Pattern Recognition Methods for the Classification of Binary Infrared Spectral Data
- 1 March 1976
- journal article
- Published by SAGE Publications in Applied Spectroscopy
- Vol. 30 (2) , 213-216
- https://doi.org/10.1366/000370276774456246
Abstract
Five pattern recognition methods are compared for their ability to classify binary infrared spectra. Included is a discussion of the time vs success balance for each of the techniques. Predictive ability decreases in the order maximum likelihood > distance > Tanimoto similarity ∼ Hamming distance > dot product. The time required for each prediction after the classifier has been developed increases in order maximum likelihood ∼ distance ∼ dot product < Tanimoto similarity ∼ Hamming distance.Keywords
This publication has 16 references indexed in Scilit:
- Automatic identification of chemical spectra. Goodness of fit measure derived from hypothesis testingAnalytical Chemistry, 1975
- A Comparison of Two Discriminant Functions for Classifying Binary Infrared DataApplied Spectroscopy, 1975
- Bayesian decision theory applied to the multicategory classification of binary infrared spectraAnalytical Chemistry, 1974
- Interpretation of steroid mass spectra with pattern recognition methodsChromatographia, 1974
- Heuristic pattern recognition analysis of carbon-13 nuclear magnetic resonance spectraJournal of the American Chemical Society, 1974
- Statistical method for the prediction of matching results in spectral file searchingAnalytical Chemistry, 1974
- Information content of mass spectra as determined by pattern recognition methodsAnalytical Chemistry, 1974
- Computer techniques for identifying low resolution mass spectraAnalytical Chemistry, 1971
- Matching of mass spectra when peak height is encoded to one bitAnalytical Chemistry, 1970
- Computerized learning machines applied to chemical problems. Convergence rate and predictive ability of adaptive binary pattern classifiersAnalytical Chemistry, 1969