User Behavior Analysis For Agricultural Electronic Commerce Platform

With the continuous improvement of the level of the information technology combined with agricultural field,the mutual competition of agricultural products e-commerce platform enterprises becomes more and more fierce.In order to keep old users and also continue to explore new users,you must improve the service quality further of its own.Under this background,the agricultural e-commerce platform enterprises must pay attention to their user data,and use data mining tools to analyze the users' behavior,to understand the characteristics of users' behavior deeply,and to provide relative agricultural products.In this thesis,the research background,purpose and practical effect was described.The summarizing of the current electricity supplier of agricultural products,the progress of data mining research in users behavior analysis,and the basic arrangement between the chapters was shown in Chapter one.The second chapter of this thesis briefly described the basic theory of data mining and the research of user analysis.Then it introduced the basic concepts and steps of data mining technology,introduced some basic knowledge of clustering algorithm,and then introduced the commonly used methods of users' behavior analysis and the concept of user behavior analysis and make some comparisons.The third chapter contained to improve the application of traditional K-Means algorithm in the field of user behavior analysis of agricultural products business platform.It embodied to improve the traditional algorithm to determine the best clustering number K value by constructing a weighted distance function to achieve the optimal K value optimization.In the process of solving the optimal K value,it makes a principle that the K value will be the best when the weight distance function reaches its minimum.Then it gave the optimal K value range and increased the speed of the improved algorithm.Experimental results showed that the improved algorithm was effective.In chapter four,according to the improved algorithm,an agricultural e-commerce platform user's behavior analysis model was established,it described the improved K-means algorithm in the model application in detail,the data pretreatment process,the methods and the results were described.The result of the analysis of one agricultural e-commerce platform user behavior showed the improved method is operational in dividing the users' behavior,assisting platform enterprises to make marketing strategy,providing different agricultural products and services for users with different needs and other aspects.Based on the established analysis model of users' behavior,the users' behavior analysis system of agricultural products e-commerce platform was designed.This chapter described the framework of the agricultural product electronic commerce platform users' behavior analysis system function realization,and described the main functions of the system,and the system tests were introduced in the fifth chapter.The last chapter summarized the research work of the full text,and put forward the direction of further research from two aspects of the theoretical research and practical application.