Automatic extraction of citation contexts for research paper summarization

Kaplan, Dain; Iida, Ryu; Tokunaga, Takenobu

doi:10.3115/1699750.1699764

Cited by 35 publications

(23 citation statements)

References 18 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, Qazvinian and Radev (2008) argued that citation sentences (i.e., set of sentences that appear in other papers and cite a given article) are useful in creating a summary of important contributions of a research paper. Kaplan, Iida, and Tokunaga (2009) introduced "citation-site" as a block of text that includes a citation and discusses the cited text. This work used a machine learning method for extracting citations from research papers and evaluates the result using an annotated corpus of 38 papers citing 4 articles.…”

Section: Introductionmentioning

confidence: 99%

Generating Extractive Summaries of Scientific Paradigms

Qazvinian¹,

Radev²,

Mohammad³

et al. 2013

jair

View full text Add to dashboard Cite

Researchers and scientists increasingly find themselves in the position of having to quickly understand large amounts of technical material. Our goal is to effectively serve this need by using bibliometric text mining and summarization techniques to generate summaries of scientific literature. We show how we can use citations to produce automatically generated, readily consumable, technical extractive summaries. We first propose C-LexRank, a model for summarizing single scientific articles based on citations, which employs community detection and extracts salient information-rich sentences. Next, we further extend our experiments to summarize a set of papers, which cover the same scientific topic. We generate extractive summaries of a set of Question Answering (QA) and Dependency Parsing (DP) papers, their abstracts, and their citation sentences and show that citations have unique information amenable to creating a summary.

show abstract

Section: Introductionmentioning

confidence: 99%

Generating Extractive Summaries of Scientific Paradigms

Qazvinian¹,

Radev²,

Mohammad³

et al. 2013

jair

View full text Add to dashboard Cite

show abstract

“…Further, the occurrence of another citation is a good indicator that one citation ends and another begins (though this is not necessarily the case, see Ref. [27]). Utilising citing sentences from other papers citing the same target, in a lateral manner, we can find often cited concepts, i.e., lexical hooks [2], that act as indicators, such as a system name "STRAND", or a method "CRF"; this allows us to detect a citing sentence even if such a lexical hook was not present in one anchor sentence, as long as it is present in another.…”

Section: Entity Coherence Feature Setsmentioning

confidence: 99%

“…Reference [27] present an algorithmic approach to identifying citation blocks using coreference-chains. They report similar coverage issues related to coreference systems and cross-domain adaptation.…”

Section: Related Workmentioning

confidence: 99%

“…A recent study [2] shows that less than 25% of negative sentiment, and half of positive, are present in the citation block's anchor sentence, and other studies [27], [58] have suggested that up to half of all citation content is beyond the anchor sentence. The detection of citation blocks (e.g., sentences {(0),(1),(2)} in Fig.…”

Section: Introductionmentioning

confidence: 99%

“…There is a wealth of research from over the decades focusing on citations and citation analysis in various forms; this includes citation network analysis, like indexes [16], [18], bibliographic coupling [29], co-citation [56], citation counts [68], and the hindex [24], analysis of citation role/function [60], [67], analysis of sociological aspects [70], domain summarisation [15], [17], [44], [50], [52], paper summarisation [27], [51], and sentiment analysis [1], [43].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Citation Block Determination Using Textual Coherence

Kaplan

Tokunaga

Teufel

2016

Journal of Information Processing

Self Cite

View full text Add to dashboard Cite

Detecting the boundaries of citations in the running text of research papers is an important task for research paper summarisation, idea attribution, sentiment analysis, and other citation-based analysis research. Recently, detecting non-explicit citing sentences has garnered some attention, but can still be seen as in its infancy. We define this task as citation block determination (CBD). In this paper we propose and investigate the effects of various types of textual coherence on CBD, positing that it is a crucial aspect of identifying citation blocks, as it is fundamental to the composition of citations themselves. We demonstrate promising results, with our method outperforming previous state-of-the-art on F 1 by a large margin, with an improvement in both precision and recall, and further provide an in-depth error analysis and discussion of why this is the case.

show abstract

Measuring academic influence: Not all citations are equal

Zhu

Turney

Lemire

et al. 2014

Asso for Info Science & Tech

194

197

View full text Add to dashboard Cite

The importance of a research article is routinely measured by counting how many times it has been cited. However, treating all citations with equal weight ignores the wide variety of functions that citations perform. We want to automatically identify the subset of references in a bibliography that have a central academic influence on the citing paper. For this purpose, we examine the effectiveness of a variety of features for determining the academic influence of a citation.By asking authors to identify the key references in their own work, we created a dataset in which citations were labeled according to their academic influence. Using automatic feature selection with supervised machine learning, we found a model for predicting academic influence that achieves good performance on this dataset using only four features.The best features, among those we evaluated, were features based on the number of times a reference is mentioned in the body of a citing paper. The performance of these features inspired us to design an influence-primed h-index (the hip-index). Unlike the conventional h-index, it weights citations by how many times a reference is mentioned. According to our experiments, the hip-index is a better indicator of researcher performance than the conventional h-index.

show abstract

Automatic extraction of citation contexts for research paper summarization

Cited by 35 publications

References 18 publications

Generating Extractive Summaries of Scientific Paradigms

Generating Extractive Summaries of Scientific Paradigms

Citation Block Determination Using Textual Coherence

Measuring academic influence: Not all citations are equal

Contact Info

Product

Resources

About