Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • 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
Data-driven Natural Language Processing for IT Incident Classification: A Comparative Analysis of Machine Learning Models
2025 (English)Independent thesis Advanced level (degree of Master (One Year)), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

In enterprise IT environments, the accuracy and timeliness of incident classification are important to maintain operational efficiency and minimize service disruption. Manual ticket handling often leads to failures and inconsistencies due to human error. The study presents machine learning models -Logistic Regression, Random Forest, and XGBoost - for the automation of IT incident ticket classification based on natural language description. Using a dataset of 47,837 real-world incidents, TF-IDF was used for feature extraction, and the models were evaluated using accuracy, F1-score, and confusion matrices. As a step towards addressing some of the interpretability concerns of enterprise AI, Local Interpretable Model-Agnostic Explanations (LIME) are integrated to analyze the model's decisions at the individual prediction level. The study highlights the trade-offs between classification performance concerning explainability. While complex ensemble methods achieve predictive accuracy, they are not as transparent as simple models. The findings provide insights into how organizations can select and operationalize explainable NLP models in IT service management contexts.

Place, publisher, year, edition, pages
2025.
Keywords [en]
Explainable AI, IT incident, Natural Language Processing, XGBoost, Random Forest, Logistic Regression
National Category
Information Systems
Identifiers
URN: urn:nbn:se:hb:diva-34016OAI: oai:DiVA.org:hb-34016DiVA, id: diva2:1981479
Subject / course
Informatics
Available from: 2025-07-09 Created: 2025-07-04 Last updated: 2025-09-24Bibliographically approved

Open Access in DiVA

2025MAGI05(2204 kB)315 downloads
File information
File name FULLTEXT01.pdfFile size 2204 kBChecksum SHA-512
d82b537b958b803c6ffac4a031a318f56175efc6db67e6118d0b71e5ad9e24682c1143cbac1b8b18e754cc111dd5ec6fd1d3038d23b17db72071b2f0fbaed08f
Type fulltextMimetype application/pdf

Information Systems

Search outside of DiVA

GoogleGoogle Scholar
Total: 317 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: 435 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • harvard-cite-them-right
  • 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