Consumer Product Recommendation System Based On E-commerce

Along with the continuous scale expansion of the e-commerce market,it provides a wide variety of goods as well as choices for the users.However,when facing such a variety of commodity information,how the users can select the goods quickly and accurately have become the concerned topic for the users and merchants.To reduce the time to select the interesting goods and improve the efficiency of purchasing for the customers,and improve the success rate of sales to create a greater economic benefit for the enterprises,this paper realizes a user recommendation system based on the electronic commerce.This system collects the information and data from the e-commerce users to analyze and mine the characteristics of the user data and shopping records,then the analysis result is used to recommend the commodities for users by using the data mining technology.This paper includes the following research content: First,based on the combination of web technology and database technology,the transaction data and browsing data during the e-business transactions can be collected.The data source about the user's implicit behaviors is collected in two ways.One is the logging data of request to access the website page by the users in the Web Server log;the other is the data collection of clients which is realized by the way of the remote proxy.Second,the collected data are preprocessed by format,thus being used as data mining input.Then,through data mining technology,which is used to analyze and mine the products that users are interested in,and according to the type of goods to match,and based on the established purchase memory function,the products of interest to the users in a given scene is recommended.Finally,the products of interest to the users which are mined from the historical data and the commodity information are recommended by using the combination of the fuzzy clustering and collaborative filtering algorithms.From the rich data information,the recommendation system which recommends goods basing on their preference and interes,enables users to quickly and accurately select the goods they need in the mass of commodity information,reduce the time wasted in the process of commodity selection,and provide users with more humanized services.