abstract |
An information computational method for classifying multivariate datasets to identify latent (unobservable) properties of members of a sample, which properties are then used for classification. The method comprises a novel combination of statistical and fuzzy logic methods whereby the latent classes of each object are identified according to the formula: n ƒ( j 1 , . . . , j K )|{( j k εS km jk } k=1 K ˜G[h ( k,j k ,{{S km } m=1 M k } k=1 K )] n wherein kε{ 1 , . . . , K} indexes the directions of the multidimensional space; j k ε{1 , . . . , N k } identifies an object in direction k; N k is the number of objects in principal direction k; j 1 , . . . , j K is a vector of one or more observations on a set of objects {j 1 , . . . , j K }; mε{ 1 , . . . , M k } indexes latent classes in direction k with M k being the number of latent classes in direction k; S km is a latent class m in direction k; G[·] is a specified univariate or multivariate distribution; f(·) and g(·) are specified functions; and the method calculates the likelihood that each object of interest belongs to each identified latent class. The invention addresses a variety of informatics problems, particularly in the field of biology, and permits a user to make reasonable inferences about underlying cause-effect relationships, such as the underlying biology of gene-expression patterns. |