Research Of Forest Products Trading Web Messages Extraction Based On Semantic

Information push service has been widely used in electronic commerce with the popularization and application of the rapid development of electronic commerce. Web information extraction technique has realized information extraction and structured storage. Due to the lack of normally organizational structure of forest products trade websites currently, users’interested content can’t be found directly from forest products trade websites and forest products trade of information push service can’t be realized. Using Web information extraction technique to extract users’interested content has become very important in the forest products trade of Web information push service.Based on the shortages of the existing Web information extraction technique in the presence of the low precision and automation, combined with the needs of Web information extraction technique in forest products trade Web information push, analyzed and used the features of forest products trade Web information, the forest products trade semantic dictionary has been constructed. The target information has been extracted by using the local features of the target information in web pages, and positioning the target information automatically by means of the semantic information entropy method that combined with the information entropy. On the basis of the partial target information that had been extracted, the target information that not be extracted has been extracted again by using the method based on template. All of the extraction results have been stored structured in the database finally.The experimental results show that this method has good performance in precision and recall. This paper pointed out the problems in this field that to be improved in the future.