Package elki.datasource.filter.normalization.columnwise
Normalizations operating on columns / variates; where each column is treated independently.
-
Class Summary Class Description AttributeWiseBetaNormalization<V extends NumberVector> Project the data using a Beta distribution.AttributeWiseBetaNormalization.Par<V extends NumberVector> Parameterization class.AttributeWiseCDFNormalization<V extends NumberVector> Class to perform and undo a normalization on real vectors by estimating the distribution of values along each dimension independently, then rescaling objects to the cumulative density function (CDF) value at the original coordinate.AttributeWiseCDFNormalization.Par<V extends NumberVector> Parameterization class.AttributeWiseMADNormalization<V extends NumberVector> Median Absolute Deviation is used for scaling the data set as follows:AttributeWiseMeanNormalization<V extends NumberVector> Normalization designed for data with a meaningful zero:
The 0 is retained, and the data is linearly scaled to have a mean of 1, by projection with f(x) = x / mean(X).AttributeWiseMinMaxNormalization<V extends NumberVector> Class to perform and undo a normalization on real vectors with respect to a given minimum and maximum in each dimension.AttributeWiseMinMaxNormalization.Par<V extends NumberVector> Parameterization class.AttributeWiseVarianceNormalization<V extends NumberVector> Class to perform and undo a normalization on real vectors with respect to given mean and standard deviation in each dimension.AttributeWiseVarianceNormalization.Par<V extends NumberVector> Parameterization class.IntegerRankTieNormalization Normalize vectors according to their rank in the attributes.IntegerRankTieNormalization.Sorter Class to sort an index array by a particular dimension.InverseDocumentFrequencyNormalization<V extends SparseNumberVector> Normalization for text frequency (TF) vectors, using the inverse document frequency (IDF).