BİLİMSEL VERİNİN SANATSAL DÖNÜŞÜMÜ: YAPAY ZEKÂ DESTEKLİ BİLİMSEL GÖRSELLEŞTİRMELER

Abstract

Bilimsel verilerin doğru bir şekilde anlaşılması, yorumlanması ve daha geniş kitlelere etkili biçimde aktarılması açısından görselleştirme son derece önemli bir araçtır. Özellikle çok boyutlu, büyük ölçekli ve soyut veri kümeleriyle çalışılan günümüz bilimsel araştırmalarında, geleneksel yöntemler artan veri miktarı ve karmaşık yapılar karşısında yetersiz kalmaktadır. Yapay zekâ; veri işleme, simülasyon, eksik veri tamamlama gibi özellikleriyle karmaşık kavramları görsel olarak doğru ve anlaşılır hale getirirken, estetik bir boyut ekleyerek bilim iletişimini sanatsal bir anlatıma dönüştürmektedir. Çalışma, bu yeni yaklaşımların bilimsel keşif süreçlerine olan katkılarını görünür kılmayı ve gelecekte daha derin bir bilimsel anlayış geliştirilmesine zemin hazırlamayı amaçlamaktadır. Bu kapsamda bilimsel verilerin görsel sunumunda yapay zekâ destekli üretim araçlarının sunduğu estetik ve anlatısal olanaklar irdelenmiştir. Betimsel analiz ve karşılaştırmalı içerik çözümlemesi yöntemleri kullanılarak, farklı yapay zekâ araçlarıyla üretilen bilimsel görseller; estetik değer, görsel algı, bilimsel doğruluk ve anlatı gücü açısından değerlendirilmiştir. Görsellerin üretiminde kullanılan istemler sabit tutulmuş; her araçtan elde edilen çıktılar, referans görseller ile bağlamsal analiz çerçevesinde karşılaştırılmıştır. Elde edilen bulgular, yapay zekâ destekli üretimlerin yalnızca teknik açıdan değil, aynı zamanda kavramsal ve estetik düzeyde de anlam üretme potansiyeline sahip olduğunu göstermektedir. Bu bağlamda çalışma hem bilimsel iletişim hem de yaratıcı görselleştirme süreçleri açısından disiplinlerarası bir katkı sunmayı hedeflemektedir.

Keywords

• Bilimsel Görselleştirme
• Yapay Zekâ
• Bilim İletişimi
• Grafik Tasarımı
• Sanat

THE ARTISTIC TRANSFORMATION OF SCIENTIFIC DATA: AI-ASSISTED SCIENTIFIC VISUALIZATIONS

Abstract

Visualization is a crucial tool for accurately understanding, interpreting, and effectively communicating scientific data to wider audiences. In contemporary research, which often involves multidimensional, large-scale, and abstract datasets, traditional methods struggle with increasing data volumes and complexity. Artificial intelligence, with its capabilities in data processing, simulation, and missing data imputation, can transform complex concepts into clear, accurate visual representations while adding an aesthetic dimension that turns scientific communication into an artistic narrative. This study explores how these new approaches contribute to scientific discovery and foster deeper understanding. Using descriptive and comparative content analysis, visuals generated with different AI tools were evaluated for aesthetic value, visual perception, scientific accuracy, and narrative strength. Prompts were kept constant, and outputs from each tool were compared using reference images within a contextual analysis framework. Findings show that AI-assisted productions can generate meaning not only technically but also conceptually and aesthetically. This study aims to make an interdisciplinary contribution to both science communication and creative visualization practices.

Keywords

• Scientific Visualization
• Artificial Intelligence
• Science Communication
• Graphic Design
• Art

