LibLINEAR

Package

weka.classifiers.functions

Synopsis

A wrapper class for the liblinear tools (the liblinear classes, typically the jar file, need to be in the classpath to use this classifier).
Rong-En Fan, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang, Chih-Jen Lin (2008). LIBLINEAR - A Library for Large Linear Classification. URL http://www.csie.ntu.edu.tw/~cjlin/liblinear/.

Available in Weka 3.6.x - 3.7.1. Available via the package management system for Weka >= 3.7.2 (LibLINEAR)

Options

The table below describes the options available for LibLINEAR.

Option	Description
SVMType	The type of SVM to use.
bias	If >= 0, a bias term with that value is added; otherwise (<0) no bias term is added (default: 1).
convertNominalToBinary	Whether to turn on conversion of nominal attributes to binary.
cost	The cost parameter C.
debug	If set to true, classifier may output additional info to the console.
doNotReplaceMissingValues	Whether to turn off automatic replacement of missing values. WARNING: set to true only if the data does not contain missing values.
eps	The tolerance of the termination criterion.
normalize	Whether to normalize the data.
probabilityEstimates	Whether to generate probability estimates instead of -1/+1 for classification problems (currently for L2-regularized logistic regression only!)
weights	The weights to use for the classes, if empty 1 is used by default.

Capabilities

The table below describes the capabilites of LibLINEAR.

Capability	Supported
Class	Nominal class, Missing class values, Binary class
Attributes	Binary attributes, Nominal attributes, Date attributes, Empty nominal attributes, Unary attributes, Numeric attributes
Min # of instances	1