Genetic algorithms and support vector machines for time series classification

被引:55
作者
Eads, D [1 ]
Hill, D [1 ]
Davis, S [1 ]
Perkins, S [1 ]
Ma, JS [1 ]
Porter, R [1 ]
Theiler, J [1 ]
机构
[1] Los Alamos Natl Lab, Nonproliferat & Int Secur Div, Los Alamos, NM 87545 USA
来源
APPLICATIONS AND SCIENCE OF NEURAL NETWORKS, FUZZY SYSTEMS, AND EVOLUTIONARY COMPUTATION V | 2002年 / 4787卷
关键词
time series classification; genetic algorithm; genetic programming; support vector machines; feature selection; lightning; tornado; n-fold cross validation;
D O I
10.1117/12.453526
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We introduce an algorithm for classifying time series data. Since our initial application is for lightning data, we call the algorithm Zeus. Zeus is a hybrid algorithm that employs evolutionary computation for feature extraction, and a support vector machine for the final "backend" classification. Support vector machines have a reputation for classifying in high-dimensional spaces without overfitting, so the utility of reducing dimensionality with an intermediate feature selection step has been questioned. We address this question by testing Zeus on a lightning classification task using data acquired from the Fast On-orbit Recording of Transient Events (FORTE) satellite.
引用
收藏
页码:74 / 85
页数:12
相关论文
共 18 条
[1]  
ABRAHART RJ, 1999, GEOCOMPUTATION 99
[2]  
[Anonymous], 1998, Genetic programming: an introduction
[3]  
[Anonymous], LIBSVM LIB SUPPORT V
[4]   On domain knowledge and feature selection using a support vector machine [J].
Barzilay, O ;
Brailovsky, VL .
PATTERN RECOGNITION LETTERS, 1999, 20 (05) :475-484
[5]  
BRILES S, 1993, P INT WORKSH ART INT
[6]  
BRUMBY SP, 2000, P SPIE, V4049
[7]   A tutorial on Support Vector Machines for pattern recognition [J].
Burges, CJC .
DATA MINING AND KNOWLEDGE DISCOVERY, 1998, 2 (02) :121-167
[8]  
CAO LJ, 2000, FEATURE SELECTION SU
[9]  
Cristianini N, 2000, Intelligent Data Analysis: An Introduction
[10]  
Hart P.E., 1973, Pattern recognition and scene analysis