The Research Of Topical Search Engine Base On Feature Model

With the development of enterprise information construction,enterprise information resources become more abundant,oil enterprise information portal website has become an important platform for enterprise information management,in order to promote the information sharing and interaction within the enterprise,we developed oilfield information search engine aiming at oilfield enterprise information.However,with the increasing amount of information,in terms of content,the information portal website of the different units within the enterprise has a certain similarity,when users want to search for information on a particular topic,the search result of the existing oilfield information search engines including many information that are unrelated to the topics,so,search results are confused,the topic of information is unclear,and lower retrieval efficiency comes with it.Therefore,how to allow users to quickly identify the sources of information on the topical is an urgent problem that the oilfield information search engine needs to solve.In view of the above problems,this paper proposes the solutions of the theme search engine based on feature model.Specific studies are as follows.Firstly,study current oilfield information search engines and propose the solutions of the theme search engine based on the feature model.Study the effect of feature model in the topical search engine.In order to improve the accuracy rate of web feature extraction and the recognizable theme of the search results of users,the feature model is applied to the theme search engine crawlers and search results.Secondly,research the structure characteristics of the enterprise information portal,a opinion is proposed that web features can be divided into internal and external features,according to the characteristics of the visual and semantic of web relationship were summarized on pages.And proposed that build web feature model combined internal feature with external feature,and use it to describe between web feature and they relations.At the same time gave a formal definition and method of feature model.Thirdly,an method of web feature extraction based on feature model was proposed after study the processed of web feature extraction methods.In the method,first feature extraction rule was extract and semantic information in the sample web were mapping to the web feature model,then extract web features according to the web feature model,the method solved the problem of incomprehensive and lack of semantic description of the existing methods of web feature extraction.Finally,research the application of web feature model in the topical search engine design and realize the system of topical search engine based on feature model and testing the feasibility and effectively of the theory.