Support Vector Machine Technique Research Articles

Enhancing waste management and prediction of water quality in the sustainable urban environment using optimized algorithm of least square support vector machine and deep learning techniques

Urban groundwater influences a wide range of processes in the natural world, including climatic, geological, geomorphic, biogeochemical, ecotoxicological, hydrological, and sanitary processes, supporting several ecological services. Waste groundwater management refers to the activities and practices used to ensure groundwater resources' sustainable use and protection. This can include monitoring and evaluating groundwater resources' status and protecting groundwater from pollution and other forms of degradation. Many research works have been implemented in managing groundwater. There need to be more parametric measurements available with the current technology to monitor water quality. Groundwater management and monitoring water quality in the urban environment is an important task, as urbanization can lead to increased contamination of groundwater sources. One method for managing and monitoring groundwater quality is proposed using a least squares support vector machine (LS-SVM) with a particle optimization algorithm. The LS-SVM with PSO algorithm I s used in groundwater management as a method for monitoring and evaluating the quality of groundwater resources. The LS-SVM is a machine learning algorithm that uses the least squares approach to model complex data relationships. The PSO algorithm is a particle optimization algorithm that optimizes the parameters of the LS-SVM model. By combining these two techniques, the LS-SVM with PSO algorithm provides a more accurate prediction of groundwater quality compared to other algorithms such as KNN and SVM. The accuracy rate of various algorithms with groundwater pollution dataset with the algorithms of KNN 75.32%, SVM 81.78%, KCM 77.16%, and proposed work of LSSVM-PSO 92.73%.

Software Defined Networking Based Detection of DDoS Attacks

Software-defined networking (SDN) separates network management from data-traffic routes. More businesses are adopting it because of its flexibility, adaptability, and ability to improve traffic movement. SDN, or security-by-design, might be a desirable alternative for securing networks. While there have been many advancements in SDN technology, it is still susceptible to DDoS assaults. The growing frequency and scope of DDoS attacks pose a threat to network security, despite the availability of various methods for detecting and countering such attacks. There are two main methods to spot a distributed denial of service (DDoS) attack: signature recognition and abnormality detection. When personal characteristics such as fingerprints or iris scans are used to verify identity. Anomaly-based detection, which relies on network behavior, employs machine learning methods. We present a strategy for SDNs to identify DDoS assaults in this article. In the proposed architecture, DDoS attacks are detected using the Advanced Support Vector Machine (ASVM) technique. When compared to the SVM method, ASVM has the benefit of significantly less testing and training time. To evaluate the effectiveness of the suggested system, we use the Hierarchical Task Analysis (HTA) method of measuring human error.

An automated drug dependence detection system based on EEG

ObjectiveSubstance abuse causes damage to the brain structure and function. This research aim is to design an automated drug dependence detection system based on EEG signals in a Multidrug (MD) abuser. MethodsEEG signals were recorded from participants categorized into MD-dependents (n = 10) and Healthy Control (HC) (n = 12). The Recurrence Plot investigates the dynamic characteristics of the EEG signal. The entropy index (ENTR) measured from the Recurrence Quantification Analysis was considered the complexity index of the delta, theta, alpha, beta, gamma, and all-band EEG signals. Statistical analysis was performed by t-test. The support vector machine technique was used for the data classification. ResultsThe results show decreased ENTR indices in the delta, alpha, beta, gamma, and all-band EEG signal and increased theta band in MD abusers compared to the HC group. That indicated the reduction of complexity in the delta, alpha, beta, gamma, and all-band EEG signals in the MD group. Additionally, the SVM classifier distinguished the MD group from the HC group with 90% accuracy, 89.36% sensitivity, 90.7% specificity, and 89.8% F1 score. Conclusions and significanceThe nonlinear analysis of brain data was used to build an automatic diagnostic aid system that could identify HC people apart from those who abuse MD.

