Generación de contenido educativo integrando modelos de difusión y estilos de aprendizaje
dc.contributor.advisor | Bolaños Martínez, Freddy | |
dc.contributor.advisor | Arango Zuluaga, Eliana Isabel | |
dc.contributor.author | Arenas Tamayo, Luis Miguel | |
dc.contributor.googlescholar | LhTcFwQAAAAJ | |
dc.contributor.orcid | Arenas Tamayo, Luis Miguel [0009-0003-7214-6025] | |
dc.contributor.orcid | Bolaños Martínez, Freddy [0000-0002-3123-5481] | |
dc.contributor.orcid | Arango Zuluaga, Eliana Isabel [0000-0002-0947-3694] | |
dc.contributor.researchgate | Luis-Arenas-Tamayo-2 | |
dc.date.accessioned | 2025-08-29T02:32:59Z | |
dc.date.available | 2025-08-29T02:32:59Z | |
dc.date.issued | 2025-08-28 | |
dc.description | Ilustraciones | spa |
dc.description.abstract | Esta tesis surge a partir del reciente auge y las crecientes capacidades de la inteligencia artificial generativa, en particular de los modelos de difusión generadores de imágenes, como herramienta para complementar contenidos pedagógicos considerando los estilos de aprendizaje como insumo. Para ello, se desarrolló una herramienta web utilizando el framework Django, servicios nube de AWS y ComfyUI, estructurada en dos etapas, una etapa de identificación, que adapta un cuestionario de estilos de aprendizaje mediante imágenes generadas, y una etapa de inferencia, en la cual se generan imágenes a demanda por parte de los participantes para la evaluación de cuatro modelos de difusión. En ambas etapas se aplicaron preguntas de percepción utilizando la escala Likert, validadas mediante el Coeficiente de Cronbach de 0.836 y 0.721 respectivamente. Los resultados evidenciaron que, en la etapa de identificación, el 60% de las respuestas fueron clasificadas como altas, el 23.78% como bajas y el 16.22% como medias. En la etapa de inferencia, se observó una leve disminución en la proporción de respuestas altas 47.83%, mientras que las respuestas medias y bajas aumentaron a 30.98% y 21.20%, respectivamente. Adicionalmente, se incorporó el análisis de los estilos de aprendizaje preferentes de los participantes, revelando patrones significativos en la percepción según cada estilo. En general, los participantes con estilos VIS (visual), REF (reflexivo), SEN (sensitivo) y ACT (activo) tendieron a otorgar calificaciones altas en ambas etapas. Los modelos de difusión fueron evaluados también mediante el CLIP Score, evidenciando preferencias diferenciadas según el estilo de aprendizaje. El modelo DALLE 3 fue el más preferido en términos generales, mientras que los participantes con estilo visual mostraron una inclinación positiva hacia Stable Diffusion XL y Stable Diffusion 3.5. Estos hallazgos indican que la percepción de utilidad o calidad de las imágenes generadas varía de acuerdo con el estilo de aprendizaje predominante, lo cual tiene implicaciones relevantes para la personalización de herramientas educativas basadas en inteligencia artificial generativa. (Tomado de la fuente) | spa |
dc.description.abstract | This thesis arises from the recent surge and growing capabilities of generative artificial intelligence particularly image generating diffusion models as a tool to supplement pedagogical content by taking different learning styles into account. For this purpose, a web based tool was developed using the Django framework, AWS cloud services and ComfyUI, structured in two stages, an identification stage, which adapts a learning style questionnaire by means of generated images, and an inference stage, in which images are generated on demand by participants for the evaluation of four diffusion models. In both stages, perception