The semantization of library resources is the general trend in a networked environment, therefore, we propose methods using natural language processing technology and machine learning to promote library resources semantization process. We develop semantic metadata online auxiliary construction platform of Chinese information resources to do semantic label of related literature, and generate initial semantic metadata by utilize traditional organizational resources (such as thesaurus) with machine learning method based on the relatively semantization of a large amount of content, and realize simultaneously implementation of semantic metadata system construction and organizational resources semantization. The system proposed in this paper can help to construct the semantic metadata of scientific and technical information based on the Web, and utilize up-to-date semantic technologies to realize domain thesaurus and effectively integrate various kinds of metadata, it can also realize the edit, development and maintenance of a certain semantic metadata project by automatic or auxiliary construct the semantic metadata system using the semi-structured text. The realization and implement of this method can provide technical support and theory reference for accelerate the progress of Chinese resource semantic annotation.
展开▼
