Research On Strategic Consumer Classification And Personalized Recommendation Based On Behavior Characteristics

With the rapid development of the Internet industry and e-commerce,consumers have become a part of life through online shopping through e-commerce platforms.Unlike ordinary consumers,which are easily tempted by price concessions,promotional advertisements,etc.,strategic consumers,as a special group of consumers,prefer to choose the timing based on individual utility,if they can be more favorable in the foreseeable future.The price of the product is purchased,even if the current product price is lower than its psychological expectation,the strategic consumer will choose to hold the currency and wait for the purchase.If the company ignores the existence of strategic consumers in marketing,it may cause up to 30%loss of revenue.E-commerce platforms often hold promotions on fixed dates to promote sales and compete for consumers.The regular organization of these events enhances the strategic waiting behavior of strategic consumers during online shopping;and the goods and services on the e-commerce platform The number of information is growing exponentially,and the resulting information overload problem is also increasing with the advent of the big data era.Therefore,identifying a strategic consumer in an e-commerce environment and recommending a product of interest for it can effectively improve the shopping experience of the strategic consumer while increasing the profit of the enterprise,while the traditional recommendation method is less by identifying the strategic consumer.Therefore,it is necessary to conduct in-depth research on strategy-based consumer classification and personalized recommendation methods based on behavioral characteristics.This paper has completed the following three aspects of research work:(1)A strategy-based consumer classification method based on behavioral characteristics is proposed.Through the analysis of the characteristics of strategic consumer behavior,it is clear that strategic consumers in the e-commerce environment mostly have the characteristics of strategic consumer behavior such as holding money and delaying purchase.To this end,the historical transaction data accumulated by the e-commerce platform is used to construct the behavior characteristics of consumers.Based on the XGBoost classification model,users are divided into strategic consumers and short-sight consumers,which lays a foundation for personalized recommendation for strategic consumers.(2)A strategy-based consumer recommendation method based on online interaction behavior is proposed.Based on the behavioral-based strategy-based consumer classification,the behavioral scoring matrix is scoring constructed using the multi-behavior data of the strategy-based consumers,and the improved SVD++algorithm is used to calculate the item sorting list and sort the items of interest to the target users.Then,the products that the user has purchased are removed,and the products with the highest ranking are recommended to the user,and finally the personalized recommendation based on the behavior characteristics is implemented.(3)Application research of A-site strategic consumer classification and personalized recommendation based on behavioral characteristics.Using the actual e-commerce A website transaction data,the behavior-based strategy-based consumer classification and personalized recommendation method proposed in this paper is applied to the website recommendation process,and the behavior is based on the consumer's behavior characteristics and actual consumption behavior data.Feature modeling and strategic consumer classification,and calculation of product recommendation list,recommending products of interest for strategic consumers.Finally,the feasibility and effectiveness of the proposed method are proved by comparing the recommendation effect evaluation index with the recommendation results formed by the classic recommendation method.The research results show that the behavior-based strategic consumer classification and personalized recommendation method proposed in this paper can provide a reference for the real-time e-commerce website to identify strategic consumers and solve their personalized recommendation problems.