Support Vector Machine Classification Algorithm Research Articles

An ImprovedD2GAN‐based oversampling algorithm for imbalanced data classification

AbstractTo address the problems of pattern collapse, uncontrollable data generation and high overlap rate when generative adversarial network (GAN) oversamples imbalanced data, we propose an imbalanced data oversampling algorithm based on improved dual discriminator generative adversarial nets (D2GAN). First, we integrate the positive class attribute information into the generator and the discriminator to ensure that the generator only generates the samples for positive class samples, which overcomes the problem of uncontrollable data generation by the generator. Second, we introduce a classifier into D2GAN for discriminating the generated samples and the original data, which avoids the overlap among the generated samples and the negative class samples, and ensures the diversity of the generated samples, the problem of pattern collapse is solved. Finally, the performance of the proposed algorithm is evaluated on 9 datasets by using SVM and neural network classification algorithm for oversampling experiments, the results show that the proposed algorithm effectively improve the classification performance of imbalanced data.

Assessment of Object-Based Classification for Mapping Land Use and Land Cover using Google Earth

<p>Land Use and Land Cover (LULC) maps perform a significant part in Remote Sensing (RS) in monitoring and analyzing earth information for human development. Because of the high spectrum variability, LULC classification from RS data is extremely difficult. The study's objective is to assess the object-based LULC classification (OBC) accuracy of composite images from the Landsat 8 OLI (Operational Land Imager) and auxiliary features using SNIC segmentation and five classifiers on Google Earth Engine (GEE). The outcome of the study indicates that the OBC with only spectral features achieves lesser accuracy because small objects on the land surface cannot be observed. But when OBC is paired with a variety of auxiliary features, the OBC may recognize small objects with greater accuracy. After many subsequent assessments the combination of the following aspect: composite image of all the seven bands, new feature set, Simple Non-Iterative Clustering (SNIC) segmentation algorithm, and Support Vector Machine (SVM) classification algorithm gives better accuracy of ~ 94.42% and kappa coefficient of 92.07. The inclusion of auxiliary features and the OBC method reduces the misclassification rate related to the confusion of bare land, uncultivated land, and urban, which provides accurate information on forest, waterbody, and bare land classes.</p>

Experimental design and data analysis and optimization of mechanical condition diagnosis for transformer sets

Abstract The typical power transformer diagnosis approach is imprecise and unstable. A support vector machine classification algorithm is proposed, by designing an algorithm program that can improve the accuracy and speed of energy transformer diagnosis, the vibration signals of the surface twisting in different states are extracted by wavelet packet energy spectrum signal processing method, it is verified that the curve similarity between the vibration simulation model and the measured data is greater than 0.98, proving the simulation model’s validity. The calculation technique of online short circuit inductance is developed from the equivalent transformer model, and the variation error of simulation results is less than 0.05% when compared to the real transformer characteristics. The suggested state diagnostic technique successfully compensates for the drawbacks of the reactance method, which is incapable of detecting and judging the slightly loose or faulty winding. The method’s accuracy and superiority, as well as the practicability of the state diagnosis system, are demonstrated.

An advance extended binomial GLMBoost ensemble method with synthetic minority over-sampling technique for handling imbalanced datasets

<span lang="EN-US">Classification is an important activity in a variety of domains. Class imbalance problem have reduced the performance of the traditional classification approaches. An imbalance problem arises when mismatched class distributions are discovered among the instances of class of classification datasets. An advance extended binomial GLMBoost (EBGLMBoost) coupled with synthetic minority over-sampling technique (SMOTE) technique is the proposed model in the study to manage imbalance issues. The SMOTE is used to solve the proposed model, ensuring that the target variable's distribution is balanced, whereas the GLMBoost ensemble techniques are built to deal with imbalanced datasets. For the entire experiment, twenty different datasets are used, and support vector machine (SVM), Nu-SVM, bagging, and AdaBoost classification algorithms are used to compare with the suggested method. The model's sensitivity, specificity, geometric mean (G-mean), precision, recall, and F-measure resulted in percentages for training and testing datasets are 99.37, 66.95, 80.81, 99.21, 99.37, 99.29 and 98.61, 54.78, 69.88, 98.77, 96.61, 98.68, respectively. With the help of the Wilcoxon test, it is determined that the proposed technique performed well on unbalanced data. Finally, the proposed solutions are capable of efficiently dealing with the problem of class imbalance.</span>

