kth.sePublications KTH
Operational message
There are currently operational disruptions. Troubleshooting is in progress.
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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
Automatisering av systematiska översikter med stora språkmodeller
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
2025 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
Automation of Systematic Reviews Using Large Language Models (English)
Abstract [sv]

Systematiska översikter är en central metod inom forskning men är ofta mycket tids- och resurskrävande. Detta är särskilt påtagligt inom snabbt växande områden som idrottsteknologi, där mängden nya publikationer ökar kontinuerligt. Syftet med detta examensarbete var att undersöka möjligheten att automatisera delar av processen genom att utveckla ett prototypverktyg som kan läsa, analysera och kategorisera vetenskapliga artiklar samt presentera resultaten i ett interaktivt gränssnitt. Prototypen byggdes med hjälp av lokalt körda språkmodeller, vilket säkerställer hög datasäkerhet, och ett anpassat analysramverk för att kategorisera artiklar utifrån population, sport, teknologi och utfall. Utvärderingen visade att verktyget fungerade väl för att identifiera sport och teknologier, medan population och utfall var mer utmanande att klassificera. Resultaten visar att ett sådant verktyg kan minska behovet av manuellt arbete, öka effektiviteten och ändå behålla tillräcklig noggrannhet. Arbetet bidrar därmed till utvecklingen av nya metoder för att stödja forskare inom idrottsteknologi och pekar på potentialen för vidareutveckling inom andra forskningsområden.

Abstract [en]

Systematic reviews are a central method in research but are often very time- and resource-intensive. This is particularly evident in rapidly growing fields such as sports technology, where the number of new publications is continuously increasing. The aim of this thesis was to investigate the possibility of automating parts of the process by developing a prototype tool that can read, analyze, and categorize scientific articles, as well as present the results in an interactive interface. The prototype was built using locally run language models, ensuring high data security, and a customized analysis framework to categorize articles based on population, sport, technology, and outcomes. The evaluation showed that the tool performed well in identifying sport and technologies, while population and outcomes were more challenging to classify. The results indicate that such a tool can reduce the need for manual work, increase efficiency, and still maintain sufficient accuracy. Thus, this work contributes to the development of new methods to support researchers in sports technology and highlights the potential for further development in other research areas.

Place, publisher, year, edition, pages
2025.
Series
TRITA-CBH-GRU ; 312
Keywords [en]
systematic reviews, sports technology, large language models, artificial intelligence, automation, text analysis, categorization, research tools
Keywords [sv]
systematiska översikter, idrottsteknologi, stora språkmodeller, artificiell intelligens, automatisering, textanalys, kategorisering, forskningsverktyg
National Category
Artificial Intelligence Computer Engineering
Identifiers
URN: urn:nbn:se:kth:diva-370596OAI: oai:DiVA.org:kth-370596DiVA, id: diva2:2001728
Subject / course
Computer Engineering with Business Economics
Educational program
Bachelor of Science in Engineering - Computer Engineering and Economics; Bachelor of Science in Engineering - Computer Engineering
Supervisors
Examiners
Available from: 2025-09-30 Created: 2025-09-27 Last updated: 2025-09-30Bibliographically approved

Open Access in DiVA

fulltext(1140 kB)51 downloads
File information
File name FULLTEXT01.pdfFile size 1140 kBChecksum SHA-512
ab9a8f914b11d89a15993ecd46891d7623c0fa9147260938c0eed92f2ad58c681abb876a8317ea2bd9ccf2db3ef849d70cc09c203d3aeebf0aeaaf2e0dff5a42
Type fulltextMimetype application/pdf

By organisation
Health Informatics and Logistics
Artificial IntelligenceComputer Engineering

Search outside of DiVA

GoogleGoogle Scholar
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: 302 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • 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