Package weka.filters.unsupervised.attribute
package weka.filters.unsupervised.attribute
-
ClassesClassDescriptionAn 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.An instance filter that adds a new attribute to the dataset.A filter that adds a new nominal attribute representing the cluster assigned to each instance by the specified clustering algorithm.An instance filter that creates a new attribute by applying a mathematical expression to existing attributes.An instance filter that adds an ID attribute to the dataset.An instance filter that changes a percentage of a given attributes values.Adds the labels from the given list to an attribute if they are missing.Centers all numeric attributes in the given dataset to have zero mean (apart from the class attribute, if set).Changes the date format used by a date attribute.Filter that can set and unset the class index.A filter that uses a density-based clusterer to generate cluster membership values; filtered instances are composed of these values plus the class attribute (if set in the input data).An instance filter that copies a range of attributes in the dataset.An instance filter that discretizes a range of numeric attributes in the dataset into nominal attributes.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.A filter for detecting outliers and extreme values based on interquartile ranges.Converts the given set of predictor variables into a kernel matrix.A filter that creates a new dataset with a boolean attribute replacing a nominal attribute.Modify numeric attributes according to a given expressionMerges two values of a nominal attribute into one value.Converts the multi-instance dataset into single instance dataset so that the Nominalize, Standardize and other type of filters or transformation can be applied to these data for the further preprocessing.
Note: the first attribute of the converted dataset is a nominal attribute and refers to the bagId.Converts all nominal attributes into binary numeric attributes.Converts a nominal attribute (i.e.Normalizes all numeric values in the given dataset (apart from the class attribute, if set).A filter that 'cleanses' the numeric data from values that are too small, too big or very close to a certain value (e.g., 0) and sets these values to a pre-defined default.Converts all numeric attributes into binary attributes (apart from the class attribute, if set): if the value of the numeric attribute is exactly zero, the value of the new attribute will be zero.A filter for turning numeric attributes into nominal ones.Transforms numeric attributes using a given transformation method.A simple instance filter that renames the relation, all attribute names and all nominal (and string) attribute values.A filter that applies filters on subsets of attributes and assembles the output into a new dataset.Discretizes numeric attributes using equal frequency binning, where the number of bins is equal to the square root of the number of non-missing values.
For more information, see:
Ying Yang, Geoffrey I.This filter should be extended by other unsupervised attribute filters to allow processing of the class attribute if that's required.Performs a principal components analysis and transformation of the data.
Dimensionality reduction is accomplished by choosing enough eigenvectors to account for some percentage of the variance in the original data -- default 0.95 (95%).
Based on code of the attribute selection scheme 'PrincipalComponents' by Mark Hall and Gabi Schmidberger.Converts a propositional dataset into a multi-instance dataset (with relational attribute).Reduces the dimensionality of the data by projecting it onto a lower dimensional subspace using a random matrix with columns of unit length (i.e.Chooses a random subset of attributes, either an absolute number or a percentage.A propositionalization filter inspired by the RELAGGS algorithm.
It processes all relational attributes that fall into the user defined range (all others are skipped, i.e., not added to the output).A filter that removes a range of attributes from the dataset.Removes attributes of a given type.This filter removes attributes that do not vary at all or that vary too much.A filter that generates output with a new order of the attributes.Replaces all missing values for nominal and numeric attributes in a dataset with the modes and means from the training data.Standardizes all numeric attributes in the given dataset to have zero mean and unit variance (apart from the class attribute, if set).Converts a string attribute (i.e.Converts String attributes into a set of attributes representing word occurrence (depending on the tokenizer) information from the text contained in the strings.Swaps two values of a nominal attribute.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.An instance filter that assumes instances form time-series data and replaces attribute values in the current instance with the equivalent attribute values of some previous (or future) instance.A filter for wavelet transformation.
For more information see:
Wikipedia (2004).