Design And Implementation Of Comparison Shopping Engine System

With a wide range of the popularity of the Internet, the number of networkterminals and the ways to connect into Internet is increasing much faster, the onlineshopping also has a great development. As the same, the concept of network shopping isalso deeply rooted among the people. Due to the increasing of the online shopping andtransactions, the number of network trading platforms and online shopping malls isbecome larger and larger. The number of online malls, sellers and commoditiesmultiplies at the same time. As a buyer, if you want to buy the goods as your wishes,you need to browse a large number of websites, to compare the prices of thecommodities, and this means a lot of time cost. In this case, a number of shoppingwebsites are created. But these shopping sites just put the goods which are chosen bythe sites from different malls together, and not do further processing and create anyclassification. And it will not be convenient to use the shopping sites. This system isdesigned to solve these problems of the buyers and improve the network shoppingexperience.The system implements the following functions to help the network shoppingbuyers get a better shopping experience:First of all, the design of data capture is flexible, easy to extend or modify. As aresult, the system can get all kinds of goods’ information from various websites, tradingplatforms and malls. Further, the system can also cover the emerging shopping networkstation or shopping platform at the first time.Secondly, when the user search goods by the system, different with the traditionalplatforms, the result this system returns will try to cover different brands, models andproducts. It will provide more useful information to users, and then users can makebetter choices. And according to the different businesses, the seller’s reputation andservice, information products of different brands of popularity and so on, theadministrator can configure to make more favored by buyers, sellers, brand businessinformation in a more forward position for the convenience of users.Third, an automatically running disaster recovery system which can also bemanually operated is implemented. Each time when the system gains data, the old datawill be saved as disaster information. If any error occurs during the information gathering, system will automatically switch to the last data which can make the systemworking to make the system working.Fourth, the cache system which has no hardware limit and can be continuallyimproved is implemented. The administrator can configure to adjust the system resourceconsumption, to adapt to all kinds of hardwares, the administrator can also improve thesystem efficiency by adjusting the static cache. The system itself will also continue toimprove the efficiency of the system through the dynamic cache.This paper uses Nutch and Lucene the two scalable, open source informationretrieval architecture with high performance as the basis for development, to solve theproblem. At the same time, for the realization based on knowledge base, which makesthe system in the process of using, the administrator can continually improve knowledgeto improve performances of the system.