was measured through Likert scale questions, validated by Cronbach’s alpha coefficients of 0.836 and 0.721, respectively. The results showed that, in the identification stage, 60% of the responses were classified as high, 23.78% as low, and 16.22% as medium. In the inference stage, there was a slight decrease in the proportion of high responses 47.83%, while medium and low responses rose to 30.98% and 21.20%, respectively. Additionally, an analysis of participants’ preferred learning styles was included, revealing significant perceptual patterns based on each style. Overall, participants with VIS (visual), REF (reflective), SEN (sensing), and ACT (active) styles tended to provide high ratings in both stages. The diffusion models were also evaluated through the CLIP Score, showing differentiated preferences depending on the learning style. The DALLE 3 model was the most widely preferred in general, whereas participants with a visual style showed a positive inclination toward Stable Diffusion XL and Stable Diffusion 3.5. These findings indicate that perceptions of the usefulness or quality of the generated images vary according to the predominant learning style, which has important implications for personalizing educational tools based on generative artificial intelligence. | eng |
dc.description.curriculararea | Ingeniería De Sistemas E Informática.Sede Medellín | |
dc.description.degreelevel | Maestría | |
dc.description.degreename | Magíster en Ingeniería - Analítica | |
dc.description.researcharea | Inteligencia Artificial en Educación | |
dc.description.technicalinfo | https://www.lmarenast.com/ | spa |
dc.format.extent | 92 páginas | |
dc.format.mimetype | application/pdf | |
dc.identifier.instname | Universidad Nacional de Colombia | spa |
dc.identifier.reponame | Repositorio Institucional Universidad Nacional de Colombia | spa |
dc.identifier.repourl | https://repositorio.unal.edu.co/ | spa |
dc.identifier.uri | https://repositorio.unal.edu.co/handle/unal/88505 | |
dc.language.iso | spa | |
dc.publisher | Universidad Nacional de Colombia | |
dc.publisher.branch | Universidad Nacional de Colombia - Sede Medellín | |
dc.publisher.faculty | Facultad de Minas | |
dc.publisher.place | Medellín, Colombia | |
dc.publisher.program | Medellín - Minas - Maestría en Ingeniería - Analítica | |
dc.relation.indexed | LaReferencia | |
dc.relation.references | Bedi, K.: AI Comics as Art: Scientific Analysis of the Multimedia Content of AI Comics in Education, 2023.– ISBN 9789532331042, p. 750–753 | |
dc.relation.references | Borah, Asha R. ; Nischith, T. N. ; Gupta, Saksham: Improved Learning Ba sed on GenAI, Institute of Electrical and Electronics Engineers Inc., 2024.– ISBN 9798350327533, p. 1527–1532 | |
dc.relation.references | Brown, Tom B. ; Mann, Benjamin: Language Models are Few-Shot Learners. (2020), 5 | |
dc.relation.references | Choi, D.H.: LYRA: An Interactive and Interactive Storyteller, 2019.– ISBN 9781728116631, p. 148–153 | |
dc.relation.references | Civitai. ComfyUI. https://civitai.com/. 2022 | |
dc.relation.references | CivitAI. Detail Tweaker XL. https://civitai.com/models/122359/detail-tweaker xl?modelVersionId=135867. 2023 | |
dc.relation.references | CivitAI. Stable Diffusion Acceleration Module- LoRA. https://civitai.com/models/195519/lcm-lora-weights-stable-diffusion-acceleration module?modelVersionId=424720. 2024 | |
dc.relation.references | ComfyUI. ComfyUI. https://github.com/comfyanonymous/ComfyUI. 2023 | |