Evaluating face recognition with different texture descriptions and convolution neural network

Extracting the remarkable attributes of the image objects is an issue of ongoing research special in the face recognition problem. This paper presents two directions. The first is a comparison between the local binary patterns (LBP) and its modified center symmetric LBP drawn from localized facial expressions and due to the efficiency, K-nearest neighbor (KNN) and the support vector machine (SVM) techniques play significant roles in this research used to implement the proposed system efficiently. The second direction proposes an efficient architecture by depending on deep learning convolution neural network (CNN) to implement face recognition. Such a design consists of two parts: a convolutional learning feature model and a classification model. The first one learns the important feature,while the second part produces a score class for each sample input. Many experiments are implemented on the known dataset once for the number of nearest neighbors (K value), and then decrease the number of expression samples for each individual the other time. The cross-validation method is used to provide a true picture of the accuracy of the face recognition system. In all experiment results, the center symmetric LBP with KNN outperforms the classic LBP. While significant progress in the results accuracy recognition ratio of the CNN model compared with other methods used.

Detecting impersonation episode using teaching learning‐based optimization and support vector machine techniques

AbstractOver the last few decades, computer and internet security has become a vital area because of eye‐opening numbers of data breaches, intruders on perilous infrastructure and malware attacks that are increasing day by day. In order to monitor abnormal activities and identify unusual attacks, several security solutions have been proposed in the recent past years. To protect computer system, intrusion detection system (IDS) opens up great opportunities to determine vulnerabilities and detect anomalies in security system. For detecting new attacks or building security solutions, in this study we present a new wrapper technique for security specialists that can help to detect more complicated attacks by combining teaching learning‐based optimization with opposition learning scheme and simulated annealing method. In order to address advance attack, firstly opposition learning strategy is used to update population generation of teaching learning‐based optimization that affect the robustness of the model after that simulated annealing is integrated into the teaching learning‐based optimization. In proposed method to choose the relevant features, we have used support vector machine as a fitness function that can be helped to recognize attacks precisely. The proposed algorithm is evaluated on three popular datasets namely NSL‐KDD, ISCX 2012 and UNSW‐NB15. Experimental results show that the proposed method is superior to other existing wrapper algorithms in terms of detection rate, accuracy and false alarm rates.

Estimation and mapping of pasture biomass in Mongolia using machine learning methods

Mongolian pasture plays an essential role in the national economy. Reliable pasture biomass estimation is indispensable to support the agricultural sector and also sustainable livelihood in the country. The aim of this study is to determine an appropriate method to estimate and map pasture biomass in a forest-steppe area of Mongolia. For this purpose, machine learning methods such as random forest (RF), support vector machine (SVM), and partial least squares regression (PLSR) were compared. As data sources, spectral indices derived from Sentinel-2B image of 2019 and field-measured biomass sample datasets were used. To determine the optimal spectral predictor variables, initially, 20 spectral indices were evaluated using the PLSR. Of these, five indices (i.e. ATSAVI2, EVI, GRVI, IPVI and MSR) with the highest correlation coefficients (r ≥ 0.94) were considered for further analysis. These indices were also examined and validated by a variable importance analysis. Then, the RF, SVM, and PLSR models were applied to predict and map pasture biomass using the selected five indices. The PLSR method demonstrated the highest accuracy with coefficient of determination (R2) =0.899 and root mean square error (RMSE)=10.560 g/m2. The SVM technique showed the second highest accuracy with R2=0.837 and RMSE = 12.881 g/m2. The RF model gave the lowest accuracy with R2=0.823 and RMSE = 13.430g/m2. Our research showed that different machine learning models might be applied (because in all cases R2>0.82) for a pasture biomass estimation and mapping in the selected test site, but the best result could be achieved by the use of the PLSR.

