Amplifying scientific paper’s abstract by leveraging data-weighted reconstruction

• Published in: Information processing & management 2016-07, Vol.52 (4), p.698-719
• Main Authors: Yang, Shansong, Lu, Weiming, Zhang, Zhanjiang, Wei, Baogang, An, Wenjia
• Format: Article
• Language: English
• Subjects: Abstracts; Amplification; Bibliographic literature; Bibliographies; Citation analysis; Data-weighted reconstruction; Document summarization; Reconstruction; Regularization; Scientific literature; Scientific papers; Sentences; Studies; Weight
• ISSN: 0306-4573; 1873-5371
• DOI: 10.1016/j.ipm.2015.12.014

•This paper explores the impact of heterogeneous bibliographic network for generating scientific paper’s amplified abstract.•The amplified abstract is generated by leveraging target scientific paper’s abstract and citation sentence’s content and structure, which is addressed through document summarization manner.•Sentence’s weight is learned by exploiting regularization for ranking on heterogeneous bibliographic network.•Data-weighted reconstruction is proposed to assign different priority to sentences when reconstructing the original document.•Various evaluation metrics are designed to validate the effectiveness of our approach. In this paper, we focus on the problem of automatically generating amplified scientific paper’s abstractwhich represents the most influential aspects of scientific paper. The influential aspects can be illustrated by the target scientific paper’s abstract and citation sentences discussing the target paper, which are provided in papers citing the target paper. In this paper, we extract representative sentences through data-weighted reconstruction approach(DWR) by jointly leveraging target scientific paper’s abstract and citation sentences’ content and structure. In our study, we make two-folded contributions. Firstly, sentence’s weight was learned by exploiting regularization for ranking on heterogeneous bibliographic network. Specially, Sentences-similar-Sentencesrelationship was identified by language modeling-based approach and added to the bibliographic network. Secondly, a data-weighted reconstruction objective function is optimized to select the most representative sentences which reconstructs the original sentence set with minimum error. In this process, sentences’ weight plays a critical role. Experimental evaluation over real dataset confirms the effectiveness of our approach.