dc.relation.references | Conruyt, N. ; S´ ebastien, V. ; S´ ebastien, O. ; Grosser, D. ; S´ ebastien, D.: From knowledge transmission to sign sharing: Semiotic web as a new paradigm for teaching and learning in the future internet. Vol. 422. 2014.– 170–188 p..– ISBN 9783642548963 | |
dc.relation.references | Cosentino, G. ; Morra, D. ; Gelsomini, M. ; Matera, M. ; Mores, M.: COBO: A Card-Based Toolkit for Co-Designing Smart Outdoor Experiences with People with Intellectual Disability. Vol. 12932 LNCS. 2021.– 149–169 p..– ISBN 9783030856229 | |
dc.relation.references | Come, Bruno ; Devanne, Maxime ; Weber, Jonathan ; Forestier, Germain: A Comparative Study of CNNs and Vision-Language Models for Chart Image Classifica tion, Science and Technology Publications, Lda, 2025.– ISSN 2184433X, p. 816–827 | |
dc.relation.references | Dehouche, N. ; Dehouche, K.: What’s in a text-to-image prompt? The potential of stable diffusion in visual arts education. En: Heliyon 9 (2023) | |
dc.relation.references | El Aissaoui, Ouafae ; El Alami El Madani, Yasser: Mining Learners’ Behaviors: An Approach Based on Educational Data Mining Techniques. En: Embedded Systems and Artificial Intelligence. Singapore : Springer Singapore, 2020.– ISBN 978–981–15 0947–6, p. 655–670 | |
dc.relation.references | Elman, Jeffrey L.: Finding Structure in Time. En: Cognitive Science 14 (1990), Nr. 2, p. 179–211 | |
dc.relation.references | Fatahi, Somayeh ; Moradi, Hadi ; Kashani-Vahid, Leila: A survey of personality and learning styles models applied in virtual environments with emphasis on e-learning environments. En: Artificial Intelligence Review 46 (2016), 10, p. 413–429.– ISSN 15737462 | |
dc.relation.references | Felder, Richard M. Learning and Teaching Styles in Engineering Education. 1988 | |
dc.relation.references | Fleming, Neil D. ; Mills, Colleen. Not Another Inventory, Rather a Catalyst for Reflection. 1992 | |
dc.relation.references | Fletscher, Luis ; Mendoza-Cardenas, Carlos ; Mercado, Jhon ; Montoya, Lina ; Valderrama, Angela ; Ramirez, Doris ; Oviedo, Diana: Enhancing Differentiated Learning in Virtual Courses Through AI-Driven Recommender Systems. En: DEMOcon 2024- 2024 IEEE Digital Education and MOOCS Conference, Proceedings (2024). ISBN 9798350353143 | |
dc.relation.references | Gal, Rinon ; Alaluf, Yuval ; Atzmon, Yuval ; Patashnik, Or ; Bermano, Amit H. ; Chechik, Gal ; Cohen-Or, Daniel. An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion | |
dc.relation.references | Gardner, H.E.: Frames of Mind: The Theory of Multiple Intelligences. Basic Books, 2011.– ISBN 9780465024346 | |
dc.relation.references | Gomede, Everton ; de Barros, Rodolfo M. ; de Souza Mendes, Leonardo: Use of deep multi-target prediction to identify learning styles. En: Applied Sciences (Switzer land) 10 (2020), 3.– ISSN 20763417 | |
dc.relation.references | Goodfellow, Ian J. ; Pouget-Abadie, Jean ; Mirza, Mehdi ; Xu, Bing ; Warde Farley, David ; Ozair, Sherjil ; Courville, Aaron ; Bengio, Yoshua: Generative Adversarial Networks. (2014), 6 | |
dc.relation.references | Gu, R. ; Xi, Z. ; Lin, B. ; Ji, Y.: Teacher-guided Autonomous Learning Enabled by Artificial Intelligence Empowered Remote Experiment Platform, 2022.– ISBN 9781665444347, p. 1089–1094 | |
dc.relation.references | Guran, A.-M. ; Cojocar, G.-S. ; Turian, A.: Towards Preschoolers’ Automatic Satisfaction Assessment. An Experience Report, 2020.– ISBN 9781728173764, p. 141 146 | |
dc.relation.references | Hassan, O. A. Learning theories and assessment methodologies- an engineering edu cational perspective. 8 2011 | |
dc.relation.references | Hemment, D. ; Currie, M. ; Bennett, S.J. ; Elwes, J. ; Ridler, A. ; Sinders, C. ; Vidmar, M. ; Hill, R. ; Warner, H.: AI in the Public Eye: Investigating Public AI Literacy Through AI Art, 2023.– ISBN 9781450372527, p. 931–942 | |
