Implementation Of Support Vector Machine Research Articles

Evaluating conversion from mild cognitive impairment to Alzheimer's disease with structural MRI: a machine learning study.

Alzheimer's disease is a disabling neurodegenerative disorder for which no effective treatment currently exists. To predict the diagnosis of Alzheimer's disease could be crucial for patients' outcome, but current Alzheimer's disease biomarkers are invasive, time consuming or expensive. Thus, developing MRI-based computational methods for Alzheimer's disease early diagnosis would be essential to narrow down the phenotypic measures predictive of cognitive decline. Amnestic mild cognitive impairment (aMCI) is associated with higher risk for Alzheimer's disease, and here, we aimed to identify MRI-based quantitative rules to predict aMCI to possible Alzheimer's disease conversion, applying different machine learning algorithms sequentially. At baseline, T1-weighted brain images were collected for 104 aMCI patients and processed to obtain 146 volumetric measures of cerebral grey matter regions [regions of interest (ROIs)]. One year later, patients were classified as converters (aMCI-c = 32) or non-converters, i.e. clinically and neuropsychologically stable (aMCI-s = 72) based on cognitive performance. Feature selection was performed by random forest (RF), and the identified seven ROIs volumetric data were used to implement support vector machine (SVM) and decision tree (DT) classification algorithms. Both SVM and DT reached an average accuracy of 86% in identifying aMCI-c and aMCI-s. DT found a critical threshold volume of the right entorhinal cortex (EC-r) as the first feature for differentiating aMCI-c/aMCI-s. Almost all aMCI-c had an EC-r volume <1286 mm3, while more than half of the aMCI-s patients had a volume above the identified threshold for this structure. Other key regions for the classification between aMCI-c/aMCI-s were the left lateral occipital (LOC-l), the middle temporal gyrus and the temporal pole cortices. Our study reinforces previous evidence suggesting that the morphometry of the EC-r and LOC-l best predicts aMCI to Alzheimer's disease conversion. Further investigations are needed prior to deeming our findings as a broadly applicable predictive framework. However, here, a first indication was derived for volumetric thresholds that, being easy to obtain, may assist in early identification of Alzheimer's disease in clinical practice, thus contributing to establishing MRI as a useful non-invasive prognostic instrument for dementia onset.

Improved Histogram of Oriented Gradient (HOG) Feature Extraction for Facial Expressions Classification

Facial expression classification system is one of the implementations of machine learning (ML) that takes facial expression datasets, undergoes training, and then utilizes the trained results to recognize facial expressions in new facial images. The recognized facial expressions include anger, contempt, disgust, fear, happy, sadness, and surprise expressions. The method employed for facial feature extraction utilizes histogram-oriented gradient (HOG). This study proposes an enhancement method for HOG feature extraction by reducing the feature dimension into multiple sub-features based on gradient orientation intervals, referred to as HOG channel (HOG-C). Classifier testing techniques are divided into two methods for comparison—support vector machines (SVM) with HOG features and SVM with HOG-C features. The testing results demonstrate that SVM with HOG achieves an accuracy of 99.9% with an average training time of 18.03 minutes, while SVM with HOG-C attains a 100% accuracy with an average training time of 18.09 minutes. The testing outcomes reveal that the implementation of SVM with HOG-C successfully enhances accuracy for facial expression classification.

Emergency patient forecasting with models based on support vector machines

&lt;p&gt;Understanding the dynamic nature of the influx of patients is crucial for efficiently managing supplies, medical personnel, and infrastructure in an emergency room (ER). While overestimation can lead to resource wastage, underestimation can result in shortages and compromised service quality. This study addresses emergency patient forecast by means of implementing support vector machine (SVM) algorithms. Along four phases (analysis, design, development, and validation), more than 50,000 ER records were preprocessed and analyzed. Traditional error metrics such as mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE) were utilized alongside monthly consolidated forecasts. To benchmark performance, actual values and forecasts derived from linear regression (LR) models were used. Experiments revealed that LR models had lower errors compared to SVM models. However, monthly consolidated forecasts showed that SVM-based models underestimated less than LR-based models. In conclusion, SVM-based models could help planners to accurately estimate the requirements for supplies and medical personnel during the period under study.&lt;/p&gt;

An Ultralow-Power Real-Time Machine Learning Based fNIRS Motion Artifacts Detection.

