Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Finding Implicit Citations in Scientific Publications: Improvements to Citation Context Detection Methods
KTH, School of Computer Science and Communication (CSC).
2015 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This thesis deals with the task of identifying implicit citations between scientific publications. Apart from being useful knowledge on their own, the citations may be used as input to other problems such as determining an author’s sentiment towards a reference, or summarizing a paper based on what others have written about it. We extend two recently proposed methods, a Machine Learning classifier and an iterative Belief Propagation algorithm. Both are implemented and evaluated on a common pre-annotated dataset. Several changes to the algorithms are then presented, incorporating new sentence features, different semantic text similarity measures as well as combining the methods into a single classifier. Our main finding is that the introduction of new sentence features yield significantly improved F-scores for both approaches.

Abstract [sv]

Detta examensarbete behandlar frågan om att hitta implicita citeringar mellan vetenskapliga publikationer. Förutom att vara intressanta på egen hand kan dessa citeringar användas inom andra problem, såsom att bedöma en författares inställning till en referens eller att sammanfatta en rapport utifrån hur den har blivit citerad av andra. Vi utgår från två nyliga metoder, en maskininlärningsbaserad klassificerare och en iterativ algoritm baserad på en grafmodell. Dessa implementeras och utvärderas på en gemensam förannoterad datamängd. Ett antal förändringar till algoritmerna presenteras i form av nya särdrag hos meningarna (eng. sentence features), olika semantiska textlikhetsmått och ett sätt att kombinera de två metoderna. Arbetets huvudsakliga resultat är att de nya meningssärdragen leder till anmärkningsvärt förbättrade F-värden för de båda metoderna.

Place, publisher, year, edition, pages
2015.
Keyword [en]
implicit citations, citation context, citations, natural language processing, nlp, machine learning, belief propagation
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-173913OAI: oai:DiVA.org:kth-173913DiVA: diva2:855986
External cooperation
SICS Swedish ICT
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2015-09-24 Created: 2015-09-22 Last updated: 2015-09-24Bibliographically approved

Open Access in DiVA

finding_implicit_citations_jonathan(978 kB)166 downloads
File information
File name FULLTEXT01.pdfFile size 978 kBChecksum SHA-512
3356762e92ae40343c1436b0ea3507b615b826c0446d19d322f78b365e66075e52731c62690b05287b30f138439262c1a3245df7be0e00ae4564421db2a3b7f2
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 166 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 217 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf