Creating the faceted citation summary of a research paper involves identifying cited texts for citation sentences (i.e., citances), classifying their discourse facets, and generating a structured summary from the cited texts. This paper proposes a supervised method for the first two tasks by classification techniques. The first task uses binary classification to distinguish relevant pairs of citances and reference sentences from irrelevant pairs. The second task applies multi-class classification to assign one of the predefined discourse facets to relevant pairs of the first task. The proposed method is evaluated using the CL-SciSumm 2016 datasets and found to be competitive in producing superior results compared to state-of-the-art methods.
All Science Journal Classification (ASJC) codes
- Human-Computer Interaction
- Hardware and Architecture
- Library and Information Sciences
- Computational Theory and Mathematics