Due to iterative matrix multiplications or gradient computations, machine learning modules often require a large amount of processing power and memory. As a result, they are often not feasible for use in wearable devices, which have limited processing power and memory. In this study, we propose an ultralow-power and real-time machine learning-based motion artifact detection module for functional near-infrared spectroscopy (fNIRS) systems. We achieved a high classification accuracy of 97.42%, low field-programmable gate array (FPGA) resource utilization of 38354 lookup tables and 6024 flip-flops, as well as low power consumption of 0.021 W in dynamic power. These results outperform conventional CPU support vector machine (SVM) methods and other state-of-the-art SVM implementations. This study has demonstrated that an FPGA-based fNIRS motion artifact classifier can be exploited while meeting low power and resource constraints, which are crucial in embedded hardware systems while keeping high classification accuracy.

Pneumonia detection by binary classification: classical, quantum, and hybrid approaches for support vector machine (SVM)

Early diagnosis of pneumonia is crucial to increase the chances of survival and reduce the recovery time of the patient. Chest X-ray images, the most widely used method in practice, are challenging to classify. Our aim is to develop a machine learning tool that can accurately classify images as belonging to normal or infected individuals. A support vector machine (SVM) is attractive because binary classification can be represented as an optimization problem, in particular as a Quadratic Unconstrained Binary Optimization (QUBO) model, which, in turn, maps naturally to an Ising model, thereby making annealing—classical, quantum, and hybrid—an attractive approach to explore. In this study, we offer a comparison between different methods: (1) a classical state-of-the-art implementation of SVM (LibSVM); (2) solving SVM with a classical solver (Gurobi), with and without decomposition; (3) solving SVM with simulated annealing; (4) solving SVM with quantum annealing (D-Wave); and (5) solving SVM using Graver Augmented Multi-seed Algorithm (GAMA). GAMA is tried with several different numbers of Graver elements and a number of seeds using both simulating annealing and quantum annealing. We found that simulated annealing and GAMA (with simulated annealing) are comparable, provide accurate results quickly, competitive with LibSVM, and superior to Gurobi and quantum annealing.

Optimizing ‘Explorer’ Rose Production Data with SVM in Smart Agriculture

In the context of modern flower cultivation, this study leverages the power of Support Vector Machines (SVM) to revolutionize the production process. By deploying SVM technology, the study aims to optimize various stages of flower production, with a keen focus on the ever-popular ‘Explorer’ rose variety. The SVM model showcases an exceptional precision rate, achieving a remarkable 99.21% accuracy in classifying production stages. This precision is further validated through k-fold cross-validation, resulting in an impressive average precision of 99.34%. The successful implementation of SVM underscores its potential to not only enhance the quality and profitability of rose production but also serve as a stepping stone for prospective investigations into the seamless integration of other machine learning algorithms within the realm of smart farming.

Quantum Machine Learning—An Overview

Quantum computing has been proven to excel in factorization issues and unordered search problems due to its capability of quantum parallelism. This unique feature allows exponential speed-up in solving certain problems. However, this advantage does not apply universally, and challenges arise when combining classical and quantum computing to achieve acceleration in computation speed. This paper aims to address these challenges by exploring the current state of quantum machine learning and benchmarking the performance of quantum and classical algorithms in terms of accuracy. Specifically, we conducted experiments with three datasets for binary classification, implementing Support Vector Machine (SVM) and Quantum SVM (QSVM) algorithms. Our findings suggest that the QSVM algorithm outperforms classical SVM on complex datasets, and the performance gap between quantum and classical models increases with dataset complexity, as simple models tend to overfit with complex datasets. While there is still a long way to go in terms of developing quantum hardware with sufficient resources, quantum machine learning holds great potential in areas such as unsupervised learning and generative models. Moving forward, more efforts are needed to explore new quantum learning models that can leverage the power of quantum mechanics to overcome the limitations of classical machine learning.

