Method for predicting personalized preference of user through eye movement data

Abstract

The invention discloses a method for predicting personalized preference of a user through eye movement data. The method comprises the steps that (1) design elements are extracted, wherein a product picture is selected, and a semiotic and morphological analysis method is adopted to perform structure disintegration and design element extraction on a pattern of the picture; (2) eye movement indicators are screened, wherein a picture type is used as an influence factor, variance homogeneity testing is performed on user eye movement indicators with browsing tasks, single-factor variance analysis is further performed, and the eye movement indicators are screened; and (3) a relation model of the design elements and the eye movement indicators is established, wherein a quantification-I theory is adopted to encode the design elements of the pattern, the qualitative variable pattern is converted into a quantitative variable pattern, a multiple regression model of the eye movement indicators and the design elements is established, and influences of different design elements on the personalized preference of the user are obtained. The method has the advantages that the relation model of the eye movement data and the design elements can be established, and the influences of different design elements on the personalized preference of the user can be reflected.

Description

Technical field [0001] The invention belongs to the technical field of product design, and specifically relates to a method for predicting user's personalized preference by using eye movement data. Background technique [0002] With the rise of user experience design, the research on user emotions has become a hot topic. The research on user emotions mainly includes psychological response measurement technology based on psychology and physiological response measurement technology based on psychology. Psychology-based psychological response measurement techniques mainly include semantic difference method, semantic method, spoken language analysis method, semantic comparison method, PANAS scale method [1] , PAD emotional measurement [2] , Product pleasure measurement questionnaire, PreMo measurement method [3] , Emocards (emotional cards) measurement method [4] , SAM (selfassessment manikin) self-assessment model, etc. [0003] In the existing technology, Liu Ye [5] Et al. analyzed t...

AI Technical Summary

Problems solved by technology

In short, the existing emotion research is mainly reflected in the emotional or rational emotion of the product as a whole, which cannot reflect the relationship between user emotion and product design elements, nor can it reflect the influence of design elements on user emotion, so that different conclusions cannot be drawn. Influence of Design Elements on User Personalization Preference

Images