Abstract: This invention is a method of training a mean-field Bayesian data reduction algorithm (BDRA) based classifier which includes using an initial training for determining the best number of levels. The Mean-Field BDRA is then retrained for each point in a target data set and training errors are calculated for each training operation. Cluster candidates are identified as those with multiple points having a common training error. Utilizing these cluster candidates and previously identified clusters as the identified target data, the clusters can be confirmed by comparing a newly calculated training error with the previously calculated common training error for the cluster. The method can be repeated until all cluster candidates are identified and tested.

Abstract: An automatic feature selection system for test data with data (including the test data and/or the training data containing missing values in order to improve classifier performance. The missing features for such data are selected in one of two ways: first approach assumes each missing feature is uniformly distributed over its range of values whereas in the second approach, the number of discrete levels for each feature is increased by one for the missing features. These two choices modify the Bayesian Data Reduction Algorithm accordingly used for the automatic feature selection.

Abstract: A data reduction method for a classification system using quantized feature vectors for each class with a plurality of features and levels. The reduction algorithm consisting of applying a Bayesian data reduction algorithm to the classification system for developing reduced feature vectors. Test data is then quantified into the reduced feature vectors. The reduced classification system is then tested using the quantized test data. A Bayesian data reduction algorithm is further provided having by computing an initial probability of error for the classification system. Adjacent levels are merged for each feature in the quantized feature vectors. Level based probabilities of error are then calculated for these merged levels among the plurality of features. The system then selects and applies the merged adjacent levels having the minimum level based probability of error to create an intermediate classification system.

Abstract: A classification system uses sensors to obtain information from which feaes which characterized a source or object to classified can be extracted. The features are extracted from the information and compiled into a feature vector which is then quantized to one of one of M discrete symbols. After N feature vectors have been quantized, a test vector having components which are defined by the number of occurrences of each of the M symbols in N the quantized vectors is built. The system combines the test vector with training data to simultaneously estimate symbol probabilities for each class and classify the test vector using a decision rule that depends only on the training and test data. The system classifies the test vector using either a Combined Bayes test or a Combined Generalized likelihood ratio test.