User-depth Customized Men’s Shirt Design Framework Based on BI-LSTM

摘要

The current user customization method of clothing has a low degree of consumer participation, which results in the lack of effective data for garment manufacturers and poorly-recognized personalized products. The personalized shirt framework based on deep learning uses the long-short-term memory neural network (LSTM) to process the textual information of the characters described by the consumers themselves, providing the garment factory with more consistent design data, replacing the traditional clothing component combination design method. This method replaces the traditional clothing component combination design method, can effectively improve the accuracy of data in personalized production, the degree of matching between user participation and recommendation system, thus improving the predictability of garment factory for personalized production and the production efficiency of factory design department.