Web mining using multi agents

Abstract

Web mining is an important sub-branch of data mining. The data mining technology adopts data integration method to generate Data warehouse. Data warehouse gives Relation Rules, Cluster characters and get the useful Module Prediction and knowledge evaluation. Web data are semi-structure and heterogeneous in nature. Web mining technology is different from pure mining technology which is on database. A Multi-Agent Cooperation Module, which acts as knowledge discovery, exaction machine, Generalization Machine and Analysis Machine has been developed. The communication between Multi-Agent system and Agent Naming Server will enable to gain the useful knowledge in their related task. newlineThe Web crawler is a part of the search engines that fetches web pages along with the other multimedia files from the Web and pre-process the page for extracting information. Web crawlers are also known as robots, spiders, worms, walkers, and wanderers. Web crawling is the process used by search engines to analyze pages from the Web. Web crawlers are essential component for many web applications including search engines, digital libraries, online marketing, and web data mining. MAN (Mobile Agent Name Management) is a three Dimensional column (Agent Naming, Server Location, Birth Time, Death Time, Mining tasks). In Multi-Agents Mining system every Mobile Agent has an Index to an ANS (Agent Naming Server), which accepts Agents searching results and simply processes it. Like router ANS can also notify its neighbor ANS by sending message, which includes valuable information of it. By exchanging information among ANS the valuable information analyzes and cluster. The ANS has a table to maintain the information about what it focuses on. newlineThere are lots of problems in the existing work. Some of them are, not traversing the secure HTTP pages, taking more time to retrieve hidden pages, less precision and less recall. newlineviii newlineCrawlers are not efficient in mobile applications. Retrieving Related and Non Related web pages are not efficient with respect to t