The Study On Intelligent Information Mining And Service Archetypal System Of College Digital Libraries On Agents

Main goal of this article is that applying intelligent Agents to digital library information gathering and service system in order to raise its personality service level.On the base of deeply analysis in the present situation of individuality information service technology and the research application in digital library, as well as understanding and introducing Agents, search engine and web information gathering technology, applies the intelligent Agents technology on library information gathering and the service system, proposes a model of intellectualized information gathering and service system based on Agents in college digital library. Revolving the construction and software realization of this model, the main work includes:First, analyzes the existing library information management system, detailed understands its function and application function. Prepares for the next step of work.Next, structures a Multi-Agent system, all of the subsystems is both mutually related and mutually independent, may study the technology which involved on each subsystem separately. This MAS works above an existing library information management system, MAS transforms user's search request to the search condition which can be carried out by the existing system through a connection and then do the search work; Additionally, inquired result is filtered and pushed to users defer to users' individuality characteristics through system study. The system searches information entirely and accurately according to the user interest, this kind of ability is decided by the study of user interest through user Agent and the complete and accurate degree to the user interest information stored in the user model database, and the completeness and accuracy of the user characteristic information unceasingly enhances thorough along with accumulation of the system study in user interest.Third, it simultaneously expressed three kind of information descriptions model in this system model: the documents, the domain and the user model. The documents is the object of information processing, the documents model establishment is the foundation of such intelligent information processing as characteristic withdrawing and documents filtering; Expression the domain model by the subject concept and the key words, highlights the information service characteristic of facing domain, and canbetter realize individuality and intellectualization of the information service; The user model manifests the user interest and intention, and is used in the way of user interest expression and mining.Fourth, the search and monitor Agent searches documents and information through domain model.Filter Agent analyses the homepage collected, extracts the documents characteristic, and forms the documents structure expression. In filter Agent, it proposes withdrawing method of the documents characteristic and key word of new domain; Through comparison the structural documents attribute and the user model, recommend Agent seeks the most similar user, and recommends the user the final documents.As well as comparison between user's similarities, thus recommend to similar user corresponding URL or other documents. Recommend Agent describes documents choice and the documents recommendation strategy between the similar users.Finally, makes the conclusion to the work had done and proposes work forecast.