scholarly journals Faster Support Vector Machines

2021 ◽  
Vol 26 (1) ◽  
pp. 1-21
Author(s):  
Sebastian Schlag ◽  
Matthias Schmitt ◽  
Christian Schulz
Keyword(s):  
Support Vector Machines ◽  
Time Complexity ◽  
Original Problem ◽  
Label Propagation ◽  
Support Vector ◽  
Propagation Algorithm ◽  
Svm Model ◽  
Vector Machines ◽  
Data Points ◽  
Classification Quality

The time complexity of support vector machines (SVMs) prohibits training on huge datasets with millions of data points. Recently, multilevel approaches to train SVMs have been developed to allow for time-efficient training on huge datasets. While regular SVMs perform the entire training in one—time-consuming—optimization step, multilevel SVMs first build a hierarchy of problems decreasing in size that resemble the original problem and then train an SVM model for each hierarchy level, benefiting from the solved models of previous levels. We present a faster multilevel support vector machine that uses a label propagation algorithm to construct the problem hierarchy. Extensive experiments indicate that our approach is up to orders of magnitude faster than the previous fastest algorithm while having comparable classification quality. For example, already one of our sequential solvers is on average a factor 15 faster than the parallel ThunderSVM algorithm, while having similar classification quality. 1

scholarly journals Application of the Artificial Neural Network and Support Vector Machines in Forest Fire Prediction in the Guangxi Autonomous Region, China

2020 ◽  
Vol 2020 ◽  
pp. 1-14
Author(s):  
Yudong Li ◽  
Zhongke Feng ◽  
Shilin Chen ◽  
Ziyu Zhao ◽  
Fengge Wang
Keyword(s):  
Neural Network ◽  
Support Vector Machines ◽  
Forest Fire ◽  
Bp Neural Network ◽  
Forest Fires ◽  
Prediction Accuracy ◽  
Support Vector ◽  
Autonomous Region ◽  
Svm Model ◽  
Vector Machines

The study of forest fire prediction is of great environmental and scientific significance. China’s Guangxi Autonomous Region has a high incidence rate of forest fires. At present, there is little research on forest fires in this area. The application of the artificial neural network and support vector machines (SVM) in forest fire prediction in this area can provide data for forest fire prevention and control in Guangxi. In this paper, based on Guangxi’s 2010–2018 satellite monitoring hotspot data, meteorology, terrain, vegetation, infrastructure, and socioeconomic data, the researchers determined the main forest fire driving factors in Guangxi. They used feature selection and backpropagation neural networks and radial basis SVM to build forest fire prediction models. Finally, the researchers use the accuracy, precision, and area under the characteristic curve (ROC-AUC) and other indicators to evaluate the predictive performance of the two models. The results showed that the prediction accuracy of the BP neural network and SVM is 92.16% and 89.89%, respectively. As both results are over 85%, the requirements of prediction accuracy is met. These results can be used for forest fire prediction in the Guangxi Autonomous Region. Specifically, the accuracy of the BP neural network was 0.93, which was higher than that of the SVM model (0.89); the recall of the SVM model was 0.84, which was lower than the BANN model (0.92), and the AUC value of the SVM model was 0.95, which was lower than the BP neural network model. The obtained results confirm that the BP neural network model can provide more prediction accuracy than support vector machines and is therefore more suitable for forest fire prediction in Guangxi, China. This research provides the necessary theoretical basis and data support for application in the field of forestry of the Guangxi Autonomous Region, China.

scholarly journals Forecasting Electric Load by Support Vector Machines with Genetic Algorithms

2005 ◽  
Vol 9 (2) ◽  
pp. 134-141 ◽  
Author(s):  
Ping-Feng Pai ◽  
◽  
Wei-Chiang Hong ◽  
Chih-Shen Lin ◽  
◽  
...  
Keyword(s):  
Genetic Algorithms ◽  
Support Vector Machines ◽  
Moving Average ◽  
Predictive Performance ◽  
Support Vector ◽  
Electric Load ◽  
Power Company ◽  
Svm Model ◽  
Vector Machines ◽  
General Regression Neural Networks

