LIBLINEAR -- A Library for Large Linear Classification

Machine Learning Group at National Taiwan University
Contributors

Version 1.31 released on June 6, 2008. We now support another multi-class SVM formula (Crammer & Singer).

Introduction

LIBLINEAR is a linear classifier for data with millions of instances and features. It supports L2-regularized logistic regression (LR), L2-loss linear SVM, and L1-loss linear SVM.

The approach for L1-SVM and L2-SVM is a coordinate descent method:

C.-J. Hsieh, K.-W. Chang, C.-J. Lin, S. S. Keerthi, and S. Sundararajan. A dual coordinate descent method for large-scale linear SVM. ICML 2008.

For LR and L2-SVM, we implement a trust region Newton method:

C.-J. Lin, R. C. Weng, and S. S. Keerthi. Trust region Newton method for large-scale logistic regression. Journal of Machine Learning Research 9(2008), 627--650.

Code used for experiments in our LIBLINEAR papers can be found here.

Main features of LIBLINEAR include

Same data format as LIBSVM, our general-purpose SVM solver, and also similar usage
Multi-class classification: one-vs-the rest and Crammer & Singer
Cross validation for model selection
Probability estimates (logistic regression only)
Weights for unbalanced data
MATLAB/Octave interface

FAQ is here

When to use LIBLINEAR but not LIBSVM

There are some large data for which with/without nonlinear mappings gives similar performances. Without using kernels, one can train a much larger set via a linear classifier. Document classification is one such application. In the following example (20,242 instances and 47,236 features; available on LIBSVM data sets), the cross-validation time is significantly reduced by using LIBLINEAR:

% time libsvm-2.85/svm-train -c 4 -t 0 -e 0.1 -m 800 -v 5 rcv1_train.binary
Cross Validation Accuracy = 96.8136%
345.569s
% time liblinear-1.21/train -c 4 -e 0.1 -v 5 rcv1_train.binary
Cross Validation Accuracy = 97.0161%
2.944s

More details can be found in Appendix B of our SVM guide.

Download LIBLINEAR

The current release (Version 1.31, June 2008) of LIBLINEAR can be obtained by downloading the zip file or tar.gz file.

The package includes the source code in C/C++. A README file with detailed explanation is provided. For MS Windows users, there is a subdirectory in the zip file containing binary executable files.

Please read the COPYRIGHT notice before using LIBLINEAR.

Interfaces to LIBLINEAR

Language	Description	Maintainers and Their Affiliation	Supported LIBLINEAR version	Link
MATLAB	A simple MATLAB interface	LIBLINEAR authors at National Taiwan University.	The latest	Included in LIBLINEAR package
Octave	A simple Octave interface	LIBLINEAR authors at National Taiwan University.	The latest	Included in LIBLINEAR package

Please send comments and suggestions to Chih-Jen Lin.