Unsupervised Metadata Extraction in Scientific Digital Libraries Using A-Priori Domain-Specific Knowledge.

摘要: Information extraction from unstructured sources is a crucial step in the semantic annotation of content. The challenge supporting an high quality automatic approach (or at least semi-automatic) order to sustain scalability semantic-enabled services future. Unsupervised information encompasses number underlying research problems, such as natural language processing, heterogeneous integration, knowledge representation, and others that are under past current investigation. In this paper we concentrate on problem unsupervised metadata Digital Libraries domain. We propose present novel focusing improvement without involving external (oracles, manually prepared databases, etc), but relying document itself its corresponding context. More specifically, focus improvements scientific papers (mainly computer science domain) collected various over Internet. Finally, compare results our with state art domain discuss future work.