Research On E-commerce Recommendation Based On Multiple Consumer Psychology

E-commerce recommendation plays an irreplaceable role in alleviating product information overload,improving consumers' personalized experience and sales conversion rate.According to the idea of recommendation,e-commerce recommendation can be divided into two types: recommendation based on correlation and recommendation based on causality.The former,such as collaborative filtering,relies heavily on data.The latter,such as commodity recommendation based on consumer psychology,has great advantages in new commodity recommendation due to the introduction of consumer psychology knowledge in the recommendation.But at present,the recommendation based on consumer psychology is usually the recommendation based on single consumer psychology.According to the common understanding of e-commerce recommendations,the more you know about a user,the more likely you are to make an accurate recommendation.Then,when all aspects of consumer psychology,such as consumer attitude,consumer motivation,consumer interest,etc.are applied to product recommendation,can the recommendation accuracy be greatly improved?How about the recommendation rules under the combination of various consumer psychology?In order to answer the above questions,this paper chooses three typical consumer psychology,namely "consumer motivation","consumer attitude" and "consumer interest",to explore the product recommendation of multiple consumer psychology.The work of this paper is as follows:(1)Clarifying the combination mode of multiple consumer psychology in commodity recommendation.On the one hand,according to the consumer psychology theory and the needs of e-commerce recommendation,select some commodity-shop attribute as attribute variable,and conduct discrete processing to the attribute information of the goods purchased by the user.On the other hand,according to the Bayesian network method,this paper analyzes two combination methods: process combination recommendation method and result combination recommendation method,and selects the process combination method to carry on the research of multi-consumer psychology product recommendation.(2)To summarize the rules of commodity recommendation based on multiple consumption psychology.A user,for example,double first construct the user's consumption psychology and triple bayesian networks under the consumption psychology,analyze the multiple type consumption psychology,using bayesian network inference method recommended goods,after82 user authentication at the same time,has been clear about the double based on consumer psychology and triple consumer psychology recommended method can effectively realize recommendation.This article analysis the single consumption psychology,dual consumption psychology and triple psychological difference in recommended the precision,found that compared with the single consumption psychology,based on double consumption psychology and triple psychological recommendation recommendation precision ascension has been basically achieved,precision is higher than general triple and double consumption psychology,consumer psychology,The accuracy rate of psychological goods recommendation with dual consumption motivation and attitude is the highest.