Based on the characters of Boolean retrieval and the structures and representation methods of inference network, an inference network was used to describe the Boolean retrieval and the inference network was combined with the traditional Boolean retrieval from the viewpoint of probabilistic retrieval.

A method of integrating the production system with the neural network is presented and the way to convert an inference network into a neural network is studied.

By the research on the existent IR(Information Retrieval) technology, a text retrieval model based on inference network is designed.

This paper describes a web page recommendation model for the semantic web which collects the annotation information in the web pages, uses the annotations to classify the web pages and stores the web page s uniform resource locator and its type in a database.

In this paper, we designed an intelligent web page recommender system, which include three sub-systems of data preprocessing, clustering, and web page recommendation.

The Web page recommendation systems can effectively predict users\' next clicking results in information retrieval process and the research will be beneficial to many applications ranging from intelligent recommendation to improving effectiveness of search engines.

The algorithm possesses preferable speed of response to satisfy the need of real time web page recommendation and also can be executed without training in advance.

The paper points out the derivation network of web services synthesis is made up of the combination of three elements of the derivation network and the model of web services,all the programs are solved and the price of the specific output projects can be calculated,among which the best solution is he minimum price of the projects.