Reaearch On E-commerce Website Emotional Design Based On Multimode Emotion Measurement

With the strengthening of economic globalization and the rapid development of internet, e-commerce has achieved rapid development. To enterprise, the interfaces of e-commerce websites are not only windows where the enterprises show their image, but also an important means to display the goods and help users to complete the purchase. Because of serious homogenization of competition among e-commerce websites, the users' emotional experiences have been considered to be one of the key factors in the success of an e-commerce website. To make the website design meet the users' emotional needs, it is needed to accurately obtain the emotional needs of the users and turn them into design elements, so as to guide the website design. However, domestic researches about website emotional design are only at the beginning. Most of the existing researches focused on the introduction of the concept and qualitative analysis. A system method of emotional design about websites has not formed. This study aims to establish the theory and method of the e-commerce website emotional design based on multimode emotion measurement, which could provide technology support for emotional website interface design. In this paper, the main work and conclusons includes:Firstly, a multimodal measurement method is proposed to measure users' emotional experiences of e-commerce websites. The relationship between the different modes of emotional experiences is analyzed. The users' emotional experiences are measured in the task of browsing and shopping. The results indicate that differences in users' emotional experiences while browsing and shopping on various e-commerce websites are primarily embodied in the indicators of subjective emotional experiences and eye movements. For the browsing task, the eye movements indicators with significant differences in different websites are fixation count, saccade count and average saccade velocity. As for the shopping task, eye movements indicators with significant differences in different websites are average pupil size, fixation count, blink count and saccade count. The users' galvanic skin response and skin temperature will increase when they browse compared with when they at rest. The users' galvanic skin response, skin temperature and respiration rate will increase when they shop compared with when they at rest. However, no significant difference in physiological responses is observed when users browse or shop on different websites. The three aspects of emotion experiences interact with each other. The proposed multimodal emotion measurement method can be used to better understand users' emotional experiences of e-commerce websites.Secondly, a method is proposed to predict users' behavior intentions to e-commerce websites. One way analysis of variance is used to extract the indicators of emotional experiences that would affect users' behavior intentions. The relationship models of emotional experiences and behavioral intentions are established based on support vector regression.The results show that, for the browsing task, the integrated using of the indicators of subjective emotional experiences, eye movements and physiological responses can predict users' approach tendencies to a website well, for the shopping task, the integrated using of the indicators of eye movements and physiological responses can predict users' approach tendencies to a website well.Thirdly, the impact of design elements of e-commerce homepages on users' emotional experiences is investigated. The key design elements that will affect users' emotional experiences of an e-commerce website homepage are identified. A focus group is used to obtain seven key design factors with two levels for each. A total of eight webpage prototypes are created by Adobe Dreamweaver CS5 according to Taguchi's Ls(23) orthogonal array. A multimodal measurement method conjointly using questionnaires, eye tracking and physiological measures is proposed to measure users' emotional experiences. The effects of various design factors on users'emotions are analyzed using the multivariate analysis of variance. Results indicate that the location of navigation of a homepage has significant effect on dominance, but the other design factors have no significant effects on the indicators of subjective emotional experiences. None of the seven design factors has significant effect on heart rate. Both the background color and the picture's interaction of a homepage have significant effects on fixation count. However, the other design factors do not have significant main effect on fixation count. None of the design factors has significant effect on saccade velocity.Fourthly, investigate the impact of design elements of the navigation of e-commerce websites on users' emotional experiences.Identify the key design elements that will affect users' emotional experiences of the navigation of e-commerce websites. A focus group is used to obtain three key design factors (depth, breadth and location of navigation) with two levels each after the investigation of existing literature and Chinese e-commerce websites. A total of four website prototypes are created by Adobe Dreamweaver CS5 according to Taguchi's L4(23) orthogonal array. A multimodal measurement method conjointly using questionnaires, eye tracking and physiological measures is proposed to measure users' emotional experiences. Task completion time is used to measure users' performance. The effects of various design factors on users' emotions are analyzed using multivariate analysis of variance. Results indicate that the depth of navigation has significant effects on pleasure, dominance, fixation count and performance. Navigation with one level obtains positive emotional experiences and better performance. The breadth of navigation has a significant effect on the performance. Navigation with seven items obtains better performance. The location of navigation has no significant effect on emotional experiences and performance.Theory and methods proposed above can not only solve practical problems in measuring users' emotional experiences of e-commerce websites, predicting users' behavior intentions to e-commerce websites, designing emotional e-commerce homepages and designing emotional e-commerce navigations, but also provide reference to design emotional products.