Computer Manual to accompany Pattern Classification

What can I do with the Computer Manual

  • Classify two dimensional, two class, data using a graphic user interface. This allows easy access to any of a large number of classification algorithms, clustering methods, and dimentionality reduction algorithms.
  • Enjoy easy access to over 50 classification algorithms and over 20 clustering and feature reduction methods.
  • Compare the performance of several algorithms on your data.
  • Find the best parameters for running an algorithm on your data.
  • Generate your own datasets either by entering their parameters (For Gaussian and Uniform distributions) or through a graphic interface.
  • View how clustering algorithms progress through their training.
  • Make nice figures of your results...

    • List of algorithms

    Classification algorithms

  • Ada Boost.
  • Backpropagation neural network (Five variants)
  • Balanced Winnow.
  • Perceptron (Several variants)
  • Relaxation (Several variants)
  • Bayesian Model Comparison
  • C4.5.
  • Cascade-Correlation type neural network.
  • Classification and regression trees (CART).
  • Component classifiers with descriminant functions.
  • Component classifiers without descriminant functions
  • Deterministic Boltzmann classifier.
  • Discrete Bayes classifier.
  • Expectation-maximization (EM).
  • Genetic algorithm.
  • Genetic programming.
  • Gibbs algorithm.
  • ID3.
  • Interactive Learning (Learning with queries).
  • Linear Least squares (LS).
  • Least-mean squares (LMS).
  • Local polynomial fitting.
  • Local boosting
  • LVQ1.
  • LVQ3.
  • Marginalization
  • Maximum likelihood (ML).
  • Maximum likelihood model comparison (ML2).
  • Multivariate adaptive regression splines.
  • Nearest Neighbors.
  • Nearest Neighbor Editing.
  • Normal density discriminant function (NDDF).
  • Optimal brain surgeon.
  • Parzen windows.
  • Probabilistic neural network.
  • Projection pursuit regression.
  • Recurrent neural network
  • Radial basis function network
  • Reduced coulomb energy algorithm (RCE).
  • Regularized descriminant analysis (RDA).
  • Store-Grabbag.
  • Stumps.
  • Support-vector machines (SVM).
  • Voted perceptron.


    • Clustering algorithms

  • Agglomerative clustering (ADDC).
  • Agglomerative hierarchical clustering algorithm (AGHC).
  • Basic iterative MSE clustering (BIMSEC).
  • Basic leader-follower clustering.
  • Competitive learning.
  • Deterministic annealing (DA)
  • Distinction sensitive learning vector quantization (DSLVQ).
  • k-Means.
  • Kohonen self-organizing feature maps (Kohonen_SOFM).
  • Fuzzy k-means.
  • Fishers' linear discriminant.
  • LVQ1
  • LVQ3
  • Minimum spanning tree.
  • Principal component analysis (PCA), also known as Karhunen-Louve transform (KLA).
  • Stepwise optimal hierarchical clustering (SOHC).
  • Stochstic simulated annealing (SA).


    • Feature selection and feature reduction algorithms

  • Culling genetic algorithm.
  • Hierarchical dimensionality reduction (HDR).
  • Independent component analysis (ICA).
  • Multidimensional scaling (MDS).
  • Non-linear principle component analysis (NLPCA).
  • Principle component analysis (PCA).
  • Information-based selection.