Fuzzy Comprehensive Evaluation Model of Project Investment Risk Based on Computer Vision Technology

With the development of the economy and real-time embedded systems and the progress of science and technology, people’s economic income forms have undergone tremendous changes, and the concept of financial management has become clearer in people’s property income arrangements. Project investment is one of the most popular financial management methods in the era of big data. Both large enterprise groups and individual petty bourgeoisie groups have begun to pay attention to the risks and benefits brought by the new financial management method of project investment. This paper’s goal is to develop a fuzzy comprehensive evaluation (FCE) model for project investment risk based on computer vision technology and explore the application of computer vision technology in project investment risk evaluation. This article first uses a real-time embedded system to understand the basic process of project investment and select 10 investment experts for risk assessment, risks, and causes of the risks through literature research and case analysis. Then, this paper establishes a model of fuzzy comprehensive evaluation of project investment risk through computer vision technology, real-time embedded systems, and neural network models in big data and artificial intelligence technology to realize the analysis and prediction of project investment risk. The fuzzy comprehensive evaluation method and analytic hierarchy process (AHP) are used in this evaluation model to evaluate and forecast project investment risks. In addition, this paper also trains and tests the risk evaluation model of this research through the support vector machine classification algorithm, the real-time embedded system, and the average random consistency index. The research shows that the fuzzy comprehensive evaluation model of this study has higher accuracy for project investment risk evaluation than other risk evaluation methods. For example, for the investment risk of chemical fiber projects, this research model evaluated the factors such as organization, management, technology, and economy and found that the risks were all higher than 21.36%, which concluded that the overall investment risk of chemical fiber projects was relatively high.

Multi-element polynomial chaos expansion based on automatic discontinuity detection for nonlinear systems

This article focuses on the stochastic modeling of nonlinear systems featuring discontinuities in their response surface. More specifically, original developments are presented to improve the efficiency of multi-element polynomial chaos expansion for such systems. First, an automated detection procedure of the discontinuities is proposed. It relies on an iterative algorithm with a polynomial annihilation edge detection method and support vector machine classification algorithms leading to the representation of the discontinuity as a B-spline curve. Based on this curve, an ad-hoc decomposition of the variable space is considered and an original approach for the application of polynomial chaos expansion on each subdomain yields a robust and versatile way to compute the response surface of the system of interest. The proposed methodology is detailed and applied to several nonlinear academic systems such as a circular discontinuity and the Duffing oscillator including one or two discontinuities in its response surface. Through these applications, it is evidenced that, compared to the classical polynomial chaos and multi-element polynomial chaos expansions, the proposed methodology yields an approximation of the discontinuous responses that is both more accurate and less computationally expensive. The influence of the main parameters of the proposed methodology is also analyzed in details. This parametric analysis underlines the robustness of the methodology and highlights the key parameters in terms of computational cost and accuracy of the discontinuities. The proposed methodology is finally applied to an industrial application, it allows to efficiently compute the surface response of an industrial compressor blade undergoing structural contacts.

DP-AOP: A novel SVM-based antioxidant proteins identifier

The identification of antioxidant proteins is a challenging yet meaningful task, as they can protect against the damage caused by some free radicals. In addition to time-consuming, laborious, and expensive experimental identification methods, efficient identification of antioxidant proteins through machine learning algorithms has become increasingly common. In recent years, researchers have proposed models for identifying antioxidant proteins; unfortunately, although the accuracy of models is already high, their sensitivity is too low, indicating the possibility of overfitting in the model. Therefore, we developed a new model called DP-AOP for the recognition of antioxidant proteins. We used the SMOTE algorithm to balance the dataset, selected Wei's proposed feature extraction algorithm to obtain 473 dimensional feature vectors, and based on the sorting function in MRMD, scored and ranked each feature to obtain a feature set with contribution values ranging from high to low. To effectively reduce the feature dimension, we combined the dynamic programming idea to make the local eight features the optimal subset. After obtaining the 36 dimensional feature vectors, we finally selected 17 features through experimental analysis. The SVM classification algorithm was used to implement the model through the libsvm tool. The model achieved satisfactory performance, with an accuracy rate of 91.076 %, SN of 96.4 %, SP of 85.8 %, MCC of 82.6 %, and F1 core of 91.5 %. Furthermore, we built a free web server to facilitate researchers' subsequent unfolding studies of antioxidant protein recognition. The website is http://112.124.26.17:8003/#/.