dc.relation.references | Hu, Edward ; Shen, Yelong ; Wallis, Phillip ; Allen-Zhu, Zeyuan ; Li, Yuanzhi ; Wang, Shean ; Wang, Lu ; Chen, Weizhu. LORA: LOW-RANK ADAPTATION OF LARGE LAN-GUAGE MODELS. 2021 | |
dc.relation.references | Iqbal, M.M. ; Saleem, Y. ; Naseer, K. ; Kim, M.: Multimedia based student-teacher smart interaction framework using multi-agents in eLearning. En: Multimedia Tools and Applications 77 (2018), p. 5003–5026 | |
dc.relation.references | Jannach, D. ; Leopold, K. ; Hellwagner, H.: An extensible framework for knowledge-based multimedia adaptation, 2004, p. 144–153 | |
dc.relation.references | Josceanu, A.M. ; Onofrei, R. ; Isopescu, R. ; Plesu, V. ; Zelch, O.: Technology pedagogy relationship in developing e-learning systems for chemical engineering educa tion, 2006.– ISBN 9788086059457 | |
dc.relation.references | KandooAI. Juggernaut XL. modelVersionId=351306. 2024 https://civitai.com/models/133005? | |
dc.relation.references | Karras, Tero ; Laine, Samuli ; Aila, Timo: A Style-Based Generator Architecture for Generative Adversarial Networks. (2018), 12 | |
dc.relation.references | Kolb, Alice Y. ; Kolb, David A. Learning Styles and Learning Spaces: Enhancing Experiential Learning in Higher Education Experience-Based Learning Systems. 2005 | |
dc.relation.references | Kolekar, Sucheta V. ; Pai, Radhika M. ; Pai, M. M. M.: Adaptive User Interface for Moodle based E-learning System using Learning Styles, Elsevier B.V., 2018.– ISSN 18770509, p. 606–615 | |
dc.relation.references | Leaver, Betty L. ; Ehrman, Madeline ; Shekhtman, Boris: Learning styles and learning strategies. Cambridge University Press, 12 2009.– 65–91 p | |
dc.relation.references | Leung, K.-K. ; Huang, W. ; Ho, J. ; Gu, J. ; Meng, Q. ; Zhang, R. ; Leung, K.-H. ; Kwan, T.H.: An Inquiry-based Genomics Intelligent Tutoring System, 2022.– ISBN 9781665491174, p. 25–30 | |
dc.relation.references | Li, M. ; Amoroso, N.: An Early Look at Applications for Artificial Intelligence Vi sualization Software in Landscape Architecture. En: Journal of Digital Landscape Ar chitecture 2023 (2023), p. 543–553 | |
dc.relation.references | Lykon. DreamShaper XL. https://civitai.com/models/112902?modelVersionId= 351306. 2024 | |
dc.relation.references | L´ opez-S´ anchez, D. ; Arrieta, A.G. ; Corchado, J.M.: Deep neural networks and transfer learning applied to multimedia web mining. Vol. 620. 2018.– 124–131 p.. ISBN 9783319624099 | |
dc.relation.references | Mejia-Escobar, C. ; Cazorla, M. ; Martinez-Martin, E.: Improving Facial Ex pression Recognition Through Data Preparation and Merging. En: IEEE Access 11 (2023), p. 71339–71360 | |
dc.relation.references | Miao, Fengchun. Inteligencia artificial y educaci´on: gu´ıa para las personas a cargo de formular pol´ıticas- UNESCO Biblioteca Digital. 2021 | |
dc.relation.references | Miao, Fengchun. Guidance for generative AI in education and research- UNESCO Biblioteca Digital. 2023 | |
dc.relation.references | Murugesan, S. ; Cherukuri, A.K.: The Rise of Generative Artificial Intelligence and Its Impact on Education: The Promises and Perils. En: Computer 56 (2023), p. 116–121 | |
dc.relation.references | Mylonas, P.: A fuzzy contextual approach towards intelligent educational content adaptation, 2007.– ISBN 9780769530406, p. 257–261 | |
dc.relation.references | OpenAI. GPT-4o. ://openai.com/index/hello-gpt-4o/. 2024 | |
dc.relation.references | OpenAI: GPT-4o System Card. (2024), 10 | |
dc.relation.references | OpenAI. Video generation models as world simulators. https://openai.com/index/video-generation-models-as-world-simulators/. 2024 | |
