Package weka.filters

Class Summary
AbstractTimeSeriesFilter An abstract instance filter that assumes instances form time-series data and performs some merging of attribute values in the current instance with attribute attribute values of some previous (or future) instance.
AddFilter An instance filter that adds a new attribute to the dataset.
AllFilter A simple instance filter that passes all instances directly through.
AttributeExpressionFilter Applys a mathematical expression involving attributes and numeric constants to a dataset.
AttributeFilter An instance filter that deletes a range of attributes from the dataset.
AttributeSelectionFilter Filter for doing attribute selection.
AttributeTypeFilter An instance filter that deletes all attributes of a specified type from the dataset.
CopyAttributesFilter An instance filter that copies a range of attributes in the dataset.
DiscretizeFilter An instance filter that discretizes a range of numeric attributes in the dataset into nominal attributes.
DynamicReductDiscretize  
EmptyAttributeFilter Removes all attributes that do not contain more than one distinct value.
Filter An abstract class for instance filters: objects that take instances as input, carry out some transformation on the instance and then output the instance.
FirstOrderFilter This instance filter takes a range of N numeric attributes and replaces them with N-1 numeric attributes, the values of which are the difference between consecutive attribute values from the original instance. eg: Original attribute values
0.1, 0.2, 0.3, 0.1, 0.3 New attribute values
0.1, 0.1, 0.1, -0.2, -0.2 The range of attributes used is taken in numeric order.
InstanceFilter Filters instances according to the value of an attribute.
MakeIndicatorFilter Creates a new dataset with a boolean attribute replacing a nominal attribute.
MergeTwoValuesFilter Merges two values of a nominal attribute.
NominalToBinaryFilter Converts all nominal attributes into binary numeric attributes.
NonSparseToSparseFilter A filter that converts all incoming instances into sparse format.
NormalizationFilter Normalizes all numeric values in the given dataset.
NullFilter A simple instance filter that allows no instances to pass through.
NumericToBinaryFilter Converts all numeric attributes into binary attributes (apart from the class attribute): if the value of the numeric attribute is exactly zero, the value of the new attribute will be zero.
NumericTransformFilter Transforms numeric attributes using a given transformation method.
ObfuscateFilter A simple instance filter that renames the relation, all attribute names and all nominal (and string) attribute values.
RandomizeFilter This filter randomly shuffles the order of instances passed through it.
ReplaceMissingValuesFilter Replaces all missing values for nominal and numeric attributes in a dataset with the modes and means from the training data.
ResampleFilter Produces a random subsample of a dataset.
RSDiscretize  
RSNumerizeNominal Creates numerical attributes of nominal ones, in case we wish to discretize them also
RSReplaceMissingValues Replaces all missing values for nominal and numeric attributes in a dataset with the modes and means from the training data.
SparseToNonSparseFilter A filter that converts all incoming sparse instances into non-sparse format.
SplitDatasetFilter This filter takes a dataset and outputs a subset of it.
SpreadSubsampleFilter Produces a random subsample of a dataset.
StringToNominalFilter Converts a string attribute (i.e. unspecified number of values) to nominal (i.e. set number of values).
SwapAttributeValuesFilter Swaps two values of a nominal attribute.
TimeSeriesDeltaFilter An instance filter that assumes instances form time-series data and replaces attribute values in the current instance with the difference between the current value and the equivalent attribute attribute value of some previous (or future) instance.
TimeSeriesTranslateFilter An instance filter that assumes instances form time-series data and replaces attribute values in the current instance with the equivalent attribute attribute values of some previous (or future) instance.