FUEL INCREASE SENTIMENT ANALYSIS USING SUPPORT VECTOR MACHINE WITH PARTICLE SWARM OPTIMIZATION AND GENETIC ALGORITHM AS FEATURE SELECTION

BBM, or fuel oil, is one of the essential needs of the Indonesian people. The government's policy regarding the increase in fuel prices raises many opinions from the public. Twitter is one of the social media that Indonesian people often use to express opinions on a topic. In this study, sentiment analysis was carried out on public opinion regarding the fuel price increase policy from Twitter social media. This research is expected to help determine public opinion regarding the fuel price increase policy with positive, neutral and negative sentiments. The sentiment analysis method used is the Support Vector Machine (SVM) classification algorithm. The results of the accuracy of SVM were compared with accuracy by adding a feature selection process. The Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) algorithms are used for the feature selection method. After several experiments using the three methods, the SVM method with the Radial Basis Function (RBF) kernel produced the best accuracy of 71.2%. The combination of the SVM method with the RBF and PSO kernels obtained an accuracy of 68.84%, and the combination of the RBF and GA kernel SVM methods obtained an accuracy of 69.52%.

Modeling of the Acute Lymphoblastic Leukemia Detection by Convolutional Neural Networks (CNNs).

The techniques differed in many of the literature on the detection of Acute Lymphocytic Leukemia from the blood smear pictures, as the cases of infection in the world and the Kingdom of Saudi Arabia were increasing and the causes of this disease were not known, especially for children, which is a serious and fatal disease. Through this work we seek to contribute to discover the blood cells affected by Acute Lymphocytic Leukem and to find an effective and fast method and to have the correct diagnosis as the time factor is important in the diagnosis and the initiation of treatment. which is based on one of the deep learning techniques that specialize in very deep networks, the use of one of the CNNs is VGG16. Detection scheme is implemented by pre-processing, feature extraction, model building, fine tuning method, classification are executed. By using VGG16 pre-trained, and using SVM and MLP classification algorithms in Machine Learning. Our results are evaluated based on criteria, such as Accuracy, Precision, Recall, and F1-Score. The accuracy results for SVM classifier MLP of 77% accuracy at 0.001 learning rate and the accuracy for SVM classifier 75% at 0.005 learning rate. Whereas, the best accuracy result for VGG16 model was 92.27% at 0.003 learning rate. The best validation accuracy result was 85.62% at 0.001 learning rate.

Comparison between Bayesian and SVM model for breast cancer risk prediction

Breast cancer ranks first in the incidence of cancer worldwide. It is a kind of malignant disease with incidence increasing year by year. Recently, machine learning algorithms widely used in cancer prediction and other fields to relieve the burden of doctors and accelerate the diagnostic process. In this work, two representative models, the support vector machine (SVM) and Bayesian classification algorithm are leveraged for breast cancer risk prediction. It is found that the effect and accuracy of the two algorithms are very different when they are used for breast cancer prediction. However, there is still a research gap in the comparison of the two algorithms in practical application. Therefore, the research topic of this paper is the comparison between the Bayesian classification algorithm and the SVM algorithm in breast cancer prediction. The research methods of this paper are as follows: firstly, the dataset is collected, then applying the Bayesian classification to process the dataset, and then the SVM algorithm is used to process the dataset. Finally, the processing results of the two algorithms are compared to comprehensively analyze and compare which algorithm is more efficient and more suitable for actual breast cancer prediction. The test results support that the SVM outperforms the Bayesian classification algorithm in the actual target tracking problem. Therefore, it is suggested to choose the SVM classification algorithm in the actual target tracking problem to boost the accuracy and efficiency of prediction results to the greatest extent.