dc.relation.references | OpenAI. 4o Image Generation. https://openai.com/index/introducing-4o-image generation/. 2025 | |
dc.relation.references | OpenAI. Logprobs. ://platform.openai.com/docs/api-reference/chat/create/. NA | |
dc.relation.references | Openai, Alec R. ; Openai, Karthik N. ; Openai, Tim S. ; Openai, Ilya S. Improving Language Understanding by Generative Pre-Training | |
dc.relation.references | OpenTRIVIA. OpenTRIVIA. https://opentdb.com/. NA | |
dc.relation.references | Pablo, Rosser. Visualizing Learning: Evolution of Styles and Pedagogical Strategies in Teacher Education. 2024 | |
dc.relation.references | Podell, Dustin ; English, Zion ; Lacey, Kyle ; Blattmann, Andreas ; Dockhorn, Tim ; M¨ uller, Jonas ; Penna, Joe ; Rombach, Robin: SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis. (2023), 7 | |
dc.relation.references | Radford, Alec ; Kim, Jong W. ; Hallacy, Chris ; Ramesh, Aditya ; Goh, Gabriel ; Agarwal, Sandhini ; Sastry, Girish ; Askell, Amanda ; Mishkin, Pamela ; Clark, Jack ; Krueger, Gretchen ; Sutskever, Ilya: Learning Transferable Visual Models From Natural Language Supervision. (2021), 2 | |
dc.relation.references | Ramesh, Aditya ; Dhariwal, Prafulla ; Nichol, Alex ; Chu, Casey ; Chen, Mark: Hierarchical Text-Conditional Image Generation with CLIP Latents. (2022), 4 | |
dc.relation.references | Rojas-Palacio, Cinthia V. ; Arango-Zuluaga, Eliana I. ; Botero-Castro, H´ector A.: Teaching control theory: a selection of methodology based on learning styles. En: DYNA (Colombia) 89 (2022), 7, p. 9–17.– ISSN 00127353 | |
dc.relation.references | Rombach, Robin ; Blattmann, Andreas ; Lorenz, Dominik ; Esser, Patrick ; Om mer, Bj¨orn: High-Resolution Image Synthesis with Latent Diffusion Models. (2021), 12 | |
dc.relation.references | Rudolph, J. ; Tan, S. ; Tan, S.: ChatGPT: Bullshit spewer or the end of traditional assessments in higher education? En: Journal of Applied Learning and Teaching 6 (2023), p. 342–363 | |
dc.relation.references | Ruiz, Nataniel ; Li, Yuanzhen ; Jampani, Varun ; Pritch, Yael ; Rubinstein, Mi chael ; Aberman, Kfir: DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation. (2022), 8 | |
dc.relation.references | Rumelhart, David E. ; Hinton, Geoffrey E. ; Williams, Ronald J.: Learning internal representations by error propagation, 1986 | |
dc.relation.references | Runpod. Runpod.io. https://www.runpod.io/. NA | |
dc.relation.references | Schuhmann, Christoph ; Beaumont, Romain ; Vencu, Richard ; Gordon, Cade ; Wightman, Ross ; Cherti, Mehdi ; Coombes, Theo ; Katta, Aarush ; Mullis, Clayton ; Wortsman, Mitchell ; Schramowski, Patrick ; Kundurthy, Srivatsa ; Crowson, Katherine ; Schmidt, Ludwig ; Kaczmarczyk, Robert ; Jitsev, Jenia: LAION-5B: An open large-scale dataset for training next generation image-text models. (2022), 10 | |
dc.relation.references | Serrano, J M. El Constructivismo hoy: enfoques constructivistas en educaci´on. 2011 | |
dc.relation.references | Sohl-Dickstein, Jascha ; Weiss, Eric A. ; Maheswaranathan, Niru ; Ganguli, Surya ; Edu, Sganguli@stanford. Deep Unsupervised Learning using Nonequilibrium Thermodynamics | |
dc.relation.references | Sorin, V. ; Barash, Y. ; Konen, E. ; Klang, E.: Creating Artificial Images for Radiology Applications Using Generative Adversarial Networks (GANs)– A Systematic Review. En: Academic Radiology 27 (2020), p. 1175–1185 | |
dc.relation.references | stability.ai. Stable Diffusion XL. stable-diffusion-sdxl-1-announcement. 2023 | |
