Construction and Aggregation of Citation Semantic Link Network

Semantic Link Network (SLN) is an extension of the hyperlink network by attaching semantic factors on links.In this paper, we propose a framework of constructing citation semantic link network (CSLN) which extends semantics of typical citation network through integrating several natural language processing techniques including citation function classification, sentiment analysis and keyword extraction. A concrete case of CSLN in opinion mining discipline is studied. Based on the exploration of CSLN from multi-perspective, we can effectively find articles with high importance, detect opinion communities and discover emergent topics among scientific documents.