Linear Kernel Research Articles

Klasifikasi Penyakit Tanaman Bawang Merah Menggunakan Metode SVM dan CNN

Shallots are one of the most widely produced crops in Enrekang Regency. The obstacle in cultivation is the presence of disease in the plant which can reduce production yields. We can recognize this disease from the spots on the leaves because these spots have unique color and texture characteristics. The aim of this research is to determine the results of the classification of shallot plant diseases which focuses on purple spot and moler disease. The classification algorithms used are CNN and SVM with RBF, linear, sigmoid and polynomial kernels. The feature extraction method used is Gray Level Co-occurance Matrix (GLCM). The analysis was carried out using 320 datasets with 2 classes, namely, purple spot disease and moler disease, each class has 160 datasets. The test results show that the CNN and SVM methods with RBF, linear and polynomial kernels get accuracy, precision, recall and F1 scores of 100% respectively. Meanwhile, the SVM method on the sigmoid kernel using texture feature extraction with the GLCM method states that the accuracy value is 75%, precision 75%, recall 73% and F1-Score 74%. So these results state that the Sigmoid method using GLCM feature extraction has the lowest value among other methods

Stacking: A novel data-driven ensemble machine learning strategy for prediction and mapping of Pb-Zn prospectivity in Varcheh district, west Iran

Various ensemble machine learning techniques have been widely studied and implemented to construct the predictive models in different sciences, including bagging, boosting, and stacking. However, bagging and boosting concentrate on minimizing variance or bias, stacking techniques aimed at reducing both by identifying the optimal integration of base learners. Moreover, while most ensemble methods simply combine identical machine learning models, stacking utilizes a meta-machine learning model to combine different base learning models, aiming to enhance the overall accuracy of generalization. Therefore, this research showed the utilization of stacking, an ensemble approach, to develop mineral prospectivity models for Pb-Zn mineralization in the Varcheh District, west Iran. To end this, various exploration evidence layers, including geochemical data, remote sensing data, geological and tectonic controls were used to construct the stacking structure. In the following, a set of five base learners were applied, containing support vector regression (SVR) using RBF, linear and polynomial kernels, the K-nearest neighbor (KNN), and linear regression. Ridge, SVR-RBF and XGBoost were used as a meta-learner to integrate the outputs of basic learners. To measure how well each model performed, ROC, F1-score and Precision metrics was carried out. Moreover, compared to the separate algorithms, the stacking-based ensemble model showed a better prediction accuracy. The findings of this study demonstrated that the ensemble model based on stacking achieved a 95% prediction rate for Pb-Zn deposits, covering only 9% of the study area. As a result, this model holds promise as an effective tool for predicting mineral prospectivity in other study areas, regardless of whether they exhibit similar or different types of mineralization.

Comparing Machine Learning Classifiers for Predicting Hospital Readmission of Heart Failure Patients in Rwanda.

High rates of hospital readmission and the cost of treating heart failure (HF) are significant public health issues globally and in Rwanda. Using machine learning (ML) to predict which patients are at high risk for HF hospital readmission 20 days after their discharge has the potential to improve HF management by enabling early interventions and individualized treatment approaches. In this paper, we compared six different ML models for this task, including multi-layer perceptron (MLP), K-nearest neighbors (KNN), logistic regression (LR), decision trees (DT), random forests (RF), and support vector machines (SVM) with both linear and radial basis kernels. The outputs of the classifiers are compared using performance metrics including the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity. We found that RF outperforms all the remaining models with an AUC of 94% while SVM, MLP, and KNN all yield 88% AUC. In contrast, DT performs poorly, with an AUC value of 57%. Hence, hospitals in Rwanda can benefit from using the RF classifier to determine which HF patients are at high risk of hospital readmission.

The statistical complexity of early-stopped mirror descent

Abstract Recently there has been a surge of interest in understanding implicit regularization properties of iterative gradient-based optimization algorithms. In this paper, we study the statistical guarantees on the excess risk achieved by early-stopped unconstrained mirror descent algorithms applied to the unregularized empirical risk. We consider the set-up of learning linear models and kernel methods for strongly convex and Lipschitz loss functions while imposing only boundedness conditions on the unknown data-generating mechanism. By completing an inequality that characterizes convexity for the squared loss, we identify an intrinsic link between offset Rademacher complexities and potential-based convergence analysis of mirror descent methods. Our observation immediately yields excess risk guarantees for the path traced by the iterates of mirror descent in terms of offset complexities of certain function classes depending only on the choice of the mirror map, initialization point, step size and the number of iterations. We apply our theory to recover, in a clean and elegant manner via rather short proofs, some of the recent results in the implicit regularization literature while also showing how to improve upon them in some settings.

