Literature search framework by analyzing key aspects

Quickly moving to an unfamiliar field for researchers is painful due to a mass of scientific articles to study without knowing their research contents. Research point and its implementation method are two helpful aspects. However these two aspects are difficult to be achieved without appropriate semantic tools. This paper proposes a literature search framework which supports a new perspective retrieval of research point and implementation method of articles. This framework contains four components: extracting the above two aspects from articles, analyzing them by semantic similarity, ontology population and index building. The experimental results verify the effectiveness of this framework. And we apply this framework to the field of computer science, demonstrate the prototype system.