PKIDiscretize
Package
weka.filters.unsupervised.attribute
Synopsis
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. Webb: Proportional k-Interval Discretization for Naive-Bayes Classifiers. In: 12th European Conference on Machine Learning, 564-575, 2001.
Options
The table below describes the options available for PKIDiscretize.
Option |
Description |
|---|---|
attributeIndices |
Specify range of attributes to act on. This is a comma separated list of attribute indices, with "first" and "last" valid values. Specify an inclusive range with "-". E.g: "first-3,5,6-10,last". |
bins |
Ignored. |
desiredWeightOfInstancesPerInterval |
Sets the desired weight of instances per interval for equal-frequency binning. |
findNumBins |
Ignored. |
ignoreClass |
The class index will be unset temporarily before the filter is applied. |
invertSelection |
Set attribute selection mode. If false, only selected (numeric) attributes in the range will be discretized; if true, only non-selected attributes will be discretized. |
makeBinary |
Make resulting attributes binary. |
useEqualFrequency |
Always true. |
Capabilities
The table below describes the capabilites of PKIDiscretize.
Capability |
Supported |
|---|---|
Class |
Relational class, Numeric class, Binary class, No class, Empty nominal class, Missing class values, Unary class, Nominal class, String class, Date class |
Attributes |
Binary attributes, String attributes, Nominal attributes, Missing values, Unary attributes, Relational attributes, Empty nominal attributes, Numeric attributes, Date attributes |
Min # of instances |
0 |