From Web Data to DIY Ideas: Material-Based Project Generation Systems
DOI:
https://doi.org/10.54097/g450x518Keywords:
Upcycling; Natural Language Processing; Material-based Generation; Sustainability; DIY Systems; Web Data Extraction; Creative Design Automation; Human-in-the-loop Learning.Abstract
This article investigates and explores the prototype of a material based project generation system, aiming to transform online DIY content into structured design knowledge for sustainable reuse. Based on an unfinished system developed by the author in his early years, this work reviews the methods of collecting unstructured data from the Internet, extracts key information through lightweight natural language processing, and normalizes it into a unified model suitable for automatic generation. Discussed baseline strategies for retrieving and editing project generation, as well as evaluation metrics that consider constraint satisfaction, feasibility, and user experience. The paper also investigates practical applications in personal creativity support, maker education, and community recycling, supported by initial human in the loop implementation and interactive segmentation code that links material understanding with basic robot operations. The challenges of multimodal alignment, richer constraint modeling, and robust dataset construction have been identified, highlighting opportunities for future integration with digital twins and automated assembly. Although the original system is still incomplete, the study concludes that a practical and impactful pipeline is achievable, demonstrating how artificial intelligence can help users recycle materials more effectively and contribute to sustainable, resource-saving designs.
References
[1] Ferraro, A., Karatzas, K., & Jones, M. (2021). Text mining for sustainability: A systematic review. Journal of Cleaner Production, 288, 125583.
[2] Etzioni, O., Banko, M., Soderland, S., & Weld, D. S. (2008). Open information extraction from the web. Communications of the ACM, 51(12), 68–74.
[3] Chen, D., & Manning, C. D. (2014). A fast and accurate dependency parser using neural networks. Proceedings of EMNLP 2014, 740–750.
[4] Dong, L., Wang, S., & Zhang, J. (2021). Constraint-based text generation: A survey. ACM Computing Surveys, 54(5), 1–34.
[5] Sung, K., Cooper, T., & Kettley, S. (2014). Individual upcycling practice: Exploring the possible determinants of upcycling based on a literature review. Sustainable Development, 22(2), 123–135.
[6] Liu, Y., Wan, X., & He, J. (2022). Multimodal knowledge extraction and generation: Recent advances and challenges. IEEE Transactions on Neural Networks and Learning Systems, 33(10), 5034–5051.
[7] Gao, X, Y., (2022). A Hybrid Search Model for Constrained Optimization. Wiley Journal.
[8] Sun, Y., Li, Z., & Zhang, H. (2020). Human-in-the-loop machine learning: A survey. ACM Computing Surveys, 53(2), 1–31.
[9] Buechley, L., & Eisenberg, M. (2008). The LilyPad Arduino: Toward wearable engineering for everyone. IEEE Pervasive Computing, 7(2), 12–15.
[10] Sun, Y., Li, Z., & Zhang, H. (2020). Human-in-the-loop machine learning: A survey. ACM Computing Surveys, 53(2), 1–31. Turner, C. J., Emmanouilidis, C., & Tiwari, A. (2021). Digital twins for sustainable smart manufacturing: A review. Advanced Engineering Informatics, 50, 101431.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
