Research And Implementation Of Intelligent Keyword Suggestion Technology For E-Commerce Product Search

With the continuous development of Internet technology,e-commerce has become a bright spot for economic development.The market scale continues to expand,the market structure continues to be optimized,and the division of labor system is gradually perfected.At the same time,the "double creation" activities vigorously promoted by the country have created favorable conditions for the development of small and medium-sized e-commerce companies.Under this background,e-commerce companies are booming and e-commerce websites are becoming increasingly sophisticated.In order to enable users to quickly locate the products they need in the mass merchandise information,the search engines become standard on the e-commerce website.The use of intelligent keyword suggestion in search engines can effectively reduce user input,eliminate query ambiguity,and improve the convenience and accuracy of information retrieval.This paper aims at e-commerce in this specific field,through data mining technology,proposes a intelligent keyword suggestion method that comprehensively considers the user's search and shopping behavior.The main work of the dissertation includes:First,this article first analyzes the keyword suggestion system and proposes a product search keyword intelligence suggestion mechanism and a product search keyword dictionary model.Due to the large amount of log in the e-commerce application,the user feedback required by the intelligent keyword suggestions requires high real-time performance.However,the mining algorithms for large-scale data all take a certain amount of time to obtain the calculation results,and the log data cannot be analyzed in real time.Therefore,this article divides the intelligent keyword suggestion system into two parts:online and offline.The offline part analyzes and calculates the logs generated before the system,stores the results into the keyword thesaurus,and sends the results to the keyword thesaurus.The user enters the keyword and obtains relevant keyword results to ensure real-time performance.Second,a intelligent keyword suggestion algorithm that comprehensively considers the user's search behavior,user shopping behavior,and merchandise inventory is proposed.In the past,the intelligent keyword suggestion system based on log mining mostly focused on the user's search behavior.There was little application of the user's shopping behavior in the intelligent keyword suggestion.The e-commerce website's user shopping behavior reflected the user's shopping tendency more.The value of excavation is huge.Furthermore,from the perspective of the merchant,recommending to the user a product that meets the needs of the user with a large amount of inventory is in line with the merchant's own interests.Therefore,this algorithm integrates the user search,shopping behavior,and inventory information of the product.Third,we designed and implemented an intelligent search system for product search keywords.Based on the above product search intelligent keyword suggestion model and product search intelligent keyword suggestion algorithm,the user log mining subsystem was designed and implemented on the Hadoop platform,and a keyword thesaurus was generated;in the online subsystem,the front end search page was implemented,Backend query processing and other functions.The experimental verification shows that the product-based intelligent keyword suggestion method based on log mining can effectively improve the accuracy of keyword suggestion,reduce user input,and has good processing performance.