Welcome to our project on Advanced Large Language Models & Visualization Tools for Data Analytics Learning! Our mission is to revolutionize the way non-computational professionals and students learn data analytics. We do this by harnessing the power of cutting-edge AI technologies like GPT-4 and the advanced visualization capabilities of tools like LIDA. Our research, supported by extensive case studies and published in specialized conferences and journals, shows how these tools can drastically improve both the speed and quality of data-related disciplines education. Join us on this exciting journey to make data analytics and data science more accessible, efficient, and engaging for everyone!
This project, based on two comprehensive studies, explores the use of advanced Large Language Models (LLMs) and visualization tools to enhance data analytics learning for students and professionals from non-computational backgrounds. The methodologies and outcomes described herein underscore the significant benefits of integrating cutting-edge AI technologies such as based on Generative AI (GenAI) into educational practices to foster a deeper understanding and more efficient execution of data-related projects.
- Promote a comprehensive understanding of data-based project pipelines.
- Enhance programming and other computational thinking-related skills through interactive AI assistance.
- Enable wider adoption of GenAI tools in educational contexts.
- Improve the efficiency and effectiveness of data-related project development.
The project unfolds in several key stages, as outlined in the case studies:
Students and professionals from non-computational backgrounds. Specifically, 88% of participants came from fields such as finance, business, social sciences, and others, while the remaining 12% were from engineering disciplines including sustainable engineering, chemical engineering, biomedical engineering, and industrial engineering.
- Traditional Approach: Participants first completed a data analytics project using standard Python packages (e.g., scikit-learn, pandas, seaborn) in Google Colab.
- ChatGPT Approach: Participants then repeated the project with conventional ChatGPT assistance, using the tool mainly for generating code snippets.
- LIDA + GPT Approach: Finally, participants completed the project using LIDA integrated with the GPT-4 API, enabling automated data summarization, exploration, and advanced visualizations in response to any prompt originating from the project’s source code itself.
to be updated
- Journal article published at Frontiers in Education open access here
- Data from Case Study used to obtain results for our journal article available here
- Conference extended abstract published at proceedings of IACEE 2024 free access here
- Conference presentation access here
When referencing this project, please use the following citation formats:
Journal Article:
Valverde-Rebaza, J., González, A., Navarro-Hinojosa, O., & Noguez, J. (2024). Advanced large language models and visualization tools for data analytics learning. Front. Educ. 9:1418006. DOI: 10.3389/feduc.2024.1418006.
@article{valverde:frontiers:24,
title={Advanced large language models and visualization tools for data analytics learning},
author={Valverde-Rebaza, J. and González, A. and Navarro-Hinojosa, O. and Noguez, J.},
journal={Front. Educ.},
volume={9},
pages={1418006},
year={2024},
doi={10.3389/feduc.2024.1418006}
}Conference Extended-Abstract:
Valverde-Rebaza, J., González, A., Navarro-Hinojosa, O., & Noguez, J. (2024). Empowering Data Analytics Learning: Leveraging Advanced Large Language Models and Visualization Tools. Proceedings of the 19th World Conference on Continuing Engineering Education, IACEE 2024, pp. 47-49. ISBN: 978-1-7327114-3-3.
@inproceedings{Valverde:iacee:24b,
author = {Valverde-Rebaza, J. and González, A. and Navarro-Hinojosa, O. and Noguez, J.},
title = {{Empowering Data Analytics Learning: Leveraging Advanced Large Language Models and Visualization Tools}},
booktitle = {Proceedings of The 19th World Conference on Continuing Engineering Education},
series = {IACEE 2024},
pages = {47--49},
isbn = {978-1-7327114-3-3},
publisher = {IACEE},
year = {2024}
}