Accession Number : ADA323964
Title : Learning Hidden Structure from Data: A Method for Marginalizing Joint Distributions Using Minimum Cross-Correlation Error.
Descriptive Note : Master's thesis,
Corporate Author : AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH
Personal Author(s) : Haynes, Antony K.
PDF Url : ADA323964
Report Date : 18 APR 1997
Pagination or Media Count : 99
Abstract : This thesis demonstrates an entropy-based, Bayesian dependency algorithm-Minimum Error Tree Decomposition II (METD2)-that performs computer-based generation of probabilistic hidden-structure domain models from a database of cases. The system learns probabilistic hidden-structure domain models from data, which partially automates the task of expert system construction and the task of scientific discovery. Existing probabilistic systems find associations among the observable variables but do not consider the presence of hidden variables, or else, do not use cross-correlation error as the metric for building the hidden structure. The algorithm decomposes a joint distribution of n observable variables into n+l observable and hidden variables. The hidden variable exists in the form of a tree consisting of n-l interior nodes. The final product of the procedure is a combined tree whose n leaves are the observable variables in a sample and whose n-l interior nodes are the marginalizations for the leaves.
Descriptors : *ALGORITHMS, *EXPERT SYSTEMS, *BAYES THEOREM, *CROSS CORRELATION, DATA BASES, MATHEMATICAL MODELS, NEURAL NETS, DATA MANAGEMENT, PROBABILITY DISTRIBUTION FUNCTIONS, LEARNING MACHINES, THESES, STATISTICAL SAMPLES, NONPARAMETRIC STATISTICS, ERROR ANALYSIS, SYSTEMS ANALYSIS, MARKOV PROCESSES.
Subject Categories : Cybernetics
Distribution Statement : APPROVED FOR PUBLIC RELEASE