Üretken Yapay Zeka Performans Kıyaslaması: ChatGPT ve DeepSeek'in Kullanıcı Etkileşimleri Üzerindeki Etkisi

Abstract

Bu çalışma, üretken yapay zekâ modelleri olan ChatGPT ve DeepSeek'in kullanıcı etkileşimleri üzerindeki performansını ve etkisini karşılaştırmalı bir analizle değerlendirmeyi amaçlamaktadır. Araştırmada, Kaggle.com'dan temin edilen ve Temmuz 2023 ile Şubat 2025 arasını kapsayan, 10.000'den fazla kullanıcı etkileşim verisi içeren sentetik bir veri seti kullanılmıştır. Makine öğrenmesi algoritmaları (regresyon, karar ağaçları, kümeleme) ve istatistiksel testler (t-test, ANOVA) aracılığıyla yanıt doğruluğu, oturum süresi, kullanıcı memnuniyeti gibi değişkenler analiz edilmiştir. Bulgular, DeepSeek'in özellikle teknik görevlerde (hata ayıklama, kod optimizasyonu) daha yüksek kullanıcı derecelendirmeleri aldığını, ChatGPT'nin ise içerik üretimi ve eğitim gibi alanlarda öne çıktığını göstermiştir. DeepSeek kullanıcıları genel olarak daha yüksek deneyim puanları ve derecelendirmeler bildirirken, daha düşük terk etme oranlarına sahip olduğu gözlemlenmiştir. Çalışma, her iki platformun güçlü yönlerini, sınırlılıklarını ve kullanıcı etkileşim dinamiklerini ortaya koyarak, platformların kullanım bağlamına göre performans farklılıkları sergilediğini ve veri gizliliği ile etik konuların önemli olduğunu vurgulamaktadır.

Keywords

• Üretken Yapay Zekâ
• ChatGPT
• DeepSeek
• Kullanıcı Etkileşimi
• Performans Kıyaslaması

Generative AI Performance Benchmarking: The Impact of ChatGPT and DeepSeek on User Interactions

Abstract

This study aims to evaluate and compare the performance and impact of generative AI models ChatGPT and DeepSeek on user interactions through a comparative analysis. The research utilizes a synthetic dataset obtained from Kaggle.com, encompassing over 10,000 user interaction records collected between July 2023 and February 2025. Variables such as response accuracy, session duration, and user satisfaction were analyzed using machine learning algorithms (regression, decision trees, clustering) and statistical tests (t-test, ANOVA). The findings indicate that DeepSeek received higher user ratings particularly in technical tasks (e.g., debugging, code optimization), while ChatGPT stood out in areas such as content generation and education. DeepSeek users generally reported higher experience scores and ratings, along with lower abandonment rates. The study highlights the strengths, limitations, and user interaction dynamics of both platforms, emphasizing that performance varies depending on the context of use and underlining the importance of data privacy and ethical considerations.

Keywords

• Generative Artificial Intelligence
• ChatGPT
• DeepSeek
• User Interaction
• Performance Comparison

References

1. Acemoglu, D., Autor, D., & Johnson, S. (2023). Can we have pro-worker ai. https://shapingwork.mit.edu/wp-content/uploads/2023/09/Pro-Worker-AI-Policy-Memo.pdf
2. Adiguzel, T., Kaya, M. H., & Cansu, F. K. (2023). Revolutionizing education with AI: Exploring the transformative potential of ChatGPT. Contemporary Educational Technology, 15(3).
3. Al-Abdullatif, A. M., & Alsubaie, M. A. (2024). ChatGPT in learning: Assessing students’ use intentions through the lens of perceived value and the influence of AI literacy. Behavioral Sciences, 14(9), 845.
4. Al-Busaidi, A. S., Raman, R., Hughes, L., Albashrawi, M. A., Malik, T., Dwivedi, Y. K., ... & Walton, P. (2024). Redefining boundaries in innovation and knowledge domains: Investigating the impact of generative artificial intelligence on copyright and intellectual property rights. Journal of Innovation & Knowledge, 9(4), 100630.
5. Alkaissi, H., & McFarlane, S. I. (2023). Artificial hallucinations in ChatGPT: implications in scientific writing. Cureus, 15(2).
6. Artetxe, M., Bhosale, S., Goyal, N., Mihaylov, T., Ott, M., Shleifer, S., & Stoyanov, V. (2021). Efficient large scale language modeling with mixtures of experts. https://arxiv.org/pdf/2112.10684
7. Bahrini, A., Khamoshifar, M., Abbasimehr, H., Riggs, R. J., Esmaeili, M., Majdabadkohne, R. M., & Pasehvar, M. (2023). ChatGPT: Applications, opportunities, and threats. In 2023 Systems and Information Engineering Design Symposium (SIEDS), 274-279.
8. Bell, R., & Bell, H. (2023). Entrepreneurship education in the era of generative artificial intelligence. Entrepreneurship Education, 6(3), 229–244. https://doi.org/10.1007/s41959-023-00099-x Borgesius, F. (2018). Discrimination, artificial intelligence, and algorithmic decision-making. Council of Europe, Directorate General of Democracy.