Study on the Relationship between the Spatial Distribution of Shared Bicycle Travel Demand and Urban Built Environment

As a green and sustainable trip mode, shared bicycles play an essential role in completing short-distance trips in cities. This paper proposes a method to analyze the impact of the urban built environment on the distribution of shared bicycles in a small-scale space. First, the Fishnet function of ArcGIS is utilized to divide the study area into grids of 500 m × 500 m. Then, three indicators are proposed to describe the characteristics of the urban built environment, including point of information (POI) comprehensive index, the intensity of public transportation coverage, spatial accessibility, providing them the ways to be assigned to the grids. Finally, the multivariable linear regression model and support vector regression (SVR) models are applied to reveal the impacts of built environment factors on the spatial distribution of shared bicycles. Results show that SVR models based on linear kernel function, Gaussian radial basis kernel function, and Polynomial kernel function can achieve better analysis results. The SVR model based on the Gaussian radial basis function shows higher explanatory power (adjusted R2 = 0.978) than the multivariable linear regression model (adjusted R2 = 0.847). This paper can aid in understanding the demand and supply of shared bicycles and help operators or governments to improve service quality.

A Hybrid ROS-SVM Model for Detecting Target Multiple Drug Types

Misleading in determining the decision to use the target drug will be fatal, even to death. This study examines five pharmacological targets designated as types A, B, C, X, and Y. Early detection of misleading drug targeting will reduce the risk of death. This study aims to develop hybrid random oversampling techniques (ROS) and support vector machine (SVM) methods. The use of the oversampling technique in this study aims to balance classes in the dataset; due to the data collection in each class, there is a relatively large gap. This study applies five schemes to see which combination of models produces the highest accuracy. This study also uses five types of SVM kernels, linear, polynomial, gaussian, RBF, and sigmoid, combined with the ROS oversampling technique. Our proposed model combines the ROS oversampling technique with a linear SVM kernel. We evaluated the proposed model and resulted in an accuracy of 97% and compared it with several experiments, including the ROS technique with a sigmoid kernel which only resulted in 50% accuracy. It can be seen from the results obtained that the linear kernel is very adaptive to data types in the form of numeric and nominal compared to other kernels. The method proposed in this study can be applied to other medical problems. Future research can be carried out using a combination of other sampling techniques with deep learning-based methods on this issue.

Economic Impact due Covid-19 Pandemic: Sentiment Analysis on Twitter Using Naïve Bayes Classifier and Support Vector Machine

Covid-19 is an outbreak caused by severe acute respiratory syndrome. Covid-19 first appeared in Indonesia on March 2, 2020, with two confirmed cases and increased to 1285 cases in 30 provinces. One of the impacts of the Covid-19 pandemic is on the economic aspect, which has experienced a drastic decline in income. This study aims to classify public opinion to determine the level of public sentiment on the economic impact of the Covid-19 pandemic and to identify parameters that influence the accuracy of the sentiment analysis classification model. The methods used in this current research are Lexicon, Support Vector Machine (SVM), and Naive Bayes Classifier (NBC). First, Lexicon is used for scoring and labeling the preprocessed data. Second, SVM is used to classify the sentiment, then find the best accuracy using linear, radial, polynomial, and sigmoid kernels. Third, NBC is used to classify sentiment as a comparison method. The results indicated that 255 tweet data consisted of 44 positive tweets (17.25%), 46 neutral tweets (18.04%), and 165 negative tweets (64.71%). Therefore, it can be inferred that the economic impact on the Indonesian people due to the Covid-19 pandemic has a high negative sentiment value. In the performance, SVM yielded a better accuracy of 100%, precision, recall, and F-measure are 1. This study proves that selecting the kernel type and applying underfitting can improve the accuracy of SVM. Also, SVM can perform well on a small amount of training data.

Machine learning‐based regional flood frequency analysis of Indian watersheds

AbstractEstimates of the flood quantile for ungauged watersheds are crucial for water resources management but challenging due to the nonlinear complex hydrological system. For ungauged watersheds, estimating flood quantiles relies on various interdependent physiometeorological variables, many of which are not adequately considered in regional flood frequency analysis (RFFA). In this study, we utilized the random forest (RF) and support vector regression (SVR) algorithms, which can learn the nonlinear relationship between the physiometeorological variables and flood quantiles for RFFA. Thirteen physiometeorological variables that were not collectively employed before were used to estimate the 10‐year, 50‐year, and 100‐year return period flood quantiles (Q10, Q50, and Q100), respectively, for 39 watersheds spread across India. The RF and SVR models were trained on 29 (75%) watersheds to estimate individual flood quantiles and were subsequently tested on the remaining ten (25%) ungauged watersheds. The R2 achieved by RF is 0.862, 0.813, and 0.845, and SVR is 0.807, 0.793, and 0.789 for Q10, Q50, and Q100, respectively. Overall, the results indicate that RF can effectively learn the nonlinear relationships, while SVR with a linear kernel requires further improvement to estimate reliable flood quantiles. The study demonstrates that machine learning algorithms, with appropriate physiometeorological input datasets, can be used to estimate flood quantiles even in the sparse data region.

