Files in this item



application/pdfZhang147.pdf (5MB)
(no description provided)PDF


application/octet-streamZhang147.epub (336kB)
(no description provided)Unknown


Title:Synopsizing “literature review” for scientific publications
Author(s):Zhang, Lin; Liu, Xiaozhong
Subject(s):citation clustering
query-centric similarity
citation summarization
citation context
publication representation
Abstract:Because the number of scientific publications in most disciplines is expanding rapidly, traditional academic search engines can hardly satisfy scholars’ need to retrieve and assimilate the information they are looking for. In this study we investigate a new summarization problem: creating a synopsis “Literature Review” of a collection of candidate cited papers in response to a query, via different methods and indicators. In more detail, we compare the use of different methods and indicators for generating citation clusters and summarized reviews by analyzing publication abstracts, citation contexts, and co-cite relationships. We also validate the usefulness of a user’s query during this process by comparing query-dependent and query-independent clustering and summarization. One interesting outcome of this study is that citation contexts are more suitable for clustering related papers, whereas abstracts are more accurate for generating longer review-like summaries. The initial user query is also helpful for enhancing clustering and summarization performance.
Issue Date:2016-03-15
Citation Info:NA
Series/Report:IConference 2016 Proceedings
Genre:Conference Paper / Presentation
Rights Information:Copyright 2016 is held by the authors. Copyright permissions, when appropriate, must be obtained directly from the authors.
Date Available in IDEALS:2016-03-08

This item appears in the following Collection(s)

Item Statistics