Design And Implementation Of E-commerce Price Comparison And Recommendation System Based On Web Crawler

With the rapid development of e-commerce,consumers shopping through the Internet are becoming more and more widespread.While many e-commerce platforms provide consumers with more product choices,they also increase the difficulty for consumers to quickly select satisfactory products.The price comparison website crawls the latest product information from various ecommerce platforms through the web crawler,and can display the price comparison information of the same product on different e-commerce websites for users,providing a reference for users to choose a more favorable e-commerce platform.However,the merchandise display mode of the price comparison network is mostly a passive display,that is,the system only displays related merchandise information according to the users' search keywords,and does not actively recommend products to users.In this paper,collaborative filtering recommendation algorithm is combined to increase the product recommendation system on the price comparison network so as to realize the transition from the passive presentation mode to the active presentation mode that can be personalized recommendation.Aiming at the problem that traditional collaborative filtering algorithms can't adapt to changes of users' interests in a timely manner,this paper proposes the concept of users' interest time and combines information entropy to improve the users' similarity calculation method,and proposes a collaborative filtering algorithm based on information entropy and users' interest time.IEICFA)is the core algorithm of the recommendation system.When crawling the e-commerce website product information,in order to solve the general network crawler can't meet the ecommerce platform product crawling demand restrictions,this article for the e-commerce platform structure differences,the main e-commerce site respectively customized different crawling The strategy implements a customized web crawler to complete the collection of product information and provide original product data for the system's price comparison and recommendation function.Finally,this paper designs and implements a web crawler-based e-commerce comparison and recommendation prototype system,which can provide users with registration login,search price,product recommendation and rating evaluation.The custom web crawler implemented can adapt to the site's structural features and complete the collection of commodity data.The proposed IEICFA recommendation algorithm can adapt to changes in users' interest over time and can effectively improve the accuracy of the recommendation.