
We have seen the Minimum Message Length (MML) and other treatments
of "simple" hypotheses over simple data such as finite choices,
e.g., multistate,
and continuous values,
e.g., normal distribution, etc..
It turns out that structured hypotheses
can be formed from combinations of simple hypotheses
plus a component to describe the particular structure.
Structured Models:
 Sequences, timeseries, grids, images, ...
 Hidden Markov Models.
 HMM & PFSA.
 Segmentation.
 Bioinformatics, DNA, proteins, genes, evolutionary trees.
 Alignment and
 Phylogenetic (Evolutionary) Trees.
 Supervised Learning,
expert systems, regressions, rule learning, ...
 DecisionTrees (ClassificationTrees),
supervised classification, and
 DecisionGraphs.
 Bayesian Networks, Causal Models, Graphical Models, ...
 Mixed Bayes Nets 
discrete & continuous (& structured) variables.
 Hybrid Models (local structure).
 Loglinear analysis,
Chordalysis MML
 ANNs, Artificial Neural Networks.
 Unsupervised Learning,
clustering, Snob, numerical taxonomy, rule discovery.
 Mixture Models,
unsupervised classification, clustering,
 in series,
 Stats. & Comp..
 Factor Analysis Models,
 Single Factor, &
 Multiple Factors.
 WB68, WF87.
 Inductive Programming (IP)
 ACSC03.
 II 1.0.
 JFP.
 ACSC06.
 ActaOe.
 IP 1.2, TR 224.
 Trees & Graphs.

