Machine Learning with BayesiaLab
BayesiaLab features a comprehensive array of highly optimized learning algorithms that can quickly uncover structures in datasets. The optimization criteria in BayesiaLab’s learning algorithms are based on information theory (e.g. the Minimum Description Length). With that, no assumptions regarding the variable distributions are made. These algorithms can be used for all kinds and all sizes of problem domains, sometimes including thousands of variables with millions of potentially relevant relationships.
Clustering in BayesiaLab covers both Data Clustering and Variable Clustering. The former applies to the grouping of records (or observations) in a dataset; the latter performs a grouping of variables according to the strength of their mutual relationships.
A third variation of this concept is of particular importance in BayesiaLab: Multiple Clustering can be characterized as a kind of nonlinear, nonparametric and nonorthogonal factor analysis. Multiple Clustering often serves as the basis for developing Probabilistic Structural Equation Models with BayesiaLab (see Chapter 8 in our book, Bayesian Networks and BayesiaLab).