Class Balancing Methods Comparison for Software Requirements Classification on Support Vector Machines

Cost, time, and development effort can increase due to errors in analyzing functional and non-functional software requirements. To minimize these errors, previous research has tried to classify software requirements, especially non-functional requirements, on the PROMISE dataset using the Bag of Words (BoW) feature extraction and the Support Vector Machine (SVM) classification algorithm. On the other hand, the unbalanced distribution of class labels tends to decrease the evaluation result. Moreover, most software requirements are usually functional requirements. Therefore, there is a tendency for classifier models to classify test data as functional requirements. Previous research has performed class balancing on a dataset to handle unbalanced data. The study can achieve better classification evaluation results. Based on the previous research, this study proposes to combine the class balancing method and the SVM algorithm. K-fold cross-validation is used to optimize the training and test data to be more consistent in developing the SVM model. Tests were carried out on the value of K in k-fold, i.e., 5, 10, and 15. Results are measured by accuracy, f1-score, precision, and recall. The Public Requirements (PURE) dataset has been used in this research. Results show that SVM with class balancing can classify software requirements more accurately than SVM without class balancing. Random Over Sampling is the class balancing method with the highest evaluation score for classifying software requirements on SVM. The results showed an improvement in the average value of accuracy, f1 score, precision, and recall in SVM by 22.07%, 19.67%, 17.73%, and 19.67%, respectively.

Capped L2,p-Norm Metric Based on Robust Twin Support Vector Machine with Welsch Loss

A twin bounded support vector machine (TBSVM) is a phenomenon of symmetry that improves the performance of the traditional support vector machine classification algorithm. In this paper, we propose an improved model based on a TBSVM, called a Welsch loss with capped L2,p-norm distance metric robust twin bounded support vector machine (WCTBSVM). On the one hand, by introducing the capped L2,p-norm metric in the TBSVM, the problem of the non-sparse output of the regularization term is solved; thus, the generalization and robustness of the TBSVM is improved and the principle of minimizing the structural risk is realized. On the other hand, a bounded, smooth, and non-convex Welsch loss function is introduced to reduce the influence of noise, which further improves the classification performance of the TBSVM. We use a half-quadratic programming algorithm to solve the model non-convexity problem caused by Welsch loss. Therefore, the WCTBSVM is more robust and effective in dealing with noise compared to the TBSVM. In addition, to reduce the time complexity and speed up the convergence of the algorithm, we constructed a least squares version of the WCTBSVM, named the fast WCTBSVM (FWCTBSVM). Experimental results on both UCI and artificial datasets show that our model can show better classification performance on classification problems.

Training Data Selection for Record Linkage Classification

This paper presents a new two-step approach for record linkage, focusing on the creation of high-quality training data in the first step. The approach employs the unsupervised random forest model as a similarity measure to produce a similarity score vector for record matching. Three constructions were proposed to select non-match pairs for the training data, with both balanced (symmetry) and imbalanced (asymmetry) distributions tested. The top and imbalanced construction was found to be the most effective in producing training data with 100% correct labels. Random forest and support vector machine classification algorithms were compared, and random forest with the top and imbalanced construction produced an F1-score comparable to probabilistic record linkage using the expectation maximisation algorithm and EpiLink. On average, the proposed approach using random forests and the top and imbalanced construction improved the F1-score by 1% and recall by 6.45% compared to existing record linkage methods. By emphasising the creation of high-quality training data, this new approach has the potential to improve the accuracy and efficiency of record linkage for a wide range of applications.

A Survey of Fuzzy Pattern Tree Classification Algorithms

Data classification algorithm is the core content of big data mining. Its main function is to extract valuable knowledge and information,analyze the characteristics of all kinds of information, and provide data basis for further research. With the wide application of data mining technology, data classification algorithms continue to emerge and gradually improve. The classic classification algorithms include decision tree classification algorithm, naive Bayes algorithm, support vector machine classification algorithm, artificial neural network classification algorithm, fuzzy pattern tree (FPT) and so on. This paper summarizes several common algorithms in data classification algorithms, and analyzes their characteristics to understand their algorithm principles and application scenarios.

