Proposed Support Vector Machine Classifier Research Articles

Wafer Defect Identification with Optimal Hyper-Parameter Tuning of Support Vector Machine Using the Deep Feature of ResNet 101

As semiconductor processing technologies continue to advance, semiconductor wafers are becoming more densely packed and intricate, resulting in a higher incidence of surface imperfections. Therefore, it is crucial to detect these defects early and accurately classify them to pinpoint the root causes of the defects in the manufacturing process, ultimately leading to improved yield. Therefore, defect detection is critical in the industrial production of monocrystalline silicon. This study employs deep learning techniques to propose a framework for detecting defects on silicon wafers, focusing on optimizing the hyperparameters of support vector machines (SVM). Three methods were utilized to fine-tune the SVM parameters: Bayesian optimization, grid search, and random search techniques. This study demonstrates how selecting optimal values for SVM parameters can lead to better classification. Additionally, real manufacturing data were utilized to evaluate the performance of the proposed SVM classifier, with a comparison to state-of-the-art techniques in the field. By using deep features from ResNet 101 and a support vector machine, this work achieves 74.5% accuracy in identifying wafer defects without employing any optimization technique. However, the performance of the model was further improved by utilizing the random search optimization technique, which yielded the best result among the three optimization techniques tested, with an accuracy of 88.1%.

Mitigation of Attacks via Improved Network Security in IoT Network using Machine Learning

In this paper, we develop a support vector machine (SVM) based attack mitigation technique from the IoT network. The SVM aims to classify the features related to the attacks based on pre-processed and feature extracted information. The simulation is conducted in terms of accuracy, precision, recall and f-measure over KDD datasets. The results show that the proposed SVM classifier obtains high grade of classification accuracy in both training and testing datasets.

Wafer Defect Identification with Optimal Hyper-Parameter Tuning of Support Vector Machine using the Deep Feature of ResNet 101

As semiconductor processing technologies continue to advance, semiconductor wafers are becoming more densely packed and intricate, resulting in a higher incidence of surface imperfections. Therefore, it is crucial to detect these defects early and accurately classify them to pinpoint the root causes of the defects in the manufacturing process, ultimately leading to improved yield. Therefore, defect detection is critical in the industrial production of monocrystalline silicon. This study employs deep learning techniques to propose a framework for detecting defects on silicon wafers, focusing on optimizing the hyperparameters of support vector machines (SVM). Three methods were utilized to fine-tune the SVM parameters: Bayesian optimization, grid search, and random search techniques. This study demonstrates how selecting optimal values for SVM parameters can lead to better classification. Additionally, real manufacturing data were utilized to evaluate the performance of the proposed SVM classifier, with a comparison to state-of-the-art techniques in the field. By using deep features from ResNet 101 and a support vector machine, this work achieves 74.5% accuracy in identifying wafer defects without employing any optimization technique. However, the performance of the model was further improved by utilizing the random search optimization technique, which yielded the best result among the three optimization techniques tested, with an accuracy of 88.1%.

Pressure-Sensor-Based Sleep Status and Quality Evaluation System

Sleep quality evaluation is a major approach to clinical diagnosis of different types of sleep andmental disturbances.How to render it possible for test subjects to complete sleep status detection and quality assessment in a natural environment is what this paper mainly highlights. Clinically, sleep disturbances are analyzed by the so-called Polysomnography (PSG). However, due to limitations in space and high cost, as well as the fact that test subjects must wear various physiological detection equipment during the test, the collected signal is susceptible to interference due to uncomfortable sensations. As such it usually does not reflect the realistic situation of the subject. For this reason,we propose in this paper a pressure-sensor-based smart mattress to realize sleep status detection and quality evaluation. Regarding sleep posture recognition, a proposed convolutional neural network (CNN) model in conjunction with a pressure distribution image formed by the pressure sensing matrix are applied. With respect to assessing the subjects’ time in bed, a support vectormachine (SVM) is used to determine sitting or lying postures, and further recognize the actual time in bed. As for sleep quality assessment, Fuzzy inference is adopted in this paper base on a set of four predefined sleep parameters. Compared with some of prior arts, the proposed systemdoes not require the test subjects to wear any equipment, and as such the subjects can complete the test in a natural environment. Experimental results show that the accuracy of the proposed SVM classifier for differentiating sitting and lying posture and that of CNN model for the recognition of four different sleep postures can be up to 99.986% and 96.987%, respectively. With the proposed model, precise sleep parameters, including time in bed, number of times of bed-leaving, number of times of body movements all night, standard deviation of time interval between body movements, and sleep posture can be provided. Moreover, the system does not use devices such as microphones or surveillance cameras to collect the data of the test subjects; thus, there is no concern about infringing the privacy of the subjects. It is believed that the system will be of considerable help and serve as an aid to clinical diagnosis of sleep disturbance.

