The invention discloses an electronic-commerce recommendation method based on the user expression analysis, mainly comprising the following steps: starting an application program of a handheld side by a user to log into an electronic-commerce website to browse commodities and shop; starting a camera of handheld equipment by the application program to capture images of the user; analyzing the images to obtain the information on the expression of the user; and amending the recommendation result or algorithm according to the feedback of the expression of the user to obtain and finally recommend the commodities satisfying the demand of the user to the user.