A Comparison of Linear Support Vector Machine Algorithms on Large Non-Sparse Datasets

Author

Lazar, Alina

Author_Institution

Dept. of Comput. Sci. & Inf. Syst., Youngstown State Univ., Youngstown, OH, USA

fYear

2010

fDate

12-14 Dec. 2010

Firstpage

879

Lastpage

882

Abstract

This paper demonstrates the effectiveness of Linear Support Vector Machines (SVM) when applied to non-sparse datasets with a large number of instances. Two linear SVM algorithms are compared. The coordinate descent method (LibLinear) trains a linear SVM with the L2-loss function versus the cutting-plane algorithm (SVMperf), which uses a L1-loss function. Four Geographical Information System (GIS) datasets with over a million instances were used for this study. Each dataset consists of seven independent variables and a class label which denotes the urban areas versus the rural areas.

Keywords

geographic information systems; pattern classification; support vector machines; L1-loss function; L2-loss function; LibLinear; SVMperf; coordinate descent method; cutting-plane algorithm; geographical information system datasets; linear classifier; linear support vector machine algorithm; nonsparse dataset; rural areas; urban areas; Accuracy; Classification algorithms; Geographic Information Systems; Machine learning; Machine learning algorithms; Support vector machines; Training; Support vector machine; linear classifier; unbalanced data;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Applications (ICMLA), 2010 Ninth International Conference on

Conference_Location

Washington, DC

Print_ISBN

978-1-4244-9211-4

Type

conf

DOI

10.1109/ICMLA.2010.137

Filename

5708960