Research On User Value Mining Method Of Shopping Guide Platform Based On RFM And BP Neural Network

In recent years,with the rapid development of Internet technology in China,the Internet has brought a lot of convenience to people's life.For example,online shopping,people only need to click the mouse or slide the mobile phone screen to buy the goods they want.Online shopping is inseparable from e-commerce platform.There are many kinds of e-commerce platforms,among which e-commerce shopping guide platform is a new type of e-commerce platform,It provides a new shopping mode for people.Users of the shopping guide platform can buy goods with lower price or higher cost performance.User scale is the first consideration in any stage of shopping guide platform.Sufficient users are the premise of efficient operation of shopping guide platform.Therefore,it is of great significance for the e-commerce shopping guide platform to fully tap the value of users and formulate different marketing strategies for users with different values based on the user data.This paper mainly studies the user value classification and user value classification prediction of shopping guide platform.Firstly,using the basic attribute information and user purchase information data collected by shopping guide platform as experimental data,this paper puts forward the user value classification method of shopping guide platform based on improved RFM model to realize the value classification of shopping guide platform users;On the other hand,according to the marked user value data for supervised learning,this paper proposes a classification and prediction method based on BP neural network for the user value of shopping guide platform to realize the classification and prediction of the user value of shopping guide platform.The user value classification method of shopping guide platform based on improved RFM model: firstly,the average order time interval of users is used to replace the latest consumption time,and the profit brought by users to the platform is used to replace the consumption amount of users,and the viscosity index of users is increased to improve the traditional RFM model.In this paper,the improved model is called RFMS model.The combination weighting method of AHP and entropy weight method is used to determine the index weight of RFMS model,and K-means clustering algorithm is used to cluster and subdivide the users of e-commerce shopping guide platform.Finally,the users are divided into four categories.According to the clustering results,the user value of each category is analyzed,and marketing suggestions are put forward.Experimental results show that,compared with other methods,the user value classification method proposed in this paper has better effect and is more suitable for shopping guide platform.Classification and prediction method of shopping guide platform user value based on BP neural network: after clustering analysis,we get the marked user value data,which can be used for supervised learning to classify and predict the value of shopping guide platform users.Firstly,select the characteristic attributes of shopping guide user value classification prediction,determine the structure of BP neural network according to the data characteristics,use particle swarm optimization algorithm to optimize BP neural network,get the PSO?BP model,PSO?BP model is used as a weak classifier,and Ada Boost algorithm is used to integrate multiple PSO?BP model is combined into a strong classifier to improve the prediction accuracy.After verification,the prediction performance of the proposed method is better than that of BP neural network and PSO?BP model and decision tree model.