The User Personalized Retrieval Model Based On Ontology Structure Research

With the rapid development of Internet and the explosive growth of information,information retireval in peopl’es production and life play an increasingly important role.Desired user in the "information trek" and "information overload" problem when the use ofinformation retireval system is basically a "one-si"ze-ifts-all, it is diiffcult to meet thepersonalized needs of different users, intelligent search engine, and based on big datainformation acquisition and processing technology is the key of the current informationretireval study subject. The main purpose of the information retrieval system is theinformation needed to satisfy the users that demand, is one of the fundamental problem is howto from a given concentration of document retireval relevant documents, and user needs, andaccording to the correlation rfom high to low order is returned to the user.In this paper, based on ontology and knowledge of the context in the process of userpersonalized retireval, semantic problems of statistical language model methods respectively,the user personalized demand made a research on the aspects of the understanding of theproblem, put forward from the ownership of the concept of words-and the co-occurrenceAngle correlation calculation method, and put the words-concept relevancy measure words-concept relations into the statistical language model to improve the computation eiffciency andretireval accuracy. In this paper，based on the study of the current information retireval modeland method, this paper expounds the design pirnciple of the personalized user retireval modelbased on ontology, basic ideas and technical rfamework, the functions of the model system.The fused ontology to the user personalized information retrieval technology, in the deepunderstanding of user requirements and semantic aspccts of resource use of domain ontologyknowledge, of documents, such as concept, words’ annotate, organization, and to build andimprove the ability of machines for the natural language understanding and the efficiency andaccuracy of retireval. The user context information retireval is the main way to deepunderstanding of user personalized demand, document context can be used to understand thecharacteirstics of the document, to facilitate users to check the matching results. Based on theabove research proposed a user personalized information retireval based on ontology prototype framework, design of various modules of the model and key technology are pointed out.Research of this paper is an exploration process, in a very baste stage, after lurthervalidation arc constantly improve, hope to be able to apply it on the Intenret，convenience forthe users to study and research.