Support vector machines (SVMs) have been successfully used in solving nonlinear regression and time series problems. However, the application of SVMs to load forecasting is very rare. Therefore, the purpose of this paper is to examine the feasibility of SVMs in forecasting electric load. In addition, the genetic algorithms are applied in the parameter selection of SVM model. Forecasting results compared with other two models, namely autoregressive integrated moving average (ARIMA) and general regression neural networks (GRNN), are provided. The experimental data are borrowed from the Taiwan Power Company. The numerical results indicate that the SVM model with genetic algorithms (SVMG) results in better predictive performance than the other two approaches.

Residual Subsidence Prediction of Abandoned Mine Goaf Based on Wavelet Support Vector Machines

2012 ◽  
Vol 524-527 ◽  
pp. 330-336 ◽  
Author(s):  
Zheng Shuai Wang ◽  
Ka Zhong Deng
Keyword(s):  
Support Vector Machines ◽  
Prediction Model ◽  
Abandoned Mine ◽  
Support Vector ◽  
Stochastic Fluctuation ◽  
Stochastic Signal ◽  
Multi Resolution Analysis ◽  
Svm Model ◽  
Vector Machines ◽  
Resolution Analysis

The prediction of residual subsidence is the fundament of stability evaluation of buildings foundation in the abandoned mine goaf, so how to get the residual subsidence with high precision is significant to reclaim the goaf for buildings. In this paper, a novel prediction model named wavelet support vector machines (WT-SVM) is proposed to forecast residual subsidence. Aiming at the stochastic fluctuation of the subsidence series, the measured data of residual subsidence were separated into components, namely, trend, oscillating sequence and stochastic signal, via wavelet multi-resolution analysis; then, the prediction model was established based on SVM regression algorithm, respectively, and the sum of the total corresponding prediction values were regarded as the final results of the residual subsidence. The predicting results of WT-SVM, SVM and BP neural network (BP-NN) were compared by a case study. The conclusions are as follows: WT-SVM model is obviously superior to other models in terms of the aspects of prediction precision, step and stability, which indicates the feasibility and effectivity of WT-SVM in predicting residual subsidence of the abandoned mine goaf.

Obtaining and Predicting of Equivalent Load Spectrum for Overhead Traveling Crane Based on Support Vector Machines

2011 ◽  
Vol 148-149 ◽  
pp. 125-129
Author(s):  
Ge Ning Xu ◽  
Xiao Peng Li ◽  
Xiao Ning Fan
Keyword(s):  
Fatigue Life ◽  
Support Vector Machines ◽  
Linear Mapping ◽  
Support Vector ◽  
Load Spectrum ◽  
Large Samples ◽  
Present Essay ◽  
Svm Model ◽  
Vector Machines ◽  
Equivalent Load

Crane load spectrum is the basis data of the evaluation of fatigue life, obtaining the large samples data need to consume large amount of time and finance. For exploring and initiating a new way of obtaining and predicting the crane load spectrum, the present essay uses Support Vector Machines (SVM) to establish the non-linear mapping relation between this type of crane's work cycles and different rated weights and lifting loads, namely equivalent load spectrum of crane. The Engineering example demonstrates: it is feasible to construct the equivalent load spectrum for some type of crane using SVM. Meanwhile, the equivalent load spectrums of the typical or future crane in this type could be obtained and predicted through the trained SVM model. It has provided the reliable data for the reliability analysis and fatigue life prediction of crane.

scholarly journals Support Vector Machines Analysis of Web Pages Binary Classification Quality

2017 ◽  
Vol 4 ◽  
Author(s):  
S.V. Voloshin ◽  
◽  
A.L. Tsaregorodtsev ◽  
E.A. Kartashev ◽  
V.V. Slavskiy ◽  
...  
Keyword(s):  
Support Vector Machines ◽  
Binary Classification ◽  
Support Vector ◽  
Web Pages ◽  
Vector Machines ◽  
Classification Quality
Sign in / Sign up

Export Citation Format

Share Document