Comparison of Satellite Imagery for Identifying Seagrass Distribution Using a Machine Learning Algorithm on the Eastern Coast of South Korea

Seagrass is an essential component of coastal ecosystems because of its capability to absorb blue carbon, and its involvement in sustaining marine biodiversity. In this study, support vector machine (SVM) technologies with corrected satellite imagery data, were applied to identify the distribution of seagrasses. Observations of seagrasses from satellite imagery were obtained using GeoEye-1, Sentinel-2 MSI level 1C, and Landsat-8 OLI satellite imagery. The satellite imagery from Google Earth has been obtained at a very high resolution, and was to be used within both the training and testing of a classification method. The optical satellite imagery must be processed for image classification, throughout which radiometric correction, sunglint, and water column adjustments were applied. We restricted the scope of the study area to a maximum depth of 10 m due to the fact that light does not penetrate beyond this level. When classifying the distribution of seagrasses present in the research region, the recently developed SVM technique achieved overall accuracy values of up to 92% (GeoEye-1), 88% (Sentinel-2 MSI level 1C), and 83% (Landsat-8 OLI), respectively. The results of the overall accuracy values are also used to evaluate classification models.

Application of Soft Computing Techniques in River Flow Modeling

Modeling of data is critical in the analysis and evaluation of hydrological behavior. River flow data is one of the most important data in explaining hydrology. Management of water resources; It takes place in the literature as an area that needs to be investigated in order to provide early warning for undesirable situations such as floods and drought. For this reason, it is of important to develop different techniques for the estimation and modeling of river flow or to make comparisons between techniques. In this study, the flow data of fourteen stations located in the Euphrates-Tigris basin between 1981 and 2010 were used. Adaptive Network Based Fuzzy Inference Systems (ANFIS), Support Vector Machine (SVM) techniques that are frequently used in the literature, and newly introduced Gaussian Process Regression (GPR), Extreme Learning Machine (ELM) and Emotional Neural Network (ENN) artificial intelligence techniques are compared. In addition, considering all performance indices, it was determined which technique gave better results with rank analysis. Although all models worked well, it was seen that the methods were ranked as ELM, GPR, ENN, SVM and ANFIS starting from the best. This has shown that ELM, GPR and ENN methods, which have been used recently in flow modeling, give better results than traditional methods with complex structures. In addition, flow values were used in the whole study and these values were examined in 3 different combinations. It was seen that the model structure that gave the best performance was the model structure that used the flow data from one, two and three days ago as an estimator. The results were analyzed with a Taylor diagram and time series graphs.

Modelling recidivism with binary models and machine learning

The article, reviewing previous study on recidivism prediction model, modeling the bina-ry outcome of recidivism with linear regression model, probit regression model and lo-gistic regression model. The application of machine learning techniques support vector machine (SVM) aids to compare the prediction result of each model and the relationships between recidivism probability and several personal characteristics have been identified. The prediction performances of models are similar with logistic regression model being the most parsimonious one and SVM method is tested to be not significantly superior in terms of prediction compared with traditional statistical models.

Data mining–based stock price prediction using hybridization of technical and fundamental analysis

PurposeThe stock market generates massive databases of various financial companies that are highly volatile and complex. To forecast daily stock values of these companies, investors frequently use technical analysis or fundamental analysis. Data mining techniques coupled with fundamental and technical analysis types have the potential to give satisfactory results for stock market prediction. In the current paper, an effort is made to investigate the accuracy of stock market predictions by using the combined approach of variables from technical and fundamental analysis for the creation of a data mining predictive model.Design/methodology/approachWe chose 381 companies from the National Stock Exchange of India's CNX 500 index and conducted a two-stage data analysis. The first stage is identifying key fundamental variables and constructing a portfolio based on that study. Artificial neural network (ANN), support vector machines (SVM) and decision tree J48 were used to build the models. The second stage entails applying technical analysis to forecast price movements in the companies included in the portfolios. ANN and SVM techniques were used to create predictive models for all companies in the portfolios. We also estimated returns using trading decisions based on the model's output and then compared them to buy-and-hold returns and the return of the NIFTY 50 index, which served as a benchmark.FindingsThe results show that the returns of both the portfolios are higher than the benchmark buy-and-hold strategy return. It can be concluded that data mining techniques give better results, irrespective of the type of stock, and have the ability to make up for poor stocks. The comparison of returns of portfolios with the return of NIFTY as a benchmark also indicates that both the portfolios are generating higher returns as compared to the return generated by NIFTY.Originality/valueAs stock prices are influenced by both technical and fundamental indicators, the current paper explored the combined effect of technical analysis and fundamental analysis variables for Indian stock market prediction. Further, the results obtained by individual analysis have also been compared. The proposed method under study can also be utilized to determine whether to hold stocks for the long or short term using trend-based research.