Support Vector Machine Based Classification of Leaf Diseases

Myanmar is well known for agricultural country; wherein about 65% of the labor force depends on agriculture. Since the leaf diseases are microscopic organism, cannot be detected normal human eyes. Leaves are special indicator to distinguish the diseases because the image information of the leaf are changed when the leaf surf the diseases. So, the image processing techniques can be used in agricultural sector. The research work presents a support vector machine classifier algorithm by using MATLAB R2017a for the classification of leaf diseases such as Alternaria Alternata, Cercospora leaf spot, Bacterial Blight and so on. In this research work, RGB color space is converted into HSI (Hue Saturation Intensity) color space. In segmentation step, k-means clustering is used to select the defected area, and it is extracted the features by using GLCM (Gray Level Co-occurrence Matrix). Prior to the features extraction, the median filter is used for getting noise free feature results. Finally, the leaf disease is classified by using support vector machine (SVM) and computes the accuracy. From the obtained results, the maximum accuracy of the system is 83%.

Effects of water stress and fertilizer stress on maize growth and spectral identification of different stresses

Water stress and fertilizer stress have a significant impact on the growth and yield of maize. In order to improve the timeliness and accuracy of irrigation and fertilizer application, it is crucial to monitor water stress and fertilizer stress rapidly and accurately. This would help in conserving water and fertilizer resources and ensuring a stable maize yield. To this end, pot experiments were set up to explore the growth differences and photosynthetic properties of maize under water stress and fertilizer stress. The hyperspectral technology was used to construct the spectral indexes that can distinguish stress types, and the classification algorithm was combined to identify stress types. The research has shown that the plant height, basal diameter, leaf area, and photosynthetic properties of maize decreased with an increase in drought stress. However, rewatering could compensate for drought stress. Furthermore, fertilizer stress also affected water uptake by plants, and high nitrogen stress had a significant negative effect on the growth of maize plants. We employed a combination of spectral indexes and the support vector machine (SVM) classification algorithm in a stepwise manner to identify stress types. Using the training dataset, we constructed six classifiers for distinguishing stress types, including the SVM classifier, K-nearest neighbor (KNN) classifier, naive Bayes (NB) classifier, decision tree (DT) classifier, random forest (RF) classifier, and AdaBoost classifier. Our results showed that the RF and AdaBoost classifiers obtained stable results in stress type differentiation, achieving accurate identification of unstressed, water stressed, and fertilizer stressed maize plants. This is expected to provide a solid basis and reference for monitoring crop stress types in agricultural fields.

674 Diffusion Basis Spectrum Imaging Predicts Comprehensive Clinical Outcomes Following Surgery for Cervical Spondylotic Myelopathy

INTRODUCTION: A major shortcoming in improving care for cervical spondylotic myelopathy (CSM) patients is the lack of robust quantitative imaging tools to guide surgical decision-making. METHODS: A prospective cohort study enrolled fifty CSM patients who underwent cervical decompressive surgery. DBSI metrics assessed white matter tract integrity by fractional anisotropy, axial diffusivity, radial diffusivity, and fiber fraction. To evaluate extra-axonal diffusion, DBSI measures restricted and non-restricted fractions. Neurofunctional status was assessed by the mJOA, MDI, and DASH. Quality of life was measured by the SF-36 PCS and MCS. The NDI was used to measure self-reported neck pain. Patient satisfaction was assessed by the NASS satisfaction index. Support vector machine classification algorithms were used to predict surgical outcomes. Specifically, three models were built for each clinical outcome measure (e.g., mJOA), including clinical, DBSI, and combined models. RESULTS: Twenty-seven mild (mJOA 15-17), 12 moderate (12-14) and 11 severe (0-11) CSM patients were enrolled. Twenty-four (60%) underwent anterior surgery compared to 16 (40%) posterior surgery. The mean (SD) follow-up was 23.2 (5.6, range 6.1-32.8) months. The best performing model was for the prediction of the NASS satisfaction index, with an accuracy [95% CI] of 94.4 [94.3, 94.8]. Conversely, the worst performing model was for the NDI, with an accuracy of 73.8 [73.6, 74.5]. When predicting improvement in the mJOA, the clinically-driven model had an accuracy of 61.9 [61.6, 62.5], compared to 78.6 [78.4, 79.2] in the DBSI model, and 90.5 [90.2, 90.8] in the combined model. CONCLUSIONS: When combined with key clinical covariates, DBSI metrics predicted improvement after surgical decompression with high accuracy. These results suggest that DBSI may serve as a non-invasive imaging biomarker for CSM.

