A Machine Learning-Based Interactive Personalized Pattern Design Knowledge Base For Hanfu  
Author Zhujun Wang


Co-Author(s) Yingmei Xing; Dagong Chen; Xuyuan Tao; Xianyi Zeng; Pascal Bruniaux


Abstract This study put forth to establish a new knowledge base (KB) for personalized Hanfu design toward e-mass customization (e-MC) using a support vector machine-based machine learning approach. The proposed KB is composed of a series of basic databases and relational models. The databases store the fundamental data related to the movements of the controlling points (CPs) and the corresponding structure lines (SLs) in the patterns. The relational models are created by SVM to express the quantified relations between the CPs’ movements and the SLs’ length variations. Based on these databases and models, the function of personalized Hanfu style can be achieved by adjusting the garment patterns adaptively. The experimental results indicate that our approach provides a feasible and valuable solution for enhancing the implementation of garment e-MC for Hanfu.


Keywords Hanfu, Garment pattern design, Knowledge base, Machine learning, E-mass customization, Industry 4.0
    Article #:  DSBFI23-139
Proceedings of 2nd ISSAT International Conference on Data Science in Business, Finance and Industry
January 8-10, 2023 - Da Nang, Vietnam