Support Vector Machine – Recursive Feature Elimination for Feature Selection on Multi-omics Lung Cancer Data

Biological data obtained from sequencing technologies is growing exponentially. Multi-omics data is one of the biological data that exhibits high dimensionality, or more commonly known as the curse of dimensionality. The curse of dimensionality occurs when the dataset contains many features or attributes but with significantly fewer samples or observations. The study focuses on mitigating the curse of dimensionality by implementing Support Vector Machine – Recursive Feature Elimination (SVM-RFE) as the selected feature selection method in the lung cancer (LUSC) multi-omics dataset integrated from three single omics dataset comprising genomics, transcriptomics and epigenomics, and assess the quality of the selected feature subsets using SDAE and VAE deep learning classifiers. In this study, the LUSC datasets first undergo data pre-processing, including checking for missing values, normalization, and removing zero variance features. The cleaned LUSC datasets are then integrated to form a multi-omics dataset. Feature selection was performed on the LUSC multi-omics data using SVM-RFE to select several optimal feature subsets. The five smallest feature subsets (FS) are used in classification using SDAE and VAE neural networks to assess the quality of the feature subsets. The results show that all 5 VAE models can obtain an accuracy and AUC score of 1.000, while only 2 out of 5 SDAE models (FS 1000 &amp; 4000) can do so. 3 out of 5 SDAE models have an AUC score of 0.500, indicating zero capability in separating the binary class labels. The study concludes that a fine-tuned supervised learning VAE model has better capability in classification tasks compared to SDAE models for this specific study. Additionally, 1000 and 4000 are the two most optimal feature subsets selected by the SVM-RFE algorithm. The SDAE and VAE models built with these feature subsets achieve the best classification results.

Imbalanced Data Classification Using SVM Based on Improved Simulated Annealing Featuring Synthetic Data Generation and Reduction

Imbalanced data classification is one of the major problems in machine learning. This imbalanced dataset typically has significant differences in the number of data samples between its classes. In most cases, the performance of the machine learning algorithm such as Support Vector Machine (SVM) is affected when dealing with an imbalanced dataset. The classification accuracy is mostly skewed toward the majority class and poor results are exhibited in the prediction of minority-class samples. In this paper, a hybrid approach combining data pre-processing technique and SVM algorithm based on improved Simulated Annealing (SA) was proposed. Firstly, the data pre-processing technique which primarily aims at solving the resampling strategy of handling imbalanced datasets was proposed. In this technique, the data were first synthetically generated to equalize the number of samples between classes and followed by a reduction step to remove redundancy and duplicated data. Next is the training of a balanced dataset using SVM. Since this algorithm requires an iterative process to search for the best penalty parameter during training, an improved SA algorithm was proposed for this task. In this proposed improvement, a new acceptance criterion for the solution to be accepted in the SA algorithm was introduced to enhance the accuracy of the optimization process. Experimental works based on ten publicly available imbalanced datasets have demonstrated higher accuracy in the classification tasks using the proposed approach in comparison with the conventional implementation of SVM. Registering at an average of 89.65% of accuracy for the binary class classification has demonstrated the good performance of the proposed works.

Support vector machine fault diagnosis based on sparse scaling convex hull

In solving the problems encountered when implementing support vector machine (SVM) fault diagnosis, the convex-hull-covering model of the dataset cannot reflect the effective distribution of the samples, and the fault identification accuracy of the original high-dimensional feature set is low. A sparse scaling convex-hull based SVM classification method is proposed and applied to the fault diagnosis of roller bearings. The dimensionality reduction of the features of the sample set is carried out by the random forest (RF) algorithm. First, the optimized sample subsets are obtained by sparse approximation, and the reduction coefficient of the convex hull of the optimized sample set is adjusted, hence the convex hulls of various sample sets are linearly separable. Second, to solve the problem of low fault recognition accuracy of the original high-dimensional feature set, the importance of features is evaluated by RF, and some redundant features are removed. Finally, the SVM model is constructed by the closest points between the convex hulls. Through fault diagnosis on two different bearing datasets, the experimental results and related theories show that the proposed method has high performance in bearing fault diagnosis.

