| dc.creator |
OLGUN, Mehmet Onur |
|
| dc.creator |
ÖZDEMİR, Gültekin |
|
| dc.date |
2012-05-31T21:00:00Z |
|
| dc.date.accessioned |
2020-10-06T10:32:17Z |
|
| dc.date.available |
2020-10-06T10:32:17Z |
|
| dc.identifier |
736666e5-292f-4849-94b0-5f1756ad861c |
|
| dc.identifier |
https://avesis.sdu.edu.tr/publication/details/736666e5-292f-4849-94b0-5f1756ad861c/oai |
|
| dc.identifier.uri |
http://acikerisim.sdu.edu.tr/xmlui/handle/123456789/63435 |
|
| dc.description |
Shewhart control charts for statistical process control are important tools to examination abnormal changes in a process. Artificial Neural Networks and Bayesian pattern recognition systems are formed to identify patterns of abnormal changes in a process to identify changes that may occur over time, to keep a process under control and to take necessary actions in a process. Classification performance of the generated pattern recognizers was measured. Six statistical features are issued from observations, that patterns were created, and classification performances were compared to improve the performance of correct classification. It is observed that Artificial Neural Networks and Bayesian pattern recognizers have higher performance after related features are defined. In conclusion, it is concluded that Bayesian pattern recognizer has better classification performance than artificial neural networks. Bayesian classifier can be used in real-time control charts for pattern recognition applications. |
|
| dc.language |
tur |
|
| dc.rights |
info:eu-repo/semantics/closedAccess |
|
| dc.title |
CONTROL CHART PATTERN RECOGNITION USING STATISTICAL-FEATURE BASED BAYES CLASSIFIER |
|
| dc.type |
info:eu-repo/semantics/article |
|