Protein classification by autofluorescence spectral shape analysis using machine learning

Depending on the relative numbers and spatial arrangement of Tryptophan (Trp; W) and Tyrosine (Tyr; Y) residues, different proteins produce distinct autofluorescence (AF) spectral shapes when excited at ∼280 nm. Yet, considering the vast number and heterogeneous forms in nature, visual analysis and precise identification of proteins based on their AF spectra is challenging and further compounded in cases when different proteins produce substantially similar AF spectral shapes. There is, thus, a serious need to develop a methodology to address this problem. The current study proposes a practical technology to quickly identify proteins using machine learning (ML) algorithms based on their AF spectra. Specifically, AF spectra of fifteen different standard proteins of varying origin with distinct structural and Trp/Tyr compositions were recorded; based on the spectral features selected by the Minimum-Redundancy-Maximum-Relevance (mRMR) algorithm, a multiclass Support Vector Machine (SVM) learning model with Radial Basis Function (RBF), Polynomial, and Linear kernels classified the proteins with high accuracy of 99.06%, 99.03%, and 98.29% respectively. Since protein identification is the key to understand biological functions and disease diagnosis, the proposed methodology could offer a viable alternative to and improve the existing protein identification techniques.

Multivariate Genomic Hybrid Prediction with Kernels and Parental Information.

Genomic selection (GS) plays a pivotal role in hybrid prediction. It can enhance the selection of parental lines, accurately predict hybrid performance, and harness hybrid vigor. Likewise, it can optimize breeding strategies by reducing field trial requirements, expediting hybrid development, facilitating targeted trait improvement, and enhancing adaptability to diverse environments. Leveraging genomic information empowers breeders to make informed decisions and significantly improve the efficiency and success rate of hybrid breeding programs. In order to improve the genomic ability performance, we explored the incorporation of parental phenotypic information as covariates under a multi-trait framework. Approach 1, referred to as Pmean, directly utilized parental phenotypic information without any preprocessing. While approach 2, denoted as BV, replaced the direct use of phenotypic values of both parents with their respective breeding values. While an improvement in prediction performance was observed in both approaches, with a minimum 4.24% reduction in the normalized root mean square error (NRMSE), the direct incorporation of parental phenotypic information in the Pmean approach slightly outperformed the BV approach. We also compared these two approaches using linear and nonlinear kernels, but no relevant gain was observed. Finally, our results increase empirical evidence confirming that the integration of parental phenotypic information helps increase the prediction performance of hybrids.

Prediction of drowsiness using EEG signals in young Indonesian drivers

Indonesia is among the countries with the highest accident rates in the world. Fatigue and drowsiness are among the main causes of the increased risks of accidents in the road transport sector. Sleep-related factors (quality and quantity, time of day) and work-related factors significantly affect the development of fatigue. The EEG signal indicator is often referred to as the gold standard for measuring fatigue and drowsiness. However, previous studies focused primarily on the trends of EEG signals under certain conditions but overlooking the development of drowsiness indicators based on EEG signals. Furthermore, existing studies still do not agree on what parameters in the EEG signal indicator are best at detecting drowsiness. Thus, this study aims to design an EEG signal-based drowsiness indicator under simulated driving conditions. Drowsy drivers were monitored through EEG signal indicators and subjective assessments. The methods used in this study include statistical significance tests, logistic regression, and support vector machine. The results showed that sleep deprivation had a significant effect on increasing alpha, beta, and theta waves. In addition, driving duration significantly increased the theta power and all EEG ratios and decreased the beta power in the alert group. The ratio of (θ + α)/β and θ/β in the SD group also showed a considerable increase in the end of driving. Furthermore, sleep status and driving duration both influenced subjective sleepiness. EEG signals combined with sleep status and driving duration factors generated acceptable model accuracies (77.1% and 90.2% in training and testing, respectively), with 90.5% sensitivity and 90% specificity in data test. Support vector machine showed better classification than that of logistics regression, with the linear kernel as the best classifier. Theta power had the highest effect in the model compared with other EEG signals.

