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
Automating Bank Transaction Classification and Improving Liquidity Forecasts
Umeå University, Faculty of Science and Technology, Department of Mathematics and Mathematical Statistics.
2025 (English)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

In many organizations, manual processes can be streamlined through machine learning. This thesis introduces an ensemble machine learning approach to automate the classification of bank transactions, which is needed in order to create liquidity forecasts. A Random Forest model achieved a classification accuracy of 98.4%. The findings indicate that machine learning can effectively automate bank transaction classification. Additionally, statistical and machine learning models, SARIMA and XGBoost, were employed to forecast transaction types and compared against the forecasts of the existing modelused by BDX. The results indicate that SARIMA and XGBoost can enhance the accuracy of forecasts compared with the model currently in use. Although SARIMA and XGBoost require a larger evaluation time frame to determine their performance relative to the current model.

Abstract [sv]

Många organisationer kan effektivisera manuella processer med hjälp av maskininlärning. Denna rapport presenterar en ensemblebaserad maskininlärningsmetod för att automatisera klassificeringen av banktransaktioner, vilket behövs för att kunna skapa likviditetsprognoser. En Random Forest-modell uppnådde en klassificeringsnoggrannhet på 98.4%. Resultaten visar att maskininlärning effektivt kan automatisera klassificeringen av banktransaktioner. Vidare användes statistiska och maskininlärningsbaserade modeller, SARIMA och XGBoost, för att prognostisera 12 månader framåt för transaktionstyperna och jämfördes med den befintliga modellen som används av BDX. Resultatet visar att SARIMA och XGBoost kan ge bättre prognoser jämfört med modellen som BDX använder idag. Däremot krävs en längre tidsperiod för att utvärdera SARIMA och XGBoost och fastställa deras prestanda i förhållande till den nuvarande modellen som används idag. 

Place, publisher, year, edition, pages
2025. , p. 43
Keywords [en]
Machine learning, Supervised learning, Forecasting, Time series forecasting, Ensemble-models, Random Forest, SARIMA, XGBoost
National Category
Mathematical sciences
Identifiers
URN: urn:nbn:se:umu:diva-235833OAI: oai:DiVA.org:umu-235833DiVA, id: diva2:1939619
External cooperation
twoday INSIKT
Educational program
Master of Science in Engineering and Management
Supervisors
Examiners
Available from: 2025-02-24 Created: 2025-02-24 Last updated: 2025-02-24Bibliographically approved

Open Access in DiVA

fulltext(1164 kB)83 downloads
File information
File name FULLTEXT01.pdfFile size 1164 kBChecksum SHA-512
35082a8e1ca27764c51093555fd403373b215d2b959194a52dc461394c0113209f8980bcddf24c71544ceb6c10ab73df24d4a29f39c88c4dc93dcbff118243ad
Type fulltextMimetype application/pdf

By organisation
Department of Mathematics and Mathematical Statistics
Mathematical sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 83 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: 6347 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