dc.relation.references | Touretzky, D. ; Gardner-McCune, C. ; Seehorn, D.: Machine Learning and the Five Big Ideas in AI. En: International Journal of Artificial Intelligence in Education (2022) | |
dc.relation.references | Treur, J.: Multidisciplinary Education Computational Modeling as the Core of a Mul tidisciplinary Curriculum. 2016.– 473–484 p | |
dc.relation.references | Vartiainen, H. ; Tedre, M.: Using artificial intelligence in craft education: crafting with text-to-image generative models. En: Digital Creativity 34 (2023), p. 1–21 | |
dc.relation.references | Vaswani, Ashish ; Shazeer, Noam ; Parmar, Niki ; Uszkoreit, Jakob ; Jones, Llion ; Gomez, Aidan N. ; Kaiser, Lukasz ; Polosukhin, Illia: Attention Is All You Need. (2017), 6 | |
dc.relation.references | Walczak, K. ; Wiza, W. ; Wojciechowski, R. ; W´ ojtowicz, A. ; Rumi´ nski, D. ; Cellary, W.: Building augmented reality presentations with web 2.0 tools. Vol. 369. 2015.– 595–605 p..– ISBN 9783319197128 | |
dc.relation.references | Wang, D. ; Han, J.: EXPLORING THE IMPACT OF GENERATIVE STIMULI ON THE CREATIVITY OF DESIGNERS IN COMBINATIONAL DESIGN, 2023, p. 1805–1814 | |
dc.relation.references | Yang, M. ; Jiang, P. ; Zang, T. ; Liu, Y.: Data-driven intelligent computational de sign for products: method, techniques, and applications. En: Journal of Computational Design and Engineering 10 (2023), p. 1561–1578 | |
dc.relation.references | Zhang, Chenshuang ; Zhang, Chaoning ; Zhang, Mengchun ; Kweon, In S.: Text to-image Diffusion Models in Generative AI: A Survey. (2023), 3 | |
dc.rights.accessrights | info:eu-repo/semantics/openAccess | |
dc.rights.license | Atribución-NoComercial 4.0 Internacional | |
dc.rights.uri | http://creativecommons.org/licenses/by-nc/4.0/ | |
dc.subject.ddc | 000 - Ciencias de la computación, información y obras generales::004 - Procesamiento de datos Ciencia de los computadores | |
dc.subject.ddc | 000 - Ciencias de la computación, información y obras generales::005 - Programación, programas, datos de computación | |
dc.subject.ddc | 370 - Educación | |
dc.subject.lemb | Inteligencia artificial | |
dc.subject.lemb | Aprendizaje automático (Inteligencia artificial) | |
dc.subject.lemb | Métodos de enseñanza | |
dc.subject.proposal | Inteligencia Artificial | spa |
dc.subject.proposal | Inteligencia Artificial Generativa | spa |
dc.subject.proposal | Modelos de Difusión | spa |
dc.subject.proposal | Estilos de Aprendizaje | fra |
dc.subject.proposal | Artificial Intelligence | eng |
dc.subject.proposal | Generative Artificial Intelligence | eng |
dc.subject.proposal | Diffusion Models | eng |
dc.subject.proposal | Learning Styles | eng |
dc.subject.proposal | DALLE | |
dc.subject.proposal | Stable Diffusion | |
dc.subject.proposal | ComfyUI | |
dc.title | Generación de contenido educativo integrando modelos de difusión y estilos de aprendizaje | spa |
dc.title.translated | Educational content generation through the integration of diffusion models and learning styles | eng |
dc.type | Trabajo de grado - Maestría | |
dc.type.coar | http://purl.org/coar/resource_type/c_bdcc | |
dc.type.coarversion | http://purl.org/coar/version/c_ab4af688f83e57aa | |
dc.type.content | Text | |
dc.type.driver | info:eu-repo/semantics/masterThesis | |
dc.type.redcol | http://purl.org/redcol/resource_type/TM | |
dc.type.version | info:eu-repo/semantics/acceptedVersion | |
dcterms.audience.professionaldevelopment | Estudiantes | |
dcterms.audience.professionaldevelopment | Investigadores | |
dcterms.audience.professionaldevelopment | Maestros | |
dcterms.audience.professionaldevelopment | Público general | |
oaire.accessrights | http://purl.org/coar/access_right/c_abf2 |