Vision-Based Detection and Classification of Used Electronic Parts

Economic and environmental sustainability is becoming increasingly important in today's world. Electronic waste (e-waste) is on the rise and options to reuse parts should be explored. Hence, this paper presents the development of vision-based methods for the detection and classification of used electronics parts. In particular, the problem of classifying commonly used and relatively expensive electronic project parts such as capacitors, potentiometers, and voltage regulator ICs is investigated. A multiple object workspace scenario with an overhead camera is investigated. A customized object detection algorithm determines regions of interest and extracts data for classification. Three classification methods are explored: (a) shallow neural networks (SNNs), (b) support vector machines (SVMs), and (c) deep learning with convolutional neural networks (CNNs). All three methods utilize 30 × 30-pixel grayscale image inputs. Shallow neural networks achieved the lowest overall accuracy of 85.6%. The SVM implementation produced its best results using a cubic kernel and principal component analysis (PCA) with 20 features. An overall accuracy of 95.2% was achieved with this setting. The deep learning CNN model has three convolution layers, two pooling layers, one fully connected layer, softmax, and a classification layer. The convolution layer filter size was set to four and adjusting the number of filters produced little variation in accuracy. An overall accuracy of 98.1% was achieved with the CNN model.

Implementation of MFCC and SVM for Voice Command Recognition as Control on Mobile Robot

The mobile robot is a system that can move according to function and task. An example is an industrial robot taking objects using a remote control system. Robots controlled using a manual remote system are generally carried out on mobile robots. Many researchers have developed manual control methods, such as image or sound-based robot control. In this study, the mobile robot was applied in an unobstructed room and controlled using voice commands. The methods used are Mel-Frequency Cepstral Coefficients (MFCC) and Support Vector Machine (SVM). MFCC is a characteristic identification of voice command patterns such as “forward”, “backward”, “left”, “right”, and “stop.” SVM is used to recognize voice command patterns based on the value of the MFCC for each pattern. The experiment has been carried out 50 times with a success rate of 96%. Overall the robot can be controlled by voice commands with good movement.

QSAR Study on Diacylgycerol Acyltransferase-1 (DGAT-1) Inhibitor as Anti-diabetic using PSO-SVM Methods

Diabetes mellitus is a chronic disease that can occurred to anyone. Up until now, there are no specific drugs that have been found which can completely cure diabetes. One of the possible steps to treat diabetes mellitus is by inhibiting the growth of Diacylglycerol Acyltransferase-1 (DGAT-1) enzyme. This study aims to build a QSAR model on DGAT-1 inhibitors as anti-diabetic using Particle Swarm Optimization (PSO) and Support Vector Machine (SVM). Acyl-CoA: DGAT1 is a mikrosomal enzyme in lipogenesis which is increased in metabolically active cells to meet nutrient requirements. Microsomal-enzymes-that have an-important in the triglyceride-synthesis process-of 1,2-diacylglycerol by-catalyzing-acyl-coa-dependent-acylations as anti-diabetics. The dataset used in this study consists of 228 samples containing molecular structures and their inhibitor activities. We reduce the number of features by removing features with a standard deviation less than the threshold value, followed by the PSO algorithm. The best-predicted result is obtained through the implementation of SVM with RBF kernel, with the score of and are 0.75 and 0.67, respectively.&#x0D;

Implementation of a sensor node for monitoring and classification of physiological signals in an edge computing system

We describe the design and development of sensor nodes, based on Edge computing technologies, for the processing and classification of events detected in physiological signals such as the electrocardiographic signal (ECG is the electrical signal of the heart), temperature, heart rate, and human movement. The edge device uses a 32-bit Tensilica microcontroller-based module with the ability to transmit data wirelessly using Wi-Fi. In addition, algorithms for classification and detection of movement patterns were implemented to be implemented in devices with limited resources and not only in high-performance computers. The Internet of Things and its application in smart environments can help non-intrusive monitoring of daily activities by implementing support vector machine (SVM is a machine learning algorithm) for implementation in embedded systems with low hardware resources. This paper shows experimental results obtained during the acquisition, transmission, and processing of physiological signals in a edge computing system and their visualization in a web application.

