Study of the neural network applied to weighted association rules mining
| Autor: | Xing-Ming Li, Tong-Yan Li |
|---|---|
| Rok vydání: | 2007 |
| Předmět: |
Association rule learning
Artificial neural network Time delay neural network business.industry Computer science Telecommunications service Sample (statistics) Machine learning computer.software_genre Set (abstract data type) ALARM Key (cryptography) Artificial intelligence Data mining business computer |
| Zdroj: | 2007 International Conference on Wavelet Analysis and Pattern Recognition. |
| DOI: | 10.1109/icwapr.2007.4420767 |
| Popis: | The mining of weighted association rules is one of the primary methods used in telecommunication alarm correlation analysis, of which weight set is a difficulty. In this study, we propose a novel method which uses neural network to identify the alarm weight. The neural network has three inputs with the key elements which reflect the importance of the telecommunication alarm. After the course of sample training, we will get the link weight. The weight of the neural network may reflect the knowledge of the experts and also can be changed automatically with the different items from the inputs. Modeling and simulation study indicate that compared with other methods of measuring alarm weight, the neural network method has more advantages. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|