Multivariate classification of BPSG thin films using Mahalanobis distances
1995
Infrared absorption spectra of borophosphosilicate glass (BPSG) thin films were collected to develop a rapid classification method for determining if the films are within the desired specifications. Classification of samples into good and bad categories was performed using principal component analysis applied to the spectra. Mahalanobis distances were used as the classification metric. The highest overall percentage of correct classification of samples based upon their spectra was 91.6%.
Keywords:
- Multivariate statistics
- Principal component analysis
- Thin film
- Absorption spectroscopy
- Mahalanobis distance
- Borophosphosilicate glass
- Infrared spectroscopy
- Statistics
- Analytical chemistry
- Materials science
- Pattern recognition
- classification methods
- multivariate classification
- semiconductor materials
- Artificial intelligence
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI