Digitala Vetenskapliga Arkivet

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
Developing an AI for Cedervall AB
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

This thesis explores the development and implementation of AI to extract accurate information from Cedervall's existing tenders. Using Microsoft's Azure platform, a Large Language Model (LLM) was built and for security reasons the data was then stored in Azure blob storage. A prototype AI model, using vector embedding was developed. Alongside that an application was developed in order to find suitable architects for tenders via a filtering algorithm. The project progressed through tasks, starting with small scale AI training and expanding to larger datasets, including tests with Azure's pre built bots. The CV application, designed with Python's Tkinter framework which ended up helping Cedervall with their tendering processes. Results demonstrate the AI's capability to provide smarter, more accurate responses for tenders. The CV application was also deployed and used by Cedervall for their employee filter process.

Abstract [sv]

Detta examensarbete innefattar utvecklingen och implementeringen av AI för att extrahera korrekt information från Cedervalls befintliga anbud. I Microsofts Azure-plattform byggdes en Large Language Model (LLM) och för säkerhetsskäl lagrades sedan all data i Azure blob lager. En prototyp AI-modell, tillsammans med vektorinbäddning utvecklades. ingenjörskonst, utvecklades. Vid sidan om utvecklades en applikation för att hitta lämpliga arkitekter för anbud via en filtreringsalgoritm. Projektet genomfördes genom att dela upp den i mindre delar, började med småskalig AI upplärning som sedan gick över till större datauppsättningar, inklusive tester med Azures förbyggda bots. CV ansökan, designades med Pythons Tkinter ramverk, som resulterade i att hjälpa Cedervall med deras anbuds processer. Resultaten visar AI:s förmåga att ge smartare, mer exakta svar för anbuden. CV applikationen lanserades och använts för att filtrera medarbetarna.

Place, publisher, year, edition, pages
2024. , p. 49
Series
TRITA-EECS-EX ; 2024:925
Keywords [en]
Large language model, Machine learning, Deep learning, Neural network, Vector embedding
Keywords [sv]
Storspråk modell, Maskininlärning, djupinlärning, neuralt nätverk, vektorinbäddning.
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-361052OAI: oai:DiVA.org:kth-361052DiVA, id: diva2:1943536
External cooperation
Cedervall AB
Supervisors
Examiners
Available from: 2025-03-17 Created: 2025-03-11 Last updated: 2025-03-17Bibliographically approved

Open Access in DiVA

fulltext(1330 kB)29 downloads
File information
File name FULLTEXT01.pdfFile size 1330 kBChecksum SHA-512
293332cf3c6068bf0d6fc7f5eb5d65bf2b395310dfd0dfb7b3f1c31c929ed50b29b429c5ad03f38bca602ef580f70115202c6fe9835de7afe9b125d080b41845
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 29 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: 331 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