Bachelor and Master Theses

Title: A Framework for Data Mining and Events Correlation on Telecommunication System Logs
Subject: Computer Science
Level: Advanced
Description: With ever increasing complexity of modern computer systems, large volumes of system events has become a rising problem. These logs are to an increasing extent consisting of concurrent events. The thesis aims to investigate how to apply data mining technology in system log analysis without relying on timestamps. The purpose of this thesis is to set up a framework which applies data mining technology on system logs to find event patterns and then performs pattern matching and correlating on system logs. Several typical data mining techniques will be adopted and evaluated in this thesis work. This work could also facilitate the understanding of system behaviour and the research of automatic troubleshooting system from a long term perspective.

Thesis Goals

  • Get acquainted with state-of-the-art of data mining and event correlation in log analysis
  • Get acquainted with T&E(Trace & Error) log collection on CPP platform
  • Select appropriate tools and algorithms that applies to the thesis problem statement.
  • Process different log types using pattern matching technologies
  • Create a proof of concept toolset

Company: Ericsson AB, kontaktperson: Alf Larsson
Prel. end date: 2011-06-16
Presentation date: 2011-06-16
Student: Zhixiang Gao
Student: Xiaoyan Wan
IDT supervisor: Daniel Flemström, +46-21-151760
Examinator: Jan Carlson
Jan Carlson, +46-21-151722

  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2018.08.31