A Multi-classification Method of Temporal Data Based on Support Vector Machine

Authors:
Zhiqing Meng;Lifang Peng;Gengui Zhou;Yihua Zhu
Affiliations:
College of Business and Administration, Zhejiang University of Technology, Zhejiang 310023, China;Library, Hunan University of Technology, Zhuzhou, 412001, China;College of Business and Administration, Zhejiang University of Technology, Zhejiang 310023, China;College of Business and Administration, Zhejiang University of Technology, Zhejiang 310023, China
Venue:
Computational Intelligence and Security
Year:
2007

Citing 2
Cited 0

Logical design for temporal databases with multiple granularities

ACM Transactions on Database Systems (TODS)
Dynamic support vector machines for non-stationary time series forecasting

Intelligent Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper studies a multi-classification method based on support vector machine for temporal data. First, we give classic classification model of support vector machine. Then, we present a support vector machine model based on multi-weighted values, which is used to deal with multi-classification problems of temporal data. We define temporal type and prediction model for the temporal data. According to the temporal type model and the support vector machine model based on multi-weighted values, we propose a multi-classification method based on the support vector machine. Finally, experiments results show that our method can effectively solve the misclassification problems of temporal data.