Analysis of bioconvection and oxytactic microorganisms in a porous cavity with nano-enhanced phase change materials and quadrant heater: Application of support vector regression based model

In this numerical study, two-dimensional simulations are performed to obtain the bioconvection flow in a quadrant heater located porous closed space saturated with nano-enhanced phase change material (NePCM). The Galerkin finite element method (GFEM) is employed to compute the governing equations for the considered system by developing a computer code. The study is performed for wide range of parameters as aspect ratio, Darcy number, Hartmann number, nanoparticle fraction, bioconvection Rayleigh number and natural convection Rayleigh number were tested. Also, the analysis is supported via artificial intelligent by using multiple support vector regression based model. The tests were performed with artificial intelligence on temperature values. After the analysis, it is found that radius of the curvilinear shaped heater plays role on both heat transfer and fluid flow, oxygen distribution, density of motile microorganism and heat capacity. Application of support vector machine technique gives good agreement with computational fluid dynamics results. Heat and mass transfer are improved with increasing of nanoparticle addition but decrease with increasing of strength of magnetic field.

Bacteria-Specific Feature Selection for Enhanced Antimicrobial Peptide Activity Predictions Using Machine-Learning Methods.

There are several classes of short peptide molecules, known as antimicrobial peptides (AMPs), which are produced during the immune responses of living organisms against various infections. In recent years, substantial progress has been achieved in applying machine-learning methods to predict the activities of AMPs against bacteria. In most investigated cases, however, the outcome is not bacterium-specific since the specific features of bacteria, such as chemical composition and structure of membranes, are not considered. To overcome this problem, we developed a new computational approach that allowed us to train several supervised machine-learning models using a specific set of data associated with peptides targeting E. coli bacteria. LASSO regression and Support Vector Machine techniques have been utilized to select, among more than 1500 physicochemical descriptors, the most important features that can be used to classify a peptide as antimicrobial or ineffective against E. coli. We then performed the classification of active versus inactive AMPs using the Support Vector classifiers, Logistic Regression, and Random Forest methods. This computational study allows us to make recommendations of how to design more efficient antibacterial drug therapies.

ANALISIS PERBANDINGAN ALGORITMA SUPPORT VECTOR MACHINE (SVM) DAN CONVOLUTIONAL NEURAL NETWORK (CNN) UNTUK SISTEM DETEKSI KATARAK

One of the most vital senses for humans is sight. Humans utilize their eyes to take in visual information that is used for a variety of tasks, but vision problems are prevalent, ranging from minor disorders to serious disorders that can result in blindness. Cataracts are one of the factors contributing to this vision loss. In Indonesia, cataracts alone account for 81 percent of cases of blindness, and 40 percent of those affected don't even realize they have them. A technique for early cataract identification using digital photographs is one way to lower the incidence of cataract-related blindness. We used Support Vector Machine (SVM) and Convolutional Neural Network (CNN) techniques to create a cataract detection system. These two techniques are used to categorize normal eye classes, immature cataracts, and mature cataracts. A digital image that has been downsized to 64x64 pixels in Joint Photographic Group (JPG) format that was taken from earlier study serves as the input data. Support Vector Machine (SVM) and Convolutional Neural Network (CNN) methods used in the cataract detection process provide optimal results. The Support Vector Machine (SVM) method itself produces an accuracy value of 96.67%, while the Convolutional Neural Network method produces a better accuracy value of 98.89%.
 Keywords: Cataract, Digital Image, Machine Learning, Support Vector Machine, Convolutional Neural Network