References

1. Alarcon, N. (2020). Using AI-based emulators to speed up simulations by billions of times. NVIDIA Developer. Erişim: 09.07.2024, https://developer.nvidia.com/blog/using-ai-based-emulators-to-speed-up-simulations-by-billions-of-times/
2. Ansys. (t.y.). The intersection of AI and simulation technology. Ansys. Erişim: 01.09.2024, https://www.ansys.com/blog/simulation-and-ai
3. Barros, Y. (2023). Unveiling the cosmic frontier: The synergy of artificial intelligence and astrophysics [Gönderi]. LinkedIn. Erişim: 22.10.2024, https://www.linkedin.com/pulse/unveiling-cosmic-frontier-synergy-artificial-yan-barros/
4. Benger, W., Haider, M., Stoeckl, J., Cosenza, B., Ritter, M., Steinhauser, D., ve Höller-Lugmayr, H. (2012). Visualization methods for numerical astrophysics. InTech. https://doi.org/10.5772/32203
5. Card, S., Mackinlay, J., ve Shneiderman, B. (1999). Readings in information visualization: Using vision to think. Morgan Kaufmann.
6. Daston, L., ve Galison, P. (2007). Objectivity. Zone Books.
7. Editverse. (t.y.). Yapay zekâ ile geliştirilmiş veri görselleştirme: Bilimsel grafiklemede yeni norm. Erişim: 15.12.2024, https://www.editverse.com/tr/AI-geli%C5%9Fmi%C5%9F-veri-g%C3%B6rselle%C5%9Ftirmesi-bilimsel-grafik-olu%C5%9Fturmada-yeni-norm-2024-2025/
8. Gajani, A. (2023). Harnessing the potential of AI in modern astronomy. Scientia. Erişim: 07.08.2024, https://scientiamag.org/harnessing-the-potential-of-artificial-intelligence-in-modern-astronomy/

