Research On Contextualized Recommendation Strategy Oriented Customer Purchase Behavior In B2C Platform

B2C(Business to Customer)is a type of e-commerce model where websites sell products and services directly to consumers.In the tide of Internet development,online shopping has been loved and sought after by the general public,and the market share and growth rate of B2 C websites have greatly increased.When consumers shop on the e-commerce platform,they are faced with the problem of information overload due to the number of goods and services is increasing.As a result,problems such as difficult to choose,longer selection times,and poor shopping experience lead to consumers,their willingness to purchase is not strong.Since the 1990 s,the development of personalized recommendation services has received more and more attention of scholars,because the personalized recommendation service can solve the problem of information overload.It taps the relationship between users and commodities as consumers.Recommend products that meet their preferences and achieve the purpose of stimulating consumer purchase intentions.However,consumer preferences are influenced by contextual factors and are characterized by uncertainty and differentiation.The current recommendation system does not address these issues well.In order to better improve consumer's willingness to purchase and satisfaction of shopping experience,the e-commerce platform needs to launch a recommendation service that considers the consumer's context and meets the consumer's preference,that is,the contextual recommendation strategy.A contextual recommendation strategy for e-commerce platform consumers is proposed to address the lack of consumer scenarios in existing recommender system of e-commerce platforms.The main content of this study is divided into two parts:First,characterize and definite the context of e-commerce platform consumers.Under the e-commerce environment,the situation that influences the purchase behavior of consumers is complicated,and different contextual factors have different influences on consumer behavior patterns.Through analyzing the characteristics of the purchase behavior of e-commerce platform consumers who have not drifted for a certain period of time,the situational factors and characteristics that affect the purchase behavior of those consumers are studied,and the e-commerce platform consumer context are characterized and defined.Second,analyze recommended strategies that consider consumer-sensitive contexts.Describe the consumer preferences from consumer behavior information,and structurally model the consumer context preference information.By describing consumer behavior information to describe consumer preferences,a structured modeling method for consumer context preference information and a recommendation method that integrates into consumers' sensitive contexts are proposed,and the concept of distributed difference is introduced into cognitive psychology.Focusing on analyzing the different influences of sensitive situation factors on consumer preferences,the numerical experiment method was used to analyze and verify the effectiveness of the proposed consumer preference extraction method and context-specific recommendation method.After theoretical research and numerical experiments,Three conclusions are drawn:(1)Using literature review to analyze the effect of context-based recommendation system on consumer purchase intention of e-commerce platform,the results show that based on personalized recommendation service,combined with the context dependence of e-commerce platform consumers to explore the customer's purchase behavior is a more suitable and novel Perspective;(2)The consumer preference extraction method based on the distributive distribution of consumer context has good user preference extraction ability;(3)The recommendation method that incorporates sensitive situations has higher recommendation accuracy than other recommendation methods and can adapt to complex situations and provide consumers with high quality product recommendation services.