An unsupervised machine learning technique to identify knock from a knock signal time-frequency analysis

Knock is a characteristic phenomenon of spark ignition engines that can result in lower fuel efficiency and higher combustion temperatures. When it is not adequately identified and controlled, knock might lead to intensive engine damage. Knock detection methods are based on recognising the large in-cylinder pressure oscillations that resonate in the combustion chamber as a consequence of the rapid auto-ignition of the end-gas. However, in-cylinder pressure sensors may be invasive and expensive. In this work, a method for knock classification is proposed based on the knock sensor measurement, which is a low-cost and easy-to-implement vibration sensor. Unfortunately, the vibration of the engine block contains information from several phenomena, and only intensive knock can be identified by classical methods. To counteract the amount of noise, machine learning (ML) techniques are applied to the knock sensor signal. A training database composed of several operating conditions is used to extract the main characteristics of knock. The signal is analysed in the time-frequency domain by using the short-time Fourier transform (STFT), the features of the signal are obtained by singular value decomposition (SVD), and finally, support vector machine (SVM) techniques are used to identify outliers. Experimental data from a spark-ignited engine at 2000 rpm and 3000 rpm with spark advance steps was used to train and validate the method. Results are compared to state-of-the-art knock recognition methods based on pressure and knock sensors. The technique was able to detect 73% of the knock cycles, which was similar to the results achieved using the pressure sensor data (77%). Compared to another method using the knock sensor signal as well, the percentage of false negative cases was reduced by 20%.

Assessment of Using Artificial Neural Network and Support Vector Machine Techniques for Predicting Wave-Overtopping Discharges at Coastal Structures

Coastal defence structures play a crucial role in protecting coastal communities against extreme weather and flooding. This study investigates artificial neural network-based approaches, such as multilayer perceptron neural network (MPNN), cascade correlation neural network (CCNN), general regression neural network (GRNN), and support vector machine (SVM) with radial-bias function for estimating the wave-overtopping discharge at coastal structures featuring a straight slope ‘without a berm’. The newly developed EurOtop database was used for this study. Discriminant analysis was performed using the principal component analysis method, and Taylor diagram visualisation and other statistical analyses were performed to evaluate the models. For predicting wave-overtopping discharge, the GRNN yielded highly accurate results. As compared to the other models, the scatter index of the GRNN (0.353) was lower than that of the SVM (0.585), CCNN (0.791), and MPNN (1.068) models. In terms of the R-index, the GRNN (0.991) was superior to the SVM (0.981), CCNN (0.958), and MPNN (0.922). The GRNN results were compared with those of the previous models. An in-depth sensitivity analysis was conducted to determine the significance of each predictive variable. Furthermore, sensitivity analysis was conducted to select the optimal validation method for the GRNN model. The results revealed that both the validation methods were highly accurate, with the leave-one-out validation method outperforming the cross-validation method by only a small margin.

Identification of dispersive soils via computational intelligence

AbstractWhen exposed to water, dispersive (D) soils are eroded and washed away by underground or surface flowing waters. Although soil dispersion is due to its chemical composition, the results of the commonly used chemical method, that is, the Sherard approach, do not match with those of the popular robust Pinhole test. Due to the deficiency of the chemical method, this study aimed to employ artificial intelligence (AI)‐based approaches for predicting the D classification of the soils. To this end, a database containing 321 records of the results of chemical and Pinhole tests on borrow soil samples was collected from various construction sites in Iran. The predictive models for soil dispersion evaluation were developed using the artificial neural network (ANN) and the support vector machine (SVM). The D classification results were presented as output classes versus target classes. Through the comparison of statistical indices, it was found that the results of the proposed models conform to those of the Pinhole test. It was also shown that the ANN model is more accurate than the SVM model for predicting the dispersion potential of the soil. Furthermore, it was indicated that the new models significantly outperform the Sherard method in determining the D classification of the soil.HighlightsDispersive soils can be classified employing artificial intelligence (AI)‐based methods.AI‐based approaches have superior predictive ability in contrast to traditional approachs.ANN and SVM techniques can accurately identify soil dispersion potential.Higher accuracy of new methods compared to common Sherard approach was validated with Pinhole test results.

