Research On Precision Marketing Of E-commerce Based On Data Mining Technology

With the development of the economy and society and the continuous advancement of Internet technology,the e-commerce industry has ushered in a development boom,and more and more business merchants have begun to get involved in the e-commerce industry.In recent years,the number of e-commerce platforms has also been increasing,and the increase in the number of platforms has led to an exponential increase in the number of merchants.Therefore,the competition in the ecommerce industry has become increasingly fierce.On the other hand,the development of the e-commerce industry has also led to changes in consumers’ consumption psychology,which increasingly pursue personalized services.Due to the imperfect service of the store itself,the loss of customers will cause irreversible damage to the store.In this context,the e-commerce industry urgently needs precision marketing.Since the e-commerce industry has a huge advantage in collecting customer data,it is a common practice in the current e-commerce industry to use big data technology to guide precision marketing.Using data mining technology to analyze the relevant data of users,classify users,and formulate different marketing strategies for different types of users can increase the loyalty of store users and achieve stable sales benefits.This paper uses a clothing e-commerce platform for a period of time sales data and commodity review data on the precise marketing of clothing e-commerce research and analysis.First,it analyzes the characteristics of clothing e-commerce sales,and combines the traditional RFM model with the clothing e-commerce The characteristics propose improved indicators: D(the difference between the last shopping and the average number of days between shopping),Q(the number of purchases in the most recent period of time),M(the average consumption amount in the most recent period of time),C(the most recent The purchase conversion rate within a period of time),a new DQMC model was formed,and the PCA algorithm was used to replace the traditional weight measurement method to calculate the weight of various indicators,the new model was used to measure user value,and the K-Means algorithm was used The users are clustered,and finally the users of the store are divided into five categories:high-value users,key users who are saved,potential users,general users,and lost users.The user value calculated by the new model is used to verify the clustering results.The verification results show that the clustering effect is consistent with the value classification effect,which confirms the effectiveness of the newly constructed model.In the end,different types of users are proposed.Marketing advice.The new model has strong guiding significance for clothing e-commerce to carry out precision marketing,and it has strong practical significance for shops to increase user loyalty and enhance shop image.