Principal curves and surfaces are nonlinear generalizations of principal components and subspaces, respectively. They can provide insightful summary of high-dimensional data not typically attainable by classical linear methods. See paper for more and details. The matlab code supplied at the authors website calculates probabilistic principle surfaces on benchmark data sets.
Model Concept(s): Methods
Simulation Environment: MATLAB (web link to model)
Implementer(s): Chang, Kuiyu [kuiyu.chang at pmail.ntu.edu.sg]
References:
Chang KY, Ghosh J. (2001). A Unified Model for Probabilistic Principal Surfaces IEEE Transactions on Pattern Analysis and Machine Intelligence. 23