Implementation Of Activity Platform Based On WeChat And Users Analysis

In recent years,WeChat has become an important marketing field in the mobile Internet era,and the market demand for WeChat marketing activities continues to expand.However,the WeChat marketing activities are still costly and difficult to manage,and a WeChat marketing platform capable of agile development and centralized management is urgently needed.At the same time,the homogeneous competition in the WeChat marketing industry is getting more and more serious,and the marketing of companies needs to be transformed into differentiated marketing.Therefore,customer segmentation has become the top priority of corporate WeChat marketing.This thesis aims at the above two points,designed and implemented a WeChat-based activity platform,and designed and implemented the user analysis function for participants of marketing activities.Firstly,from the actual work and related literature,this thesis determined the research objectives of designing and implementing the WeChat-based activity platform and user analysis function,investigated the relevant research background and status,and the related knowledge of the ThinkPHP,MVC design model,customer segmentation,clustering algorithms and AHP(the analytic hierarchy process)was introduced in detail.Secondly,the functional requirements of the activity platform based on WeChat were given in the way of case figures.According to the analysis of the requirements,the appropriate development tools and framework were selected,and the feasibility is demonstrated.The design and implementation of the system were discussed from the overall architecture and the functional modules respectively.Through the system testing and case analysis,the functionality and stability of the system were proved.The system was optimized by fixing the slow loading problem of marketing activities in the testing process.Finally,this thesis analyzed the objectives of the user analysis function,and designed the analysis process that K-means clustering first and then AHP.To overcome the shortcomings of initial cluster centers selection and distance judgment criteria of K-means algorithm,this thesis proposes an initial centers optimization by selecting the non-isolated and non-noise points apart as far as possible from each other as the initial cluster centers and a distance criteria optimization by weighting distances in different dimensions.Through simulation,the accuracy,performance and stability of algorithms were compared and the effectiveness of the optimization was verified.For the results obtained from AHP,through the analysis of the follow-up marketing activity data,the validity of the user analysis function was proved.The system proposed in this thesis helps companies solving the problems of high marketing costs and management difficulties in WeChat marketing,meanwhile offers more accurate WeChat users segment through the user analysis function,and finally achieve the fundamental purpose of promoting companies in WeChat marketing.