基于数据挖掘的信息过滤系统的设计与实现-计算机软件与理论专业论文.docxVIP

现。关键词： 现。 关键词： 信息过滤数据挖掘WinsockSPI聚类分析 分类号：TP309 lI AbstraetAbstract Abstraet Abstract Internet has developed rapidly during the past years and it becomes the largest information net around the world．But when people share the convenience brought by Internet．the harmful information and data comes along诵th the ones beneficial．So the question that how we can get the prime part from the abundant database comes to as and it’s becoming one of the important domain in the net-tech research． The main task of the Info-Filtering system is to purify the data packet from the web sites in the Intemet．And most of the Web sites organize their data by the semi-structured Html page．So this article focuses on the approaches to check the web data from Interact．It contains the points below： 1．The introduction to the current technique of information filtering． First，it expound the development of information filtering and the pivotal technique．The article also lists some defects of the current information filtering system includes the bad veracity，the low rate，the bad agility and so on． 2．Bringing the new scheme forward． one of the most important task of the information filtering system is to establish a correct，reliable and exact warehouse which contains the samples．This article makes scheme which lead the system to renew itself and organize the samples automatically．So the system’s working velocity grows faster． 3．Because the final purpose of the information filtering system is to decide whether the data packet is good or bad so that to hold it up or not．This article brings out scheme to do this．It’s detached into two parts includes ascertaining its topic and quality．To complete the first one，we use Bayes Technique；to accomplish the second one，we use KNN model． 4．Establing the framework of the information system based on Data mining． Based on data mining and the technique of dealing with the data packet from Internet，the article designed a framework of the information system which contai