The Research Of Information Automatic Collection Based On The Agent Of Internet

With the development of internet and relative technology, Internet , especially the www has already become the platform that people send or acquire information. The methods gathering information resource from the internet have become more and more important nowadays and in the future.However, it's inconvenient when people face enormous information in the internet .Two points, first, named information over loading that means information active or downloaded can' t bo quick in the uptake, Second , the users can' t define the expression for the information resource they want to search for(because of no appropriate expression approach according to current technology) and don't know how to search for the information resources more exactly and efficiently, that is named resource wildering.The search of information automatic collection based on the agent of internet is an agent used by individual or common enterprise. It can learn itself time after time to master the interest of the users and send to user filtered information in this way to realize the intelligentized initiative information service. Practising the agent according to the interests and feedbacks of users and gathering the fresh interests of users, in this way people can get individual service and resolve the two previous problems. The article introduced Agent machine learning Chinese word segmentation. Generally, focused on over agent's intelligibility, substitution, go-aheadism. This information automatic-acquiring system has been set up based on the agent and combined previous technologies.This article is composed of three parts ,data collection , data filtering and machine learning .These three parts were assembled organically and enhanced the intelligence as far as it can at every point to improve on the traditionary word segmentation algorithm and inductive learnings...