A Novel Criticality Analysis Technique for Detecting Dynamic Disturbances in Human Gait

The application of machine learning (ML) has made an unprecedented change in the field of medicine, showing a significant potential to automate tasks and to achieve objectives that are closer to human cognitive capabilities. Human gait, in particular, is a series of continuous metabolic interactions specific for humans. The need for an intelligent recognition of dynamic changes of gait enables physicians in clinical practice to early identify impaired gait and to reach proper decision making. Because of the underlying complexity of the biological system, it can be difficult to create an accurate detection and analysis of imbalanced gait. This paper proposes a novel Criticality Analysis (CA) methodology as a feasible method to extract the dynamic interactions involved in human gait. This allows a useful scale-free representation of multivariate dynamic data in a nonlinear representation space. To quantify the effectiveness of the CA methodology, a Support Vector Machine (SVM) algorithm is implemented in order to identify the nonlinear relationships and high-order interactions between multiple gait data variables. The gait features extracted from the CA method were used for training and testing the SVM algorithm. The simulation results of this paper show that the implemented SVM model with the support of the CA method increases the accuracy and enhances the efficiency of gait analysis to extremely high levels. Therefore, it can perform as a robust classification tool for detection of dynamic disturbances of biological data patterns and creates a tremendous opportunity for clinical diagnosis and rehabilitation.

A novel event-related fMRI supervoxels-based representation and its application to schizophrenia diagnosis

Background and Objective: The schizophrenia diagnosis represents a difficult task because of the confusing descriptions of symptoms given by the patient, their similarity among several disorders, the lower familiarity with genetic predisposition, and the probably inadequate response to the treatment. Neuro-biological markers of schizophrenia, as a quantitative relationship between the psychiatrist’s reports and the biology of the brain, could be used. Functional Magnetic Resonance Imaging (fMRI) obtain the subject’s performance in cognitive tasks and may find significant differences between the patient’s data and controls. The input data of classifiers may imply alterations in diagnosis; therefore, it is essential to ensure an adequate representation to describe the entire dataset classified.Methods: We propose a supervoxels-based representation calculated by two main steps: the short-range connectivity, supervoxels’ generation using a Fuzzy Iterative Clustering algorithm, and the long-range connectivity, employing Detrended Cross-Correlation Analysis among supervoxels. The unrelated supervoxels, through a statistical test based on critical points calculated empirically, are removed. The remainder supervoxels are the input for feature selectors to extract the discriminative supervoxels. We implement support vector machine classifiers using the correlation coefficient of the significant supervoxels. The dataset of 1.5 Tesla was downloaded from the SchizConnect site, where the fMRI data, during an auditory oddball task, was acquired. We calculate the performance of the classifiers using a leave-one-out cross-validation and compute the area under the Receiver Operating Characteristic curve and a permutation test to ensure no bias in the classifiers.Results: According to the permutation test, with p-values less than the significance level of 0.05, the classifiers extract discriminative class structure from data where no bias is shown. Our supervoxels-based representation gets the maximum values of sensitivity, specificity, and accuracy of 92.9%, 100%, and 96.4%, respectively. The discriminative brain regions, to discern among patients and controls, are extracted; these regions also are mentioned by the related works.Conclusions: The proposed representation, based on supervoxels, is a data-driven model that does not use predefined models of the signal nor pre-relocated brain regions of interest. The results are competitive against the related works, and the relevant supervoxels are related to the schizophrenia diagnosis.

