An interactive knowledge-based recommender system for fashion product design in the big data environment

Abstract

In this paper, we originally propose an interactive, knowledge-based design recommender system (IKDRS) for relevant personalised fashion product design schemes with their virtual demonstrations for a specific consumer. This system enables the iterative interaction between virtual product demonstration and the designer’s professional knowledge and perception in order to find the best existing design solution, i.e. combination of basic garment elements. To develop this system, the anthropometric data and designer’s perception of body shapes are first acquired by using a 3D body scanning system and a sensory evaluation procedure. Next, an instrumental experiment is realised for measuring the technical parameters of fabrics and five sensory experiments are carried out in order to acquire design knowledge. The acquired data are used to classify body shapes and model the relations between human bodies, fashion themes and design factors by using fuzzy techniques. From these models, we set up an ontology-based design knowledge base, including key data and relevant relation models. This knowledge base can be updated in a big data environment by progressively learning from new design cases. On this basis, we propose an interactive, personalised design recommender system. This system works through a newly proposed design process: consumers’ emotional requirement identification – design schemes generation – recommender – 3D virtual prototype display and evaluation – design factors adjustment. This process can be performed repeatedly until the designer is satisfied. The proposed system has been validated through a number of successful real design cases.