, S., Shibly, M.A.R., Hossain, Z.M., Akbar, M.M., Khan, Z. (2025). Leveraging the Domain Adaptation of Retrieval Augmented Generation Models for Question Answering and for Hallucination Reduction. In: Latifi, S. (eds) The 22nd International Conference on Information Technology-New Generations (ITNG 2025). ITNG 2025. Advances in Intelligent Systems and Computing, vol 1463. Springer, Cham. https://doi.org/10.1007/978-3-031-89063-5_42
31. Steinmaurer, A., Dengel, A., Comanici, M., Buchner, J., Memminger, J., & Gütl, C. (2025, February). Immersive learning in history education: exploring the capabilities of virtual avatars and large Language models. In International conference on immersive learning (pp. 363-374). Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-80475-5_26
32. Sun, Y., Yang, W., & Liu, Y. (2024, June). The application of constructing knowledge graph of oral historical archives resources based on LLM-RAG. In Proceedings of the 2024 8th International Conference on Information System and Data Mining (pp. 142-149). https://doi.org/10.1145/3686397.3686420
33. Şmşek, S. (2005). “People’s Houses” as a nationwide project for ideological mobilization in early Republican Turkey. Turkish Studies, 6(1), 71-91. https://doi.org/10.1080/1468384042000339339
34. Tonmoy, S. M., Zaman, S. M., Jain, V., Rani, A., Rawte, V., Chadha, A., & Das, A. (2024). A comprehensive survey of hallucination mitigation techniques in large language models. arXiv preprint arXiv:2401.01313. https://doi.org/10.48550/arXiv.2401.01313
35. Topal, İ. H. (2025). Theses Written by Turkish Authors About ChatGPT in The Field of Education and Training: A Content Analysis. Journal of AI, 9(1), 56-79. https://doi.org/10.61969/jai.1633592
36. Trust, T., & Maloy, R. W. (2023). AI in the history classroom: Strategies for rethinking teaching and learning. Teaching History, 57(3), 10-15. https://search.informit.org/doi/10.3316/informit.322842139237015
37. Vanichvasin, P. (2021). Chatbot Development as a Digital Learning Tool to Increase Students' Research Knowledge. International Education Studies, 14(2), 44-53. https://doi.org/10.5539/ies.v14n2p44
38. Washburn, J., & McCutchen, J. M. (2024). AI meets AI: ChatGPT as a pedagogical tool to teach American Indian history. Critical Humanities, 2(2), 1-27. https://doi.org/10.33470/2836-3140.1037
39. Wulgaert, R. (2023). Ithaca AI meets ancient Greek: Muses and robots in the classroom. Teaching History, 57(3), 16-20. https://search.informit.org/doi/10.3316/informit.322879405179531