A Multilingual Information Extraction Framework for Digital Library

摘要

As the influence of the library digitization movement is getting wider and wider, more and more countries have started their own digital library projects. Therefore, due to its importance for content integration and knowledge discovery, information extraction for different languages is becoming a key problem for the development of digital library. In this paper, we present an information extraction framework that suits for digitized textbooks in different languages. To achieve multilingual adaptation, our framework introduces language independent features that adopt domain characters to generate extractors for a certain textbook. At last, we also present some results of a preliminary experiment to show the feasibility of the framework.