Data Mining
 
This section contains the following topics:
Genetic Algorithms
*Chromosome
*GA_CHROMOSOME Function—Creates a data structure containing unencoded and encoded phenotype information.
*Individuals
*GA_INDIVIDUAL Function—Creates a data structure from user supplied phenotypes.
*GA_MUTATE Function—Performs the mutation operation on an individual’s chromosome.
*GA_DECODE Function—Decodes an individual’s chromosome into its binary, nominal, integer and real phenotypes.
*GA_ENCODE Function—Encodes an individual’s binary, nominal, integer and real phenotypes into its chromosome.
*Population
*GA_POPULATION Function—Creates a population data structure from user supplied individuals.
*GA_RANDOM_POPULATION Function—Creates a population data structure from randomly generated individuals.
*GA_GROW_POPULATION Function—Adds the individuals in the array individual to an existing population.
*GA_MERGE_POPULATION Function—Creates a new population by merging two populations with identical chromosome structures.
*Genetic Algorithm Search and Optimization
*GENETIC_ALGORITHM Function—Optimizes a user-defined fitness function using a tailored genetic algorithm.
Naive Bayes
*NAIVE_BAYES_TRAINER Function—Trains a Naive Bayes classifier.
*NAIVE_BAYES_CLASSIFICATION Function—Classifies unknown patterns using a previously trained Naive Bayes classifier.
Neural Networks
*Neural Network Data Structures
*MLFF_NETWORK_INIT Function—Creates a multilayered feedforward neural network.
*MLFF_NETWORK Function—Links and modifies a multilayered feedforward neural network.
*MLFF_INITIALIZE_WEIGHTS Function—Initializes weights for multilayered feedforward neural networks prior to network training using one of four user selected methods.
*Forecasting Neural Networks
*MLFF_NETWORK_TRAINER Function—Trains a multilayered feedforward neural network using quasi-Newton backpropagation.
*MLFF_NETWORK_FORECAST Function—Calculates forecasts using trained multilayered feedforward neural networks.
*Classification Neural Networks
*MLFF_CLASSIFICATION_TRAINER Function—Trains a multilayered feedforward neural network.
*MLFF_PATTERN_CLASSIFICATION Function—Trains a multilayered feedforward neural network for classification.
*Preprocessing Filters
*SCALE_FILTER Function—Scales or unscales continuous data prior to its use in neural network training, testing, or forecasting.
*TIME_SERIES_FILTER Function—Converts time series data to the format required for processing by a neural network.
*TIME_SERIES_CLASS_FILTER Function—Converts time series data sorted within nominal classes in decreasing chronological order to a useful format for processing by a neural network.
*UNSUPERVISED_NOMINAL_FILTER Function—Converts nominal data into a series of binary encoded columns for input to a neural network.
*UNSUPERVISED_ORDINAL_FILTER Procedure—Converts ordinal data into proportions.