Application Of ANN Technology And PSO In Document Feature Description

Popularization and development of Internet technology make Internet become the most abundant computer network of the current information resources,how to search information effectively in Internet has become one of the problems that people generally care about. In order to realize information retrieval, file information in Internet must be compared with the retrieval requirements, and then the document which satisfies the user's needs can be present to users. In information retrieval the document is described to be document feature, how to present document feature accurately is the base of information retrieval. Document feature is the most important problem in document retrieval fields, it decides the classification and aggregation of the document, and also decides the user's reaction ability to the document retrieval system (recall rate, precision rate).Information retrieval technology is developing in individualization and intelligent direction, individualizing and intelligent information retrieval feedbacks the retrieval results to the users as the retrieval needs of users. This passage analyzes the advantage and disadvantage of the Boolean Model and the Vector Space Model in information retrieval model: the Boolean Model is simple to realize, but its classification ability is limited and the logic expression formula is strict. The Vector Space Model maps the document to the continuous vector fields, it provides the basis to the weights computation of multi-elements dealing, but it is easy to ignore the common feature of the document set. This passage analyzes the feasibility of...