International Conference on Information Technology and Computer Science, 3rd (ITCS 2011)
41 Application of Data Mining in Alarm Processing of Communication Network
Download citation file:
- Ris (Zotero)
- Reference Manager
As the telecommunication network grows rapidly in scale and complexity, it produces daily a large amount of alarm data which contains hidden valuable information. The information can be used in finding problems and possibly in predicting severe faults. The supervisors of network are finding it increasingly difficult to cope with the volume of alarm data. Data mining plays an important role in the alarm processing. In this paper, alarm correlation rules are discovered from the alarm databases of telecommunication network management systems by using technology of association rule mining and alarm sequence analysis. Aprior Model are used to discover the information. This paper proposes a multi-layer filter mechanism for data cleaning. The mechanism not only discards redundant alarm data, but also solves the time synchronization problem exists in the original alarms, and greatly reduces the volume of alarm data and the number of rules in the results. The proposed mining algorithm and filter mechanism is implemented and applied to test against a set of real alarm data provided by a mobile company. The experiment results show that the proposed mechanism of data cleaning are able to improve the execution of the mining algorithm significantly and the information obtained from the alarm data is of great importance from the perspective of network supervisors for alarm control and alarm prediction.