9. Gibson, J. J. (1979). The ecological approach to visual perception: Classic edition. Houghton Mifflin.
10. Gould, J. (2023). Art and science: Close cousins or polar opposites? Nature. Erişim: 06.08.2024, https://www.nature.com/articles/d41586-023-03389-5
11. Gould, J. (2023). Scientific illustration: Striking the balance between creativity and accuracy. Nature. Erişim: 06.08.2024, https://www.nature.com/articles/d41586-023-03391-x
12. Groenewald, E., Kumar, N., ve Groenewald, C. A. (2024). AI-assisted astronomical data analysis: Unveiling patterns and phenomena in the universe. Naturalista Campano, 28(1), 1916–1925.
13. He, S., Li, Y., Feng, Y., Ho, S., Ravanbakhsh, S., Chen, W., ve Póczos, B. (2019). Learning to predict the cosmological structure formation. Proceedings of the National Academy of Sciences, 116(28), 13825–13832. https://doi.org/10.1073/pnas.1821458116
14. Hutson, M. (2020). From models of galaxies to atoms, simple AI shortcuts speed up simulations by billions of times. Science. Erişim: 09.08.2024, https://www.science.org/content/article/models-galaxies-atoms-simple-ai-shortcuts-speed-simulations-billions-times
15. Jarolim, R. (t.y.). Solar image enhancement with artificial intelligence. Erişim: 22.08.2024, https://est-east.eu/?option=com_content&view=article&id=911&lang=en&Itemid=622
16. Jones, J. (2012). Science is more beautiful than art. The Guardian. Erişim: 15.07.2024, http://www.guardian.co.uk/artanddesign/jonathanjonesblog/2012/sep/19/science-more-beautiful-than-art
17. Kahneman, D. (2011). Thinking, fast and slow. Farrar, Straus and Giroux.
18. Kasim, M. F., Watson-Parris, D., Deaconu, L., Oliver, S., Hatfield, P., Froula, D. H., Gregori, G., Jarvis, M., Khatiwala, S., Korenaga, J., Topp-Mugglestone, J., Viezzer, E., ve Vinko, S. M. (2022). Building high accuracy emulators for scientific simulations with deep neural architecture search. Machine Learning: Science and Technology, 3(1), 015013. https://doi.org/10.1088/2632-2153/ac3ffa
19. Kim, J. J. H., Um, R. S., Lee, J. W. Y. vd. (2024). Generative AI can fabricate advanced scientific visualizations: Ethical implications and strategic mitigation framework. AI Ethics. https://doi.org/10.1007/s43681-024-00439-0
20. Kolijn, E. (2013). Observation and visualization: Reflections on the relationship between science, visual arts, and the evolution of the scientific image. Antonie van Leeuwenhoek, 104, 597–608. https://doi.org/10.1007/s10482-013-9951-z
21. Krenn, M., Pollice, R., Guo, S. Y., Aldeghi, M., Cervera-Lierta, A., Friederich, P., Gomes, G. D., Hase, F., Jinich, A., Nigam, A., Yao, Z., ve Aspuru-Guzik, A. (2022). On scientific understanding with artificial intelligence. Nature Reviews Physics, 4, 761–769. https://doi.org/10.48550/arXiv.2204.01467
22. Li, Y., Ni, Y., Croft, R., Matteo, T., Bird, S., ve Feng, Y. (2021). AI-assisted superresolution cosmological simulations. Proceedings of the National Academy of Sciences, 118(19). https://doi.org/10.1073/pnas.2022038118
23. Medeiros, L., Psaltis, D., Lauer, T., ve Özel, F. (2023). The image of the M87 black hole reconstructed with PRIMO. The Astrophysical Journal Letters, 947(7). https://doi.org/10.3847/2041-8213/acc32d
24. NASA. (2011). NASA's Swift satellite spots black hole devouring a star. Erişim: 02.08.2024, https://svs.gsfc.nasa.gov/10807/
25. NASA. (2018). The Pillars of Creation. Erişim: 11.08.2024, https://www.nasa.gov/image-article/pillars-of-creation/
26. NASA. (2023). Dark matter: Roman Space Telescope. Erişim: 05.08.2024, https://science.nasa.gov/mission/roman-space-telescope/dark-matter/
27. Nature. (2021). Collaborations with artists go beyond communicating the science. Erişim 21.11.2024, https://www.nature.com/articles/d41586-021-00469-2
28. Pandey, Y., Gautam, V., Mowade, V., Telang, S., ve Bhagat, S. (2024). A review on image enhancement using AI. International Journal for Multidisciplinary Research (IJFMR), 6(5). https://doi.org/10.36948/ijfmr.2024.v06i05.29653
29. Pittalwala, I. (2021). Algorithm helps speed up scientific simulations of the universe. Inside UCR. Erişim: 17.09.2024, https://insideucr.ucr.edu/stories/2021/05/11/algorithm-helps-speed-simulation-vast-complex-universes
30. Ravanbakhsh, M., Nabi, M., Sangineto, E., Marcenaro, L., Regazzoni, C., ve Sebe, N. (2017). Abnormal event detection in videos using generative adversarial nets. In Proceedings of the IEEE International Conference on Image Processing (ICIP),1577–1581. IEEE. https://doi.org/10.1109/ICIP.2017.8296547
31. Root-Bernstein, R. (2000). Science in culture. Nature, 407, 134. https://doi.org/10.1038/35025133
32. Samsel, F., Bartram, L. R., ve Bares, A. (2018). Art, affect, and color: Exploring the impact of artistic visualization on affective response. In 2018 IEEE VIS Arts Program (VISAP), 1–9. https://doi.org/10.1109/VISAP45312.2018.9046053
33. Shirasaki, M., Moriwaki, K., Oogi, T., Yoshida, N., Ikeda, S., ve Nishimichi, T. (2021). Noise reduction for weak lensing mass mapping: An application of generative adversarial networks to Subaru Hyper Suprime-Cam first-year data. Monthly Notices of the Royal Astronomical Society, 504(2), 1825–1839. https://doi.org/10.1093/mnras/stab982
34. Sönmez, Ü. (2024). Grafik tasarım tarihinde teknolojik ilerlemeler ve yapay zekânın yaratıcılığa etkileri. Yeni Yüzyıl’da İletişim Çalışmaları, 2(9), 37–43.
35. Sutter, P. (2023). AI is already helping astronomers make incredible discoveries. Here’s how. Space. Erişim: 18.11.2024, https://www.space.com/ai-astronomy-discoveries
36. Team Emb. (2023). AI in astrophysics: Unlocking mysteries of the universe. EMB Global Blog. Erişim: 20.09.2024, https://blog.emb.global/ai-in-astrophysics/
37. Wainwright, S. (2023). AI in astronomy: Unlocking the secrets of the cosmos. Space. Erişim: 17.09.2024, https://www.space.com/ai-in-astronomy
38. Wang, H., Ye, J., Hao, J., Hou, Y., Hou, Y., Wang, Z., Xiao, S., Luo, Y., ve Zeng, W. (2023). Generative adversarial networks for scientific discovery: An overview. Nature, 620, 47–60. https://doi.org/10.1038/s41586-023-06221-2