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
Data Mining of Trouble Tickets for Automatic Action Recommendation
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Systems and Control.
2017 (Swedish)Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

This work investigates the possibility of applying machine learning and data mining to the problem of finding solutions to software and hardware problems arising in telecommunication systems. Trouble ticket data is analyzed using traditional data mining techniques and more complex machine learning models, including neural networks, to find out which types of models are suitable for the task. Results show that there are relevant correlations in the data which enables the root cause to predicted with up to 90% accuracy in the case of the most common root cause, and up to 70% when classifying between up to 20 root causes. These predictive models could be used to assist the engineers by giving them probably suggestions for the root cause, and potentially save time in the troubleshooting process. Relatively simple data mining and linear models performed best which shows that this could be implemented in practice given that these methods are fast, robust, memory efficient and easy to implement. Neural networks were also investigated but gave no significant improvement in the results, athough there are indications that they could outperform linear models if more training data was available. A large fraction of the collected data could not be used for analysis because of missing values and other inconsistencies, which highlights the importance of defining standards for the data collection process. This would lead to higher quality data, and allow the trained models to be more general and perform well in multiple locations.

Place, publisher, year, edition, pages
2017. , p. 32
Series
UPTEC F, ISSN 1401-5757 ; 17012
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:uu:diva-316894OAI: oai:DiVA.org:uu-316894DiVA, id: diva2:1079257
External cooperation
Ericsson AB, DUIC S&T
Educational program
Master Programme in Engineering Physics
Supervisors
Examiners
Available from: 2017-03-08 Created: 2017-03-08 Last updated: 2018-01-13Bibliographically approved

Open Access in DiVA

fulltext(578 kB)470 downloads
File information
File name FULLTEXT01.pdfFile size 578 kBChecksum SHA-512
7aedcac04c9014a37655aa9a515fdd670e0bf7483c32ce6bd252f454ff05f1e4fa34949a06d8ca67a5a69174fa821902030ecee1e47a92f10b5e4e1416a60b90
Type fulltextMimetype application/pdf

By organisation
Division of Systems and Control
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 470 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: 525 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