Performance Analysis of Support Vector Machine Learning Based Carrier Aggregation Resource scheduling in 5G Mobile Communication

The 5G cellular system will enable heterogeneous networks that incorporate 5G, 4G, Wi-Fi, and other telecommunication connectivity. This will result in huge changes in existing infrastructure and architecture network. Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), and massive Machine Type Communication (mMTC) are some of the 5G's basic application cases. The variety of heterogeneous networks makes it challenging to manage, network traffic, inter-cell, and intra-cell interfaces. Due to diverse interfaces and data transmission rates, network traffic in heterogeneous networks can be difficult to manage and control. Advanced techniques including Machine Learning (ML), Artificial Intelligence (AI), and Deep Learning (DL) are used in the 5G Network to overcome such issues. In this paper, a Support Vector Machine (SVM) learning-based resource allocation approach for heterogeneous networks is designed. A machine learning method is proposed to enhance the performance of 5G NR mobile networks. This method is used to upgrade the throughput of the overall networks. The SVM learning function detects by splitting network data of channel state information (CSI). The considered channel state data of two variables as labels with X1 and X2 and trained using SVM-LINEAR, SVM-RBF, and SVM-SIGMOID. The proposed SVM classifier by splitting data for training 80% and testing 20%. A Static Error rate is set as a threshold limit at 10% (0.10) of BLER (Block Error Rate). If the error rate passed the threshold, then update Carrier Aggregation with aggregated component else set carrier aggregation with ordinary carrier component values to Zero. The proposed approach further compares with the round robin (RR), best channel quality indicator (BCQI), fractional frequency reuse (FFR), MU-MIMO, and beamforming resource scheduling techniques. The innovative results indicated that the proposed approach enhanced spectral efficiency, average cell throughput, edge throughput peak throughput, and fairness index.

BFS-SVM Classifier for QoS and Resource Allocation in Cloud Environment

In cloud computing Resource allocation is a very complex task. Handling the customer demand makes the challenges of on-demand resource allocation. Many challenges are faced by conventional methods for resource allocation in order to meet the Quality of Service (QoS) requirements of users. For solving the about said problems a new method was implemented with the utility of machine learning framework of resource allocation by utilizing the cloud computing technique was taken in to an account in this research work. The accuracy in the machine learning algorithm can be improved by introducing Bat Algorithm with feature selection (BFS) in the proposed work, this further reduces the inappropriate features from the data. The similarities that were hidden can be demoralized by the Support Vector Machine (SVM) classifier which is also determine the subspace vector and then a new feature vector can be predicted by using SVM. For an unexpected circumstance SVM model can make a resource allocation decision. The efficiency of proposed SVM classifier of resource allocation can be highlighted by using a single-cell multiuser massive Multiple-Input Multiple Output (MIMO) system, with beam allocation problem as an example. The proposed resource allocation based on SVM performs efficiently than the existing conventional methods; this has been proven by analysing its results.

A Machine Learning Approach for the Classification of Falls and Activities of Daily Living in Agricultural Workers

Population aging is a global trend, and the highest proportion of elderly people in the workforce per unit of population is found in agricultural areas. However, few systematic studies have been conducted on farmer falls in the field of agricultural machinery. This study focuses on the application of classification methods for monitoring devices to detect fall/nonfall movements of farmworkers, where agricultural biomechanical factors are considered in detecting activities of daily living. In this study, we recorded and analyzed original acquisition datasets of signals obtained from two accelerometers and one gyroscope for 40 healthy individuals who performed various falls and activities of daily living (ADLs). Spatial characteristics were used to train the machine-learning classifiers to distinguish between fall and non-fall events. Supervised machine learning experiments evaluated the effectiveness of the proposed approach: the k-nearest neighbors (kNN) and support vector machine (SVM) algorithms achieved roc auc-scores of 0.999 in differentiating between falls and ADLs (binary-class classification). Moreover, an artificial neural network (ANN) classifier showed the highest performance in terms of classification roc auc-scores of 1.0. The evaluation metric demonstrated the highest performance in the analysis and evaluation of the signal obtained from the S2 acceleration sensor with a measurement range of ±16 g. The proposed SVM classifier evaluations showed a 0.988 roc auc-score for sensor tests in multi-class classification, along with the highest performance in terms of the F1-score and Matthews Correlation Coefficient (MCC) over 84% in the multi-class classification model for distinguishing each of ADLs and Fall using ±16 g acceleration sensor.

