Product's Arrangement And Involvement In Recommender Interfaces Influence On Product Selection

The recommendation system is designed to help users deal with information overload and efficient selection of optimal projects from many complex types.Today,the information on the network is expanding,becoming more complex,and the update speed is fast.We urgently need a good recommendation system.Help us overcome this excess information.However,most of the research on the recommended interface emphasizes the accuracy of the algorithm,rather than the user-centered application of the recommended interface.Therefore,this study aims to examine whether the newly developed recommendation interface can help users select products more efficiently by comparing the impact of the two mobile end product arrangement methods and the high and low two degrees of involvement on the recommended interface product selection.Whether it can make users notice more product information.This experimental study is divided into three parts.The first part is a preliminary experiment,which is designed to screen out the high and low involvement products as experimental materials,and to understand the factors that the subjects mainly pay attention to when purchasing these products.The preliminary experiment adopts Vaughn.Scale to assess the degree of involvement of the product,and the use of questionnaires to assess the factors that the subject will consider when purchasing the product as the basis for classifying the recommended product.The second part is the formal experiment 1,a behavioral experiment.Involvement of products(high and low)and two kind of recommended interface(integrated sorting,sorting)was used as variables of this experiment,and the subjects are presented by E-prime.The different recommendation interfaces are presented in turn,the presentation of the materials is balanced by the Latin square design.The subjects can browse the interface at will and make a purchase selection in the recommendation interface.The third part is the formal experiment 2,the eye movement experiment,aiming to further explore the subjects' different eye movement regular and decision-making quality when purchasing choice in different recommended interface.This experiment uses the same experimental design as Experiment 1,using RED desktop eye tracker to present stimulus materials and recording eye movement data,and also gives the subjects different presentations in turn.The stimulating materials allow the participants to browse through the simulation recommendation interface with different product arrangement and different intrusion products,and make purchase choices.Then,the questionnaire is used to evaluate the decision quality of the product selection.Finally,The eye movement data of the subjects were exported and analyzed.The results show that in the first experiment,the interaction effect of the arrangement and the degree of involvement on the decision time is significant.In Experiment 2,the interaction effect of the arrangement and the degree of involvement on the residence time and the number of fixation points is significant.Under the two levels of the degree of involvement,the difference in the residence time and the number of fixation points at each level of the arrangement is significant.The interaction effect of arrangement and involvement degree on decision quality score is also significant.Under the two levels of arrangement,the difference in quality scores of each level of involvement degree variable is significant.Under the experimental conditions of this experiment,the following conclusions can be drawn:(1)The recommended interface arrangement has an impact on the user's product selection efficiency.In the recommended interface of the classification arrangement,the user is more efficient in searching for the required goods.(2)The degree of involvement of the product in the recommended interface has an impact on the user's product selection efficiency,and the selection efficiency of the low-involved product is higher than the selection of the high-entangled product.(3)Under the condition of low entanglement,the recommended interface arrangement has a significant impact on the quality of the user's decision-making.The quality of the decision-making in the recommended interface of the classified arrangement is better than that of the integrated arrangement.