Semantic Search for Scientific Publications Based on Rhetorical Structure

Most scientific papers have their own rhetorical structures, which have deeply rooted in the minds of both authors and readers, such as background, problem and discussion. However, most existing search engines for scientific publications haven‘t made good use of such semantic information. In fact, each reader would be interested in different semantic modules of a paper, that is, certain concepts or entities mentioned in different semantic parts represent various indications. In this paper, we design and implement a semantic search platform that aims to provide semantic search for scientific publications based on rhetorical structure. To provide better results, we initiate with the semantic model of scientific papers, so as to meet the special attention of the semantic module in papers for readers.