The Pattern Analysis Library (PALib) is a pattern classification/recognition library for C++ programmers. The library consists of numerical and statistical routines which range from statistical decision theory, parametric and non-parametric learning algorithms, and linear classification to supervised and unsupervised machine learning methods. PALib is portable across all platforms which support the ANSI-compliant C++ language and STL.

The library covers the following areas:

* Bayesian Decision Theory.
* Information Theory.
* Linear Programming.
* Linear Discriminant Functions.
* Support Vector Machines.
* Hidden Markov Model.
* Bayesian Networks.
* Dempster-Shafer Theory.
* Principal Component Analysis.
* Independent Component Analysis.
* Maximum Likelyhood Estimation.
* Expectation-Maximization Algorthim.
* Parzen Window.
* k-Nearest Neighbor Estimation.
* k-Means Clustering.
* Multi-Layer Perceptrons.
* Radial Basis Function Networks.
* Boltzmann Networks.
* Genetic Algorithms.
* Genetic Programming.
* Kalman Filtering.
* Decision Trees.
* Expert Systems.
* Fuzzy Inference Systems.