High-dynamic-range 3D measurement for E-beam fusion additive manufacturing based on SVM intelligent fringe projection system

Electron Beam Metal Additive Manufacturing (EBAM) has been developed over recent years because of its advantages in manufacturing internal features and complex structures with relatively high productivity. The process proceeds by layer by layer melting and re-solidification of metal powder utilising an electron beam energy source. Following solidification of the build layer, the surface becomes ‘shiny’ with high reflectivity which makes in-process inspection of the surface of the build layer using fringe projection, difficult. To address this issue, a novel intelligent fringe projection technique using a support-vector-machine (SVM) algorithm is proposed to measure the 3D topography of high dynamic range surfaces on a layer by layer basis within the EBAM machine. To facilitate the SVM implementation a range of EBAM manufactured surfaces are utilised as samples for training and classification. The training measurements are based on different exposure times and saturated pixels are utilised as feature vectors to predict tested samples categories. Training errors are evaluated, and the correct recognition rate is 91%, which indicates the proposed training method can effectively predict the categories of the surfaces. Examples of melting edge swelling and powder bed inspections during a part build are used to demonstrate the system capability for inspection high dynamic range measurement within the EBAM machine. The whole inspection process lasts less than 5 s with two measurements which minimises the time penalty for the manufacturing process. Experimental results showed that the powder and the melting surface defects could be efficiently inspected using the proposed technology and the measurement result could be fed back to the build process to improve the processing quality.

Automated Linking PUBMED Documents with GO Terms Using SVM

We have developed an automated linking scheme for PUBMED citations with GO terms using SVM (Support Vector Machine), a classifica tion algorithm. The PUBMED database has been essential to life science re searchers with over 12 million citations. More recently GO (Gene Ontology) has provided a graph structure for biological process, cellular component, and molecular function of genomic data. By text mining the textual content of PUBMED and associating them with GO terms, we have built up an ontological map for these databases so that users can search PUBMED via GO terms and conversely GO entries via PUBMED classification. Conse quently, some interesting and unexpected knowledge may be captured from them for further data analysis and biological experimentation. This paper reports our results on SVM implementation and the need to parallelize for the training phase.

Monitoring Urban Expansion Using Remote-Sensing Data Aided by Google Earth Engine

Urbanizationis accompanied by rapid social and economic development, while the process of urbanization causes the degradation of the natural ecology. Direct loss in vegetation biomass from areas with a high probability of urban expansion can contribute to the total emissions from tropical deforestation and land-use change. Monitoring of urban expansion is essential for more efficient urban planning, protecting the ecosystem and the environment. In this paper, we use remote sensing data aided by Google Earth Engine (GEE) to evaluate the urban expansion of the city of Isfahan in the last thirty years. Thus, in this paper we use Landsat satellite images from 1986 and 2019, integrated into GEE, implementing Support vector machine (SVM) classification method. The accuracy assessment for the classified images showed high accuracy (95-96%), while the results showed a significant increase in the urban area of the city of Isfahan, occupying more than 70% of the study area. For future studies, we recommend a more detailed investigation about the city expansion and the negative impacts that may occur due to urban expansion.

Sign Language Recognition Using Multiple Kernel Learning: A Case Study of Pakistan Sign Language

All over the world, deaf people use sign language as the only reliable source of communication with each other as well as with normal people. These communicating signs are made up of the shape of the hand and movement. In Pakistan, deaf people use Pakistan sign language (PSL) as a means of communication with people. In scientific literature, many studies have been done on PSL recognition and classification. Most of these work focused on colored-based hands while some others are sensors and Kinect-based approaches. These techniques are costly and also avoid user-friendliness. In this paper, a technique is proposed for the recognition of thirty-six static alphabets of PSL using bare hands. The dataset is obtained from the sign language videos. At a later step, four vision-based features are extracted i.e. local binary patterns, a histogram of oriented gradients, edge-oriented histogram, and speeded up robust features. The extracted features are individually classified using Multiple kernel learning (MKL) in support vector machine (SVM). We employed a one-to-all approach for the implementation of basic binary SVM into the multi-class SVM. A voting scheme is adopted for the final recognition of PSL. The performance of the proposed technique is measured in terms of accuracy, precision, recall, and F-score. The simulation results are promising as compared with existing approaches.