Application of collaborative filtering algorithm with improved trust degree in clothing personalization and customization designs

Purpose This study aims to achieve efficient and accurate clothing personalization recommendation. Design/methodology/approach The study designed an improved clothing collaborative filtering (CF) recommendation model, which integrates user-item rating trust and clothing preference trust to improve the traditional CF recommendation algorithm. Findings Verified by different types of datasets, the trust-based CF algorithm designed by the study takes smaller values in the average absolute value error and root mean square error, and introducing the trust factor significantly improves the recommendation accuracy compared with other models. Research limitations/implications The recommendation algorithm designed in the study is based on text information to extract interest features. Practical implications Future research work can pay more attention to clothing pattern, color and style features. Originality/value The clothing personalized recommendation model designed by the study improves in recommendation accuracy, accelerates the digital construction process of the clothing industry and is conducive to the development of the clothing personalized customization system.