Pulmonary Nodule Detection Using V-Net and High-Level Descriptor Based SVM Classifier

Early detection of the pulmonary nodule is critical to increase the five-year survival rate of lung cancer. Many computer-aided diagnosis (CAD) systems have been proposed for nodule detection to assist radiologists in diagnosis. Along this direction, this paper proposes a novel automated pulmonary nodule detection model using the modified V-Nets and a high-level descriptor based support vector machine (SVM) classifier. The former is for nodule candidate detection and the latter is for false positive (FP) reduction. A hard mining scheme for retraining is devised to improve the FP reduction performance. The proposed SVM classifier, which employs more critical features of CT images, performs superior in FP reduction than other SVM based classifiers and CNN classifiers. Experimental results using the LIDC-IDRI dataset are presented to demonstrate the effectiveness of the proposed CAD model.

Segmentation and Classification of Image Abnormalities in Retinal Fundus using Discrete Wavelet Transforms

Glaucoma disease diagnosis greatly based on the accurate retinal image segmentation and classification of images. Segmentation means to divide the images into a patchwork of regions, each of which is “homogeneous”, that is the “same” in some sense. Using discrete wavelet transform, the segmented images are classified by Support Vector Machine (SVM) classifiers to classify the Glaucoma images.The proposed Support Vector Machine classifier is used to extract the information rely on the Region of Interest (ROI) from original retinal fundus image. Thus the classification result are used to find the normal and abnormal image and also to compute the normal and abnormal accuracies.We observed an accuracy of around 93% using data set by SVM classifier.

Supervised cross-fusion method: a new triplet approach to fuse thermal, radar, and optical satellite data for land use classification

This study presents a new fusion method namely supervised cross-fusion method to improve the capability of fused thermal, radar, and optical images for classification. The proposed cross-fusion method is a combination of pixel-based and supervised feature-based fusion of thermal, radar, and optical data. The pixel-based fusion was applied to fuse optical data of Sentinel-2 and Landsat 8. According to correlation coefficient (CR) and signal to noise ratio (SNR), among the used pixel-based fusion methods, wavelet obtained the best results for fusion. Considering spectral and spatial information preservation, CR of the wavelet method is 0.97 and 0.96, respectively. The supervised feature-based fusion method is a fusion of best output of pixel-based fusion level, land surface temperature (LST) data, and Sentinel-1 radar image using a supervised approach. The supervised approach is a supervised feature selection and learning of the inputs based on linear discriminant analysis and sparse regularization (LDASR) algorithm. In the present study, the non-negative matrix factorization (NMF) was utilized for feature extraction. A comparison of the obtained results with state of the art fusion method indicated a higher accuracy of our proposed method of classification. The rotation forest (RoF) classification results improvement was 25% and the support vector machine (SVM) results improvement was 31%. The results showed that the proposed method is well classified and separated four main classes of settlements, barren land, river, river bank, and even the bridges over the river. Also, a number of unclassified pixels by SVM are very low compared to other classification methods and can be neglected. The study results showed that LST calculated using thermal data has had positive effects on improving the classification results. By comparing the results of supervised cross-fusion without using LST data to the proposed method results, SVM and RoF classifiers showed 38% and 7% of classification improvement, respectively.

Hybrid enhanced ICA & KSVM based brain tumor image segmentation

<span>Medical image processing is an important aspect in diagnosis and treatment strategy. The tremendous volume of medical data has accelerated the need for automated analysis of this image, more so in the case Magnetic Resonance Imaging (MRI). An improved K-means algorithm and EM algorithm have been combined in the proposed approach to produce a hybrid strategy for better clustering and segmentation using Enhanced ICA. A classifier for based on Support Vector Machine (SVM) has been formulated and employed for the classification of brain tumors in Magnetic Resonance Images (MRI). The proposed SVM classifier used a kernel in the form of Gaussian radial basis function kernel (GRB kernel) to improve the classifier performance. The performance of the classifier has been validated through expert clinical opinion and calculation of performance measures. The results amply illustrate the suitability of the proposed classifier.</span>

Performing Subjectivity Classification in Text Using Support Vector Machine

In this paper, I address the problem of the subjectivity classification in text. The subjective text is opinion bearing, whereas the objective text is text without expressing opinions. The supervised learning technique namely, Support Vector Machine (SVM) is used to classify the text as subjective and objective. A publically available dataset of drug reviews is used to conduct the experiments using WEKA platform. The experimental results show that the proposed SVM classifier performed better than the other classifiers

