The CAIMAN method (Classification And Influence Matrix ANalysis) is a classification method exploiting the properties of the diagonal terms of the influence matrix, also called leverages. Depending on the classification problem, Caiman method can be used in three different ways:

 discriminant classifier (D-Caiman), i.e. as the classical classification methods, able to assign each object to one among the available classes;
 class modelling classifier (M-Caiman), which is also able to reject objects, i.e. objects can be evaluated as belonging to none of the available classes, or objects can be considered as confused, i.e. potentially belonging to at least two classes and thus not assigned;
 asymmetric classifier (A-Caiman), where the method is focused in modelling a selected class (class A) as distinct from all the other objects not belonging to that class (Not-A).
If you use the Caiman tool we would appreciate a reference to the following paper:

R.Todeschini, D.Ballabio, V.Consonni, A.Mauri and M.Pavan
CAIMAN (Classification and Influence Matrix Analysis): a new approach to the classification based on leverage-scaled functions.
Chemom.Intell.Lab.Syst., 87 (2007) 3-17

You can freely download the modules of Caiman. All the three methods are available as MATLAB modules: DMCaiman (D-Caiman and M-Caiman) and ACaiman (A-Caiman).

download caiman here

All the CAIMAN modules are structured into a GUI interface (see the picture below), in order to make the calculations simplier. In this interface now it's also possible to perform Variable Selection and Discriminant Analysis. A help file (in two formats: caiman.chm and caiman.htm) is also provided . In this file you can find:
 a full explanation about the CAIMAN classification approach
 overview on the classification
 overview on the most common methods