9. BytePlus. (2025). DeepSeek-Coder-V2: Revolutionizing AI-powered code generation. https://www.byteplus.com/en/topic/375561?title=deepseek-coder-v2-revolutionizing-ai-powered-code-generation
10. Chein, J., Martinez, S., & Barone, A. (2024). Can human intelligence safeguard against artificial intelligence? Exploring individual differences in the discernment of human from AI texts. Research Square, rs-3. CTOL Digital. (2025). Is DeepSeek Truly Open Source or Just Following Industry Norms? https://www.ctol.digital/news/is-deepseek-truly-open-source-or-following-industry-norms/ Dataloop. (t.y.). DeepSeek Coder V2 Instruct · Models. https://dataloop.ai/library/model/deepseek-ai_deepseek-coder-v2-instruct/
11. DeepSeek R1: DeepSeek-V3. (t.y.). https://deepseeksr1.com/v3/
12. DeepSeek-AI. (2024). DeepSeek-V3 Technical Report. https://arxiv.org/pdf/2412.19437
13. DeepSeek-Coder. (t.y.). DeepSeek Coder. https://github.com/deepseek-ai/DeepSeek-Coder
14. DeepSeek-R1. (t.y.). DeepSeek-R1 Training Overview. https://github.com/deepseek-ai/DeepSeek-R1 DeepSeek-V3. (t.y.). DeepSeek-V3. https://github.com/deepseek-ai/DeepSeek-V3
15. Du, N., Huang, Y., Dai, A. M., Tong, S., Lepikhin, D., Xu, Y., ... & Cui, C. (2022, June). Glam: Efficient scaling of language models with mixture-of-experts. https://arxiv.org/pdf/2112.06905
16. Dwivedi, Y. K., Kshetri, N., Hughes, L., Slade, E. L., Jeyaraj, A., Kar, A. K., & Wright, R. (2023). Opinion Paper:“So what if ChatGPT wrote it?” Multidisciplinary perspectives on opportunities, challenges and implications of generative conversational AI for research, practice and policy. International Journal of Information Management.
17. Eren, F., Dülek, L. N., Uraz, Ö. A., Kuşcu, B., & Sakallı, M. (2024). Yapay zeka endeksi kapsamında ülke bazlı yapay zeka politika stratejilerinin etkinliği: Performans ve verimlilik değerlendirmesi incelenmesi. Bilgi ve İletişim Teknolojileri Dergisi, 6(2), 113-148.
18. Firat, M. (2023). What ChatGPT means for universities: Perceptions of scholars and students. Journal of Applied Learning and Teaching, 6(1), 57-63.
19. Guo, D., Zhu, Q., Yang, D., Xie, Z., Dong, K., Zhang, W., ... & Liang, W. (2024). DeepSeek-Coder: When the large language model meets programming -- The rise of code intelligence. https://arxiv.org/pdf/2401.14196
20. Gupta, M. (2025). RLHF (OpenAI) vs Simple RL (DeepSeek). Medium. https://medium.com/data-science-in-your-pocket/rlhf-openai-vs-simple-rl-deepseek-93054cb509a4
21. Hariri, W. (2023). Unlocking the potential of ChatGPT: A comprehensive exploration of its applications, advantages, limitations, and future directions in natural language processing. https://arxiv.org/abs/2304.02017
22. IBM. (2025a). DeepSeek: sorting through the hype. https://www.ibm.com/think/topics/deepseek
23. IBM. (2025b). DeepSeek's reasoning AI shows power of small models, efficiently. https://www.ibm.com/think/news/deepseek-r1-ai
24. Information Commissioner's Office (ICO). (2023). Guidance on AI and data protection. ICO.
25. Liu, H., Zhou, Y., Li, M., Yuan, C., & Tan, C. (2024). Literature meets data: A synergistic approach to hypothesis generation. https://arxiv.org/pdf/2410.17309
26. Manik, M. M. H. (2025). ChatGPT vs. DeepSeek: A comparative study on AI-Based code generation. https://arxiv.org/pdf/2502.18467
27. Markov, T., Zhang, C., Agarwal, S., Eloundou, T., Lee, T., Adler, S., Jiang, A., & Weng, L. (2023). New and improved content moderation tooling. https://openai.com/blog/new-and-improved-content-moderation-tooling/
28. Mesko, B. (2017). The role of artificial intelligence in precision medicine. Expert Review of Precision Medicine and Drug Development, 2(5), 239-241.
29. Mogavi, R. H., Deng, C., Kim, J. J., Zhou, P., Kwon, Y. D., Metwally, A. H. S., ... & Hui, P. (2024). ChatGPT in education: A blessing or a curse? A qualitative study exploring early adopters’ utilization and perceptions. Computers in Human Behavior: Artificial Humans, 2(1).
30. Nazir, A., & Wang, Z. (2023). A comprehensive survey of ChatGPT: Advancements, applications, prospects, and challenges. Meta-radiology.
31. Neptune.ai (2025). Reinforcement learning from human feedback for LLMs. https://neptune.ai/blog/reinforcement-learning-from-human-feedback-for-llms OpenAI. (2023). GPT-4 Technical Report. https://arxiv.org/pdf/2303.08774 Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C., Mishkin, P., ... & Lowe, R. (2022). Training language models to follow instructions with human feedback. https://arxiv.org/pdf/2203.02155
32. Qawqzeh, Y. (2024). Exploring the influence of student interaction with ChatGPT on critical thinking, problem solving, and creativity. International Journal of Information and Education Technology, 14(4), 596-601.
33. Ray, P. P. (2023). ChatGPT: A comprehensive review on background, applications, key challenges, bias, ethics, limitations and future scope. Internet of Things and Cyber-Physical Systems, 3, 121-154.
34. Reuters. (2024). Italy fines OpenAI 15 million euros over privacy rules breach. Reuters. https://www.reuters.com/technology/italy-fines-openai-15-million-euros-over-privacy-rules-breach-2024-12-20/ Rivas, P., & Zhao, L. (2023). Marketing with chatgpt: Navigating the ethical terrain of gpt-based chatbot technology. AI, 4(2), 375-384.
35. Rozado, D. (2023). The political biases of ChatGPT. Social Sciences, 12(3), 148.