Crop Classification Based on Differential Characteristics of <inline-formula> <tex-math notation="LaTeX">$H/\alpha$ </tex-math> </inline-formula> Scattering Parameters for Multitemporal Quad- and Dual-Polarization SAR Images

Crop-type classification is one of the most significant applications in polarimetric synthetic aperture radar (PolSAR) imagery. As a remote sensing technique, PolSAR has been proved to have the ability to provide high-resolution information of illustrated objects. However, single-temporal PolSAR data are restricted to provide sufficient information for crop identification due to the complicated condition of varying morphology within various growing stages. With an increasing number of spaceborne PolSAR systems launched, a large amount of real PolSAR data are being generated and used to provide great opportunities for multitemporal analysis. The main contribution of this paper is to improve crop classification accuracy with various features of classical $H$ / $\alpha$ parameters. First, in order to deal with dual-PolSAR data, $H$ / $\alpha$ decomposition algorithm for quad-PolSAR is modified to suit to the case of dual polarization. Second, according to the differential scattering characteristics of main crops, a new parameter is innovatively defined to measure the differential characteristics in the $H$ / $\alpha$ classification plane. Third, crop types are discriminated by applying a supervised classification method with the newly defined parameter. Furthermore, the correctness of the parameter is verified with simulated and real Sentinel-1 data as well as AirSAR data. Finally, the performances of the classification method are investigated by the comparison with complex Wishart, Freeman–Wishart, and support vector machine (SVM) classifiers. Hence, the experimental results show that the proposed method and SVM classifier with the newly defined parameter have the ability to improve crop classification accuracy.

PCG Classification Using Multidomain Features and SVM Classifier.

This paper proposes a method using multidomain features and support vector machine (SVM) for classifying normal and abnormal heart sound recordings. The database was provided by the PhysioNet/CinC Challenge 2016. A total of 515 features are extracted from nine feature domains, i.e., time interval, frequency spectrum of states, state amplitude, energy, frequency spectrum of records, cepstrum, cyclostationarity, high-order statistics, and entropy. Correlation analysis is conducted to quantify the feature discrimination abilities, and the results show that “frequency spectrum of state”, “energy”, and “entropy” are top domains to contribute effective features. A SVM with radial basis kernel function was trained for signal quality estimation and classification. The SVM classifier is independently trained and tested by many groups of top features. It shows the average of sensitivity, specificity, and overall score are high up to 0.88, 0.87, and 0.88, respectively, when top 400 features are used. This score is competitive to the best previous scores. The classifier has very good performance with even small number of top features for training and it has stable output regardless of randomly selected features for training. These simulations demonstrate that the proposed features and SVM classifier are jointly powerful for classifying heart sound recordings.

Smartphone Based Fall Detection Using Thresholding and Machine Learning Approach

Fall detection is an approach, which is used to detect fall event when some person falls unconsciously and sends this information to the caretaker of that person. Fall detection methods can be broadly classified in machine learning based methods and threshold based methods. Our method is based on the combined approach of both threshold based method and machine learning based method. In this proposed method, first we calculate the variance of every five continuous accelerometer values because there is a sudden change in acceleration values during fall which increases variance very suddenly. Then we have used support vector machine (SVM) and K-nearest neighbor (KNN) classifiers to classify ADL and fall activities. In this proposed method SVM classifier using Gaussian kernel function performs better than other classifiers.

Acoustic NLOS Identification Using Acoustic Channel Characteristics for Smartphone Indoor Localization.

As the demand for indoor localization is increasing to support our daily life in large and complex indoor environments, sound-based localization technologies have attracted researchers’ attention because they have the advantages of being fully compatible with commercial off-the-shelf (COTS) smartphones, they have high positioning accuracy and low-cost infrastructure. However, the non-line-of-sight (NLOS) phenomenon poses a great challenge and has become the technology bottleneck for practical applications of acoustic smartphone indoor localization. Through identifying and discarding the NLOS measurements, the positioning performance can be improved by incorporating only the LOS measurements. In this paper, we focus on identifying NLOS components by characterizing the acoustic channels. Firstly, by analyzing indoor acoustic propagations, the changes of acoustic channel from the line-of-sight (LOS) condition to the NLOS condition are characterized as the difference of channel gain and channel delay between the two propagation scenarios. Then, an efficient approach to estimate relative channel gain and delay based on the cross-correlation method is proposed, which considers the mitigation of the Doppler Effect and reduction of the computational complexity. Nine novel features have been extracted, and a support vector machine (SVM) classifier with a radial-based function (RBF) kernel is used to realize NLOS identification. The experimental result with an overall 98.9% classification accuracy based on a data set with more than 10 thousand measurements shows that the proposed identification approach and features are effective in acoustic NLOS identification for acoustic indoor localization via a smartphone. In order to further evaluate the performance of the proposed SVM classifier, the performance of an SVM classifier is compared with that of traditional classifiers based on logistic regression (LR) and linear discriminant analysis (LDA). The results also show that a SVM with the RBF kernel function method outperforms others in acoustic NLOS identification.