498 Feasibility and Efficacy of Postoperative Diffusion Basis Spectrum Imaging to Evaluate Recovery After Surgery in Cervical Spondylotic Myelopathy

INTRODUCTION: Previous studies evaluating diffusion tensor imaging (DTI) in cervical spondylotic myelopathy (CSM) have demonstrated limitations in assessing postoperative outcomes following surgery. Diffusion basis spectrum imaging (DBSI), capable of providing granular assessments of spinal cord microstructure, may help address these shortcomings. METHODS: Twenty-seven CSM patients underwent preoperative diffusion-weighted MRI followed by cervical decompressive surgery and repeat imaging at two-years follow-up. DTI and DBSI metrics were extracted from the C3 spinal cord level. Diffusion-weighted MRI anisotropic tensors evaluated white matter tracts through fractional anisotropy, axial diffusivity, radial diffusivity, and fiber fraction. DBSI isotropic tensors assessed extra-axonal pathology through restricted and non-restricted fraction. Improvement after surgery was defined as a 2-point increase in the mJOA at two-years follow-up. Support vector machine classification algorithms were used to predict surgical outcomes. RESULTS: Fifteen mild (mJOA 15-17), 7 moderate (12-14) and 5 severe (0-11) CSM patients underwent surgical decompression and were followed-up to 23.2 (5.6, range 11.1-32.8) months postoperatively. Among DBSI metrics, isotropic diffusion components changed significantly from baseline to latest follow-up (p < 0.01), whereas anisotropic components remained stable (p > 0.05). When predicting improvement in the mJOA, SVM models built with DTI metrics had an accuracy of 74.1 [95% CI: 73.6, 74.6] compared to 85.2 [84.5, 85.3] in the DBSI model. CONCLUSIONS: Postoperative DBSI metrics predicted long-term outcomes with high accuracy. DBSI may have utility in assessing recovery trajectories for CSM patients following surgery.

ConvNeXt-Haze: A Fog Image Classification Algorithm for Small and Imbalanced Sample Dataset Based on Convolutional Neural Network

Aiming at the different abilities of the defogging algorithms in different fog concentrations, this paper proposes a fog image classification algorithm for a small and imbalanced sample dataset based on a convolution neural network, which can classify the fog images in advance, so as to improve the effect and adaptive ability of image defogging algorithm in fog and haze weather. In order to solve the problems of environmental interference, camera depth of field interference and uneven feature distribution in fog images, the CutBlur-Gauss data augmentation method and focal loss and label smoothing strategies are used to improve the accuracy of classification. It is compared with the machine learning algorithm SVM and classical convolution neural network classification algorithms alexnet, resnet34, resnet50 and resnet101. This algorithm achieves 94.5% classification accuracy on the dataset in this paper, which exceeds other excellent comparison algorithms at present, and achieves the best accuracy. It is proved that the improved algorithm has better classification accuracy.

Computer Vision and Machine Learning in Paddy Diseases Identification and Classification: A Review

Food security of the Indian population depends upon the agriculture. The annual crop loss due to the pestsand diseases is a severe problem which is addressed by the researcher time and again Paddy is one of the common crops found in most of the agriculture areas in India. Paddy plants exhibit health condition mainly in stem and leaf. Farmers are still using the traditional method and experts suggestions in identifying the diseases. Research in disease identification and classification process is going on for the past three decades but its automation is still an open challenge. Image segmentation based computer vision techniques, supervised and unsupervised classification algorithms of machine learning are widely used in the analysis of vegetation health analysis. Pattern recognition is playing a major role in identifying the features causing the diseases. The focus of the work is on investigating the performance of various segmentation techniques with support vector machine and K-Neural Network classification algorithms using statistical features. The hybrid IP-PR (Image Processing-Pattern Recognition)techniques are used for paddy disease classification..The performance of the classification algorithm is primarily depending upon the intelligence of the segmentation techniques. The properly segmented outputs give a better accuracy.