Improved Probabilistic Neural Network Algorithm for Chemical Sensor Array Pattern Recognition

Abstract

An improved probabilistic neural network (IPNN) algorithm for use in chemical sensor array pattern recognition applications is described. The IPNN is based on a modified probabilistic neural network (PNN) with three innovations designed to reduce the computational and memory requirements, to speed training, and to decrease the false alarm rate. The utility of this new approach is illustrated with the use of four data sets extracted from simulated and laboratory-collected surface acoustic wave sensor array data. A competitive learning strategy, based on a learning vector quantization neural network, is shown to reduce the storage and computation requirements. The IPNN hidden layer requires only a fraction of the storage space of a conventional PNN. A simple distance-based calculation is reported to approximate the optimal kernel width of a PNN. This calculation is found to decrease the training time and requires no user input. A general procedure for selecting the optimal rejection threshold for a PNN-based algorithm using Monte Carlo simulations is also demonstrated. This outlier rejection strategy is implemented for an IPNN classifier and found to reject ambiguous patterns, thereby decreasing the potential for false alarms.

Study of probabilistic neural networks to classify the active compounds in medicinal plants

“…The selection of this value is crucial because it determines the shape of the Gaussian function. A large radius possesses a smooth shape and has the advantage of interpolation, and a small radius leads to a sharp shape and reduces the overlap between adjacent samples [21]. But too small a spread cannot generalize well, because unknown samples only lie in the region that Gaussian function enclosing can be generalized.…”

A full-scale prototype multisensor system for fire detection and situational awareness

“…Naval Research Laboratory (NRL) have demonstrated significant improvements in the accuracy, sensitivity and response times in fire and smoke detection using multicriteria approaches that utilize probabilistic neural network algorithms to combine data from various fire sensors. 5,6 Other efforts at NRL have confirmed the viability of using a multisensor, multicriteria approach with data fusion for the detection of chemical agents 7,8 and unexploded ordinance. 9,10 Likewise, multisensor detection systems have shown a number of advantages over comparable single sensor systems for the detection of chemical weapons agents and toxic industrial chemicals (CWA/TIC), as evidenced by a number of successful, commercially available, multisensor-based detection systems for CWA/TIC applications.…”

A comparison of the performance of statistical and fuzzy algorithms for unexploded ordnance detection

“…The probabilistic neural network (PNN) is based upon Bayes' classification method [47], [49]- [51]. The basis of the classification method is given in the following equation, where and are the prior probabilities for class and and are the costs of misclassification, and and are the true probability density functions:…”