A boosted SVM classifier trained by incremental learning and decremental unlearning approach

Kashef, Rasha

doi:10.1016/j.eswa.2020.114154

Cited by 50 publications

(19 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For a good separation, some margin

must be set with a nonlinear kernel function. A good solution to this problem is to use the nonlinear SVM [ 42 ] model. In this form, the PCA method becomes supervised.…”

Section: Methodsmentioning

confidence: 99%

“…The task is to determine a classifier that allows to divide the entire

space into two separate areas corresponding to the −1, 1 classes and to classify features to principal components as good as possible. For this purpose, as already mentioned, the nonlinear support vector machine approach [ 42 ] can be used. If

denotes a symmetric function of two kernel vectors such that

and

, then the transform can be specified as

, where:

…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Application of Feature Extraction Methods for Chemical Risk Classification in the Pharmaceutical Industry

Topolski

2021

Sensors

View full text Add to dashboard Cite

The features that are used in the classification process are acquired from sensor data on the production site (associated with toxic, physicochemical properties) and also a dataset associated with cybersecurity that may affect the above-mentioned risk. These are large datasets, so it is important to reduce them. The author’s motivation was to develop a method of assessing the dimensionality of features based on correlation measures and the discriminant power of features allowing for a more accurate reduction of their dimensions compared to the classical Kaiser criterion and assessment of scree plot. The method proved to be promising. The results obtained in the experiments demonstrate that the quality of classification after extraction is better than using classical criteria for estimating the number of components and features. Experiments were carried out for various extraction methods, demonstrating that the rotation of factors according to centroids of a class in this classification task gives the best risk assessment of chemical threats. The classification quality increased by about 7% compared to a model where feature extraction was not used and resulted in an improvement of 4% compared to the classical PCA method with the Kaiser criterion, with an evaluation of the scree plot. Furthermore, it has been shown that there is a certain subspace of cybersecurity features, which complemented with the features of the concentration of volatile substances, affects the risk assessment of chemical hazards. The identified cybersecurity factors are the number of packets lost, incorrect Logins, incorrect sensor responses, increased email spam, and excessive traffic in the computer network. To visualize the speed of classification in real-time, simulations were carried out for various systems used in Industry 4.0.

show abstract

“…For a good separation, some margin

must be set with a nonlinear kernel function. A good solution to this problem is to use the nonlinear SVM [ 42 ] model. In this form, the PCA method becomes supervised.…”

Section: Methodsmentioning

confidence: 99%

“…The task is to determine a classifier that allows to divide the entire

denotes a symmetric function of two kernel vectors such that

and

, then the transform can be specified as

, where:

…”

Section: Methodsmentioning

confidence: 99%

Application of Feature Extraction Methods for Chemical Risk Classification in the Pharmaceutical Industry

Topolski

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Support Vector Machine (SVM) is a machine learning method developed on the basis of statistical learning theory (Kashef, 2021). The basic principle is to minimize the structured risk by calculating the minimum value of the empirical risk, so as to improve the learning ability and stability of machine learning.…”

Section: Svmmentioning

confidence: 99%

Application of machine learning algorithms for the prediction of flame temperature in small-scale burner fueled with ethanol-diesel fuel blends

JIN

Zhu

2022

JTST

View full text Add to dashboard Cite

Application of machine learning algorithms for the predictionof flame temperature in small-scale burner fueled with ethanol-diesel fuel blends IntroductionAs a light petroleum product, diesel fuel is widely used in ships, large-scale vehicles, and locomotives in railway. However, contaminants such as oxides of nitrogen (NOx) and particulate matter (PM) are generated with the use of diesel fuel. It was found that diesel fuel can be partially replaced by oxygenated fuels, reducing contaminant emissions from combustion. Consequently, many researchers have devoted their attention to the study of oxygenated fuels, such as methanol, ethanol and biodiesel (Hansen et al., 2005;Moon et al., 2010;Yao et al., 2010). Ethanol contains oxygen in 34.7wt% and its required air is theoretically nearly half as diesel when burned (Lu et al., 2021). Therefore, diesel fuel blended with ethanol can make combustion more fully, which reduces soot emissions.Ethanol-diesel fuel blends requires appropriate equipment to achieve efficient combustion. The small-scale burner is the key equipment of small-electromechanical system and this burner has the advantage of high energy density. However, in the small scale, many new problems and challenges arise, such as incomplete combustion, unstable flame, and large heat loss (

show abstract

“…However, there are certain challenges of applying SVM over large datasets. Targeting these challenges, ample research has been conducted in recent years [17], [18], [19], [20], [21].…”

Section: Svm For Large Datasetsmentioning

confidence: 99%

Random Partition Based Adaptive Distributed Kernelized SVM for Big Data

et al. 2022

View full text Add to dashboard Cite

In this paper, we present a distributed classification technique for big data by efficiently using distributed storage architecture and data processing units of a cluster. While handling such large data, the existing approaches consider specific data partitioning techniques which demand complete data be processed before partitioning. This leads to an excessive overhead of high computation and data communication. The proposed method does not require any pre-structured data partitioning technique and is also adaptive to big data mining tools. We hypothesize that an effective aggregation of the information generated from data partitions by subprocesses of the complete learning process can lead to accurate prediction results while reducing the overall time complexity. We build three SVM based classifiers, namely one phase voting SVM (1PVSVM), two phase voting SVM (2PVSVM), and similarity based SVM (SIMSVM). Each of these classifiers utilizes the support vectors as the local information to construct the synthesized learner for efficiently reducing the training time and ensuring minimal communication between processing units. In this context, an extensive empirical analysis demonstrates the effectiveness of our classifiers when compared to other existing approaches on several benchmark datasets. However, among existing methods and three of our proposed (1PVSVM, 2PVSIM, and SIMSVM) methods, SIMSVM is the most efficient. Considering MNIST dataset, SIMSVM achieves an average speedup ratio of 0.78 and minimum scalability of 73% when the data size is scaled up to 10 times. It also retains high accuracy (99%) similar to centralized approaches.

show abstract

A boosted SVM classifier trained by incremental learning and decremental unlearning approach

Cited by 50 publications

References 35 publications

Application of Feature Extraction Methods for Chemical Risk Classification in the Pharmaceutical Industry

Application of Feature Extraction Methods for Chemical Risk Classification in the Pharmaceutical Industry

Application of machine learning algorithms for the prediction of flame temperature in small-scale burner fueled with ethanol-diesel fuel blends

Random Partition Based Adaptive Distributed Kernelized SVM for Big Data

Contact Info

Product

Resources

About