Hardware Realization of an Innovative Disturbance Detection Algorithm for Control Strategy of Solid-State Transfer Switch

An accurate and fast Solid-State Transfer Switch (SSTS) has the potential to automate conventional utility network. Presently, It consumes most of the transfer time in detecting the condition for initiating the transfer process. Recent developments in the power semiconductor devices technology have reduced the switching time to the tune of nanoseconds. Hence, time consumed by the disturbance detection is significant in SSTS. IEEE Std. 1100-2005 and IEEE Std. 446-1995 requires fast load transfer for SSTS. Realization of an innovative machine learning technique is implemented on three AVR microcontrollers for each of three phases. The microcontroller executes segmentation of acquired signal and calculates derivative of each segment. From each pre-processed 10-sample segment, mean absolute deviation (MAD) and energy (E) features are extracted. On the basis of feature values, SVM classifier with the linear kernel detects disturbance. Logical OR gate generates transfer enable signal to initiate the transfer process. Results of case studies appreciate the ability of proposed hardware to detect common disturbances in power system under different operating conditions. The hardware provides a feasible solution for solid state transfer switch with required accuracy and speed.

A Novel Event-Driven Spiking Convolutional Neural Network for Electromyography Pattern Recognition.

Electromyography (EMG) pattern recognition is an important technology for prosthesis control and human-computer interaction etc. However, the practical application of EMG pattern recognition is hampered by poor accuracy and robustness due to electrode shift caused by repeated wearing of the signal acquisition device. Moreover, the user's acceptability is low due to the heavy training burden, which is caused by the need for a large amount of training data by traditional methods. In order to explore the advantage of spiking neural network (SNN) in solving the poor robustness and heavy training burden problems in EMG pattern recognition, a spiking convolutional neural network (SCNN) composed of cyclic convolutional neural network (CNN) and fully connected modules is proposed and implemented in this study. High density surface electromyography (HD-sEMG) signals collected from 6 gestures of 10 subjects at 6 electrode positions are taken as the research object. Compared to CNN with the same structure, CNN-Long Short Term Memory (CNN-LSTM), linear kernel linear discriminant analysis classifier (LDA) and spiking multilayer perceptron (Spiking MLP), the accuracy of SCNN is 50.69%, 33.92%, 32.94% and 9.41% higher in the small sample training experiment, 6.50%, 4.23%, 28.73%, and 2.57% higher in the electrode shifts experiment respectively. In addition, the power consumption of SCNN is about 1/93 of CNN. The advantages of the proposed framework in alleviating user training burden, mitigating the adverse effect of electrode shifts and reducing power consumption make it very meaningful for promoting the development of user-friendly real-time myoelectric control system.

Ebct: Using entropy balancing for continuous treatments to estimate dose–response functions and their derivatives

Interest in evaluating dose–response functions of continuous treatments has been increasing recently. To facilitate the estimation of causal effects in this setting, I introduce the ebct command for the estimation of dose–response functions and their derivatives using entropy balancing for continuous treatments. First, balancing weights are estimated by numerically solving a globally convex optimization problem. These weights eradicate Pearson correlations between covariates and the treatment variable. Because simple uncorrelatedness may be insufficient to yield consistent estimates in the next step, higher moments of the treatment variable can be rendered uncorrelated with covariates. Second, the weights are used in local linear kernel regressions to estimate the dose–response function or its derivative. To perform statistical inference, I use a bootstrap procedure. The command also provides the option of producing publication-quality graphs for the estimated relationships.

AN EMD AND IMF ENERGY ENTROPY-BASED OPTIMIZED FEATURE EXTRACTION AND CLASSIFICATION SCHEME FOR SINGLE TRIAL EEG SIGNAL

For the purpose of improving the classification accuracy of single-trial electroencephalogram (EEG) signal during motor imagery (MI) process, this study proposed a classification method which combines intrinsic mode functions (IMFs) energy entropy and improved empirical mode decomposition (EMD) scheme. Singular value decomposition (SVD), Gaussian mixture model (GMM), EMD and IMF energy entropy were employed for the newly designed scheme. After removing noise and artifacts from acquired EEG signals in EEGLAB, SVD was applied, and the singular values were clustered by GMM. The insignificant characteristics indicated by the small SVD values were then removed, and the signals were reconstructed, feeding to EMD algorithm. Those IMFs mapping to [Formula: see text], [Formula: see text], [Formula: see text] and [Formula: see text] frequencies were selected as the major features of the Electroencephalogram signal. The SVM classifier with Radial Basis Function Neural Network, linