ASMS 2009
Statistical Concepts
Document Actions
I will cover here only some special algorithms which can be used for metabolomics or structure elucidation of small molecules. This includes feature selection, classification and regression and some machine learning algorithms and genetic algorithms.
For machine learning purposes we usually apply diverse feature selection methods (sometimes based on the following 15 methods or genetic feature selection or PLS feature selection) and we use classification and regression methods from the following table (sometimes together with meta-learning concepts like bagging, boosting, voting, stacking).
Generalized Linear Models (GLM)
General Discriminant Analysis
Binary logit (logistic) regression
Binary probit regression
Nonlinear models Binary logit (logistic) regression
Binary probit regression
Multivariate adaptive regression splines (MARS)
Tree models Standard Classification Trees (CART)
Standard General Chi-square Automatic Interaction Detector (CHAID)
Exhaustive CHAID
Boosting classification trees
M5 regression trees
Neural Networks Standard General Chi-square Automatic Interaction Detector (CHAID)
Exhaustive CHAID
Boosting classification trees
M5 regression trees
Multilayer Perceptron
neural network (MLP)
Radial Basis Function neural network (RBF)
Machine Learning neural network (MLP)
Radial Basis Function neural network (RBF)
Support Vector Machines (SVM)
Naive Bayes classifier
k-Nearest Neighbors (KNN)
k-Nearest Neighbors (KNN)
Created by
zwluxx
Last modified 2008-07-09 07:54 PM
Last modified 2008-07-09 07:54 PM