Wavelet statistical texture features‐based segmentation and classification of brain computed tomography images

A computer software system is designed for segmentation and classification of benign and malignant tumour slices in brain computed tomography images. In this study, the authors present a method to select both dominant run length and co-occurrence texture features of wavelet approximation tumour region of each slice to be segmented by a support vector machine (SVM). Two-dimensional discrete wavelet decomposition is performed on the tumour image to remove the noise. The images considered for this study belong to 208 tumour slices. Seventeen features are extracted and six features are selected using Student's t-test. This study constructed the SVM and probabilistic neural network (PNN) classifiers with the selected features. The classification accuracy of both classifiers are evaluated using the k fold cross validation method. The segmentation results are also compared with the experienced radiologist ground truth. Quantitative analysis between ground truth and the segmented tumour is presented in terms of segmentation accuracy and segmentation error. The proposed system provides some newly found texture features have an important contribution in classifying tumour slices efficiently and accurately. The experimental results show that the proposed SVM classifier is able to achieve high segmentation and classification accuracy effectiveness as measured by sensitivity and specificity.

Wafer Classification Using Support Vector Machines

Increasing yield is a primary concern to integrated circuit manufacturing companies as it dictates the readiness of a new process for high volume manufacturing. In order to expedite the process of discovering yield issues, companies have started looking for ways to perform early prediction for such issues. This paper suggests the use of the support vector machines (SVMs) for early wafer classification. The choice of SVM is motivated by the model's ability to effectively classify multivariate, multimodal, and inseparable data points. This model uses multidimensional hyperplanes to separate and classify wafers into low-yield and high-yield classes. This paper includes a proposal for how the classification model can be applied for yield classification and how it can be adaptively updated in a manufacturing environment. We show how the values for the SVM parameters can be selected for best yield classification. Furthermore, performance evaluation is conducted on real manufacturing data, comparing the proposed SVM classifier to state of the art. Results show that in all cases, SVM consistently outperforms other methods with and without adaptive model updates. The experiments also show that all classifiers' performances depend on yield thresholds. It is also shown that the classification model can be built and executed using a reduced set without compromising its accuracy.

Land-cover classification of partly missing data using support vector machines

Land-cover classification based on multi-temporal satellite images for scenarios where parts of the data are missing due to, for example, clouds, snow or sensor failure has received little attention in the remote-sensing literature. The goal of this article is to introduce support vector machine (SVM) methods capable of handling missing data in land-cover classification. The novelty of this article consists of combining the powerful SVM regularization framework with a recent statistical theory of missing data, resulting in a new method where an SVM is trained for each missing data pattern, and a given incomplete test vector is classified by selecting the corresponding SVM model. The SVM classifiers are evaluated on Landsat Enhanced Thematic Mapper Plus (ETM + ) images covering a scene of Norwegian mountain vegetation. The results show that the proposed SVM-based classifier improves the classification accuracy by 5–10% compared with single image classification. The proposed SVM classifier also outperforms recent non-parametric k-nearest neighbours (k-NN) and Parzen window density-based classifiers for incomplete data by about 3%. Moreover, since the resulting SVM classifier may easily be implemented using existing SVM libraries, we consider the new method to be an attractive choice for classification of incomplete data in remote sensing.

Wood defect classification based on image analysis and support vector machines

This paper addresses the issue of automatic wood defect classification. A tree-structure support vector machine (SVM) is proposed to classify four types of wood knots by using images captured from lumber boards. Simple and effective features are proposed and extracted by partitioning the knot images into three distinct areas, followed by utilizing a novel order statistic filter to yield an average pseudo color feature in each area. Excellent results have been obtained for the proposed SVM classifier that is trained by 800 wood knot images. Performance evaluation has shown that the proposed SVM classifier resulted in an average classification rate of 96.5% and false alarm rate of 2.25% over 400 test knot images. Future work will include more extensive tests on large data set and the extension of knot types.