SVM-Based Optimization on the Number of Data Streams for Massive MIMO Systems

For millimeter-wave (mmWave) massive multiple-input and multiple-output (MIMO) systems, hardware cost and power consumption can be reduced via hybrid precoding, while keeping almost the same spectral efficiency (SE) as the fully digital precoding scheme. However, for different hardware configuration and channel environments, system SE varies with the number of transmitted data streams at the transmitter, and hence this should be determined primarily before hybrid precoding design. In this article, for single-user massive MIMO systems, a regression model based on the support vector machine technique is proposed, through which the optimal number of data streams can be obtained automatically based on the propagation characteristics of wireless channels, the number of antennas, and RF chains. To validate the correctness of the proposed model, simulation results are presented under various parameter settings.

An intelligent algorithm for lung cancer diagnosis using extracted features from Computerized Tomography images

According to the World Health Organization, lung cancer is a leading cause of death worldwide. This research aims to process the Computerized Tomography (CT) images of lung cancer patients for the early diagnosis of the disease. The images are processed using Convolutional Neural Network (CNN) in the first approach, where Artificial Neural Network (ANN) is employed to classify the images. In the second approach, the images are pre-processed and segmented before utilizing CNN and ANN. In the third method, all the pre-processed and segmented images are converted to numerical data via specific feature extraction algorithms in the last step. Besides, dimensional reduction and feature selection algorithms are employed to classify with three machine learning techniques, i.e., Gradient Boosting (GB), Random Forest (RF), and Support Vector Machine (SVM). An extensive comparative analysis is made to come up with the best technique. The comparisons are made by evaluating the methodologies on a set of lung CT scan images collected from a medical center. The results show that when either SVM or RF classification techniques are used, a 95% accuracy is obtained in diagnosing lung cancer.

Machine vision-based gradient-boosted tree and support vector regression for tool life prediction in turning

One of the essential elements of automated and intelligent machining processes is accurately predicting tool life. It also helps in achieving the goal of producing quality products with reduced production costs. This work proposes a computer vision-based tool wear monitoring and tool life prediction system using machine learning methods. Gradient-boosted trees and support vector machine (SVM) techniques are used to predict tool life. The experimental investigation on the CNC machine is conducted to study the applicability of the proposed tool wear monitoring system. Experiments are performed using workpiece material made of alloy steel and PVD-coated cutting inserts, and flank wear is monitored. An imaging system consisting of an industrial camera, lens, and LED ring light is mounted on the machine to capture tool wear zone images. Images are then processed by algorithms developed in MATLAB®. Boosted tree methods and the SVM methodology have 96% and 97% prediction accuracy, respectively. Validation tests are carried out to determine the accuracy of proposed models. It is observed that the prediction accuracy of boosted three and SVM is good, with a maximum error of 5.89% and 7.56%, respectively. The outcome of the study established that the developed system can monitor the tool wear with good accuracy and can be adopted in industries to optimize the utilization of tool inserts.

Comparative Analysis of the Implementation of Support Vector Machines and Long Short-Term Memory Artificial Neural Networks in Municipal Solid Waste Management Models in Megacities

The development of methodologies to support decision-making in municipal solid waste (MSW) management processes is of great interest for municipal administrations. Artificial intelligence (AI) techniques provide multiple tools for designing algorithms to objectively analyze data while creating highly precise models. Support vector machines and neuronal networks are formed by AI applications offering optimization solutions at different managing stages. In this paper, an implementation and comparison of the results obtained by two AI methods on a solid waste management problem is shown. Support vector machine (SVM) and long short-term memory (LSTM) network techniques have been used. The implementation of LSTM took into account different configurations, temporal filtering and annual calculations of solid waste collection periods. Results show that the SVM method properly fits selected data and yields consistent regression curves, even with very limited training data, leading to more accurate results than those obtained by the LSTM method.