Quality Engineering Using Robust Design
Quality Engineering Using Robust Design
Fast Training of Support Vector Machines for Regression
IJCNN '00 Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN'00)-Volume 5 - Volume 5
A tutorial on support vector regression
Statistics and Computing
An overview of statistical learning theory
IEEE Transactions on Neural Networks
| Hi-index | 0.00 |
This paper investigated the ability of the etched wall angle and depth controllable. The silicon plates with a patterned wet film photo resistance as a base substrate are used to demonstrate this research. The reactive ion etching (RIE) is main process for hard disk drive slider fabrication. This process is much more complicated to set its parameters to the slider with the right customer specification. Therefore, this paper presents a hybrid response surface methodology (RSM) based on robust parameter design (RPD) concept and data mining (DM) for the multiresponse optimization of a RIE process. The approach firstly, a designed experiment (DOE) was employed to collect the process data and to indicate the critical parameters of the process. Then, support vector regression (SVR) was used to establish the nonlinear multivariate relationships between process parameters and responses. Data obtained from DOE were used in the training process. Last but not least, the regression decision tree and the domain engineering knowledge were opted for the initial point of optimization algorithm as well. Finally, the reduced gradient search algorithm, a hill-climbing procedure and desirability function were adapted to the DOE Model. While grid search and desirability function were adapted to the SVR model to find the optimum parameter setting. The technique with the highest prominent accuracy performance was selected to build a RIE process model which is SVR. As a result, the optimum condition from the final model is effectively enabled to apply in the real production based on its confirmation experiment.