A Comparative study of Artificial Neural Networks Using Reinforcement learning and Multidimensional Bayesian Classification Using Parzen Density Estimation for Identification of GC-EIMS Spectra of Partially Methylated Alditol Acetates
This work addresses the problem of identifying complex chemical spectra for researchers in analytical chemistry, but it is incremental as it compares existing methods on a specific dataset.
The study developed a web-based pattern recognition system for identifying gas chromatography-electron impact mass spectra of partially methylated alditol acetates, comparing a Bayesian classifier with Parzen density estimation and an artificial neural network trained with reinforcement learning. Results showed both systems performed well with low noise, but the neural network was better at handling missing data, while performance degraded with lower signal-to-noise ratios.
This study reports the development of a pattern recognition search engine for a World Wide Web-based database of gas chromatography-electron impact mass spectra (GC-EIMS) of partially methylated Alditol Acetates (PMAAs). Here, we also report comparative results for two pattern recognition techniques that were employed for this study. The first technique is a statistical technique using Bayesian classifiers and Parzen density estimators. The second technique involves an artificial neural network module trained with reinforcement learning. We demonstrate here that both systems perform well in identifying spectra with small amounts of noise. Both system's performance degrades with degrading signal-to-noise ratio (SNR). When dealing with partial spectra (missing data), the artificial neural network system performs better. The developed system is implemented on the world wide web, and is intended to identify PMAAs using submitted spectra of these molecules recorded on any GC-EIMS instrument. The system, therefore, is insensitive to instrument and column dependent variations in GC-EIMS spectra.