Proceedings of the 2007 International Conference on Intelligent Systems and Knowledge Engineering (ISKE 2007)
Session: Support Vector Machine
14 articles
Proceedings Article
Learn from the Information Contained in the False Splice Sites as well as in the True Splice Sites using SVM
Shuo Xu, Fujing Ma, Lan Tao
In splice sites prediction, the information contained in false splice sites is often ignored, which has been recognized to be very valuable [1]-[2]. In this paper, three novel encoding approaches, MCM with DTF, MCM with UTF and WAM with UTF, are described, all of which consider the information both in...
Proceedings Article
Short-term Load Forecasting Based on FCM and Complex Gaussian Wavelet SVM
Yongkang Zheng, Weirong Chen, Chaohua Dai, Shengyong Ye
Complex Gaussian wavelet support vector machine (CGW-SVM) is constructed with complex Gaussian wavelet kernel function for short-term load forecasting (STLF). Based on the chaotic characteristics of short-term load time series, the series is reconstructed with phase space reconstruction theory (PSRT)....
Proceedings Article
Model Identification and Control for Nonlinear Discrete-time System with Time Delay: A Support Vector Machine Approach
Zhong Weimin
In this paper, a design procedure of support vector machine (SVM) based model identification and control strategy for stable nonlinear discrete-time process with input-output form is proposed. In order to be able to implement the control structure, the both inverse and straight model representation and...
Proceedings Article
Parameters selection of SVM for function approximation based on Differential Evolution
ZHOU Shaowu, WU Lianghong, YUAN Xiaofang, TAN Wen
Support vector machines (SVM) is a new machine learning method, and it has the ability to approximate nonlinear functions with arbitrary accuracy. Right setting parameters are very crucial to learning results and generalization ability of SVM. In this paper, parameters selection is regarded as a compound...
Proceedings Article
Multicategory Nonparallel Proximal Support Vector Machine
Xubing Yang, Songcan Chen, Zhisong Pan
We propose a multicategory classifier, termed as Multicategory Nonparallel Proximal Support Vector Machine (MNPSVM), which is in the spirit of proximal SVMs via generalized eigenvalues (GEPSVM). Difference from GEPSVM lie in that: 1) MNPSVM keeps the genuine rather than approximate geometrical interpretation...
Proceedings Article
Research on SVM ensemble and its application to remote sensing classification
Heng-nian Qi, Mei-li Huang
The paper analyzes the key concepts, theories and methods of machine learning ensemble, and reviews the related studies on support vector machine (SVM) ensemble. The experiments on the remote sensing classification show that SVM ensemble is more accurate than single SVM. To obtain an effective SVM ensemble,...
Proceedings Article
An Improved Caching Strategy for Training
Liang Zhou, Fen Xia, Yanwu Yang
Computational complexity is one of the most important issues while dealing with the training of Support Vector Machines(SVMs), which is done by solving corresponding linear constrained convex quadratic programming problems. The state-ofthe- art training of SVMs takes iterative decomposition strategies...
Proceedings Article
Improving Trained LS--SVM Performance with New Available Data
Yangguang Liu, Bin Xu, Jun Liu
Learning is obtaining an underlying rule by using training data sampled from the environment. In many practical situations in inductive learning algorithms, it is often expected to further improve the generalization capability after the learning process has been completed if new data are available. One...
Proceedings Article
Transient Stability Assessment of Power System Based on Support Vector Machine
Shengyong Ye, Yongkang Zheng, Qingquan Qian
Machine learning methods are promising tools to transient stability assessment (TSA) of power system. Support vector machine (SVM) is used to assess the transient stability of power system after faults occur on transmission lines. Single machine attributes were studied as inputs of the SVM classifier....
Proceedings Article
An Ensemble Method for Multi-class and Multi-label Text Categorization
Bo-Feng Zhang, Xin Xu, Jinshu Su
A method for multi-class and multi-label automated text categorization based on twin-SVM with naïve Bayes ensemble is proposed. Twin-SVM classifiers give a solution to the multi-label problem. For multi-class situation, naïve Bayes classifier constrains the belonging scope of a testing sample within...
Proceedings Article
On System Identification Based on Online Least Squares Support Vector Machine
Bin Liu, Zhiping Wang
System identification is a fundamental topic of control theory, and LS-SVM has been applied to system identification. An online training algorithm of LS-SVM for system identication is presented, which is suitable for the data set supplied in sequence rather than in batch. The online algorithm avoids...
Proceedings Article
Research of the speaker verification based on the SVM-GMM mixture model
Cui Xuan, Deng Bo, Zhuang Wen
We put forward a new SVM-GMM mixture model to improve recognition rate of the speaker verification system in the paper. Support vector machines (SVM) and Gaussian mixture model (GMM) are widely applied to the speaker verification, but both have some disadvantages. We present a new approach for speaker...
Proceedings Article
Electrocardiogram Classification Method Based on SVM
Xiao Tang, Mo Zhiwen
Heart disease is one of the main diseases threatening human beings health, and electrocardiogram is the important basis of diagnosing cardiovascular disease. Therefore, the computer auto analysis of ECG remains the research hotspot in medical Engineering. Since the distinctiveness and variability of...
Proceedings Article
SVM-based analysis and prediction on network traffic
Weidong Luo, Xingwei Liu, Jian Zhang
With continuous scale-up of the network and increase of the kinds of the services on the network, more and more people pay attention to the modeling and prediction for network traffic. Recently, SVM (Support Vector Machine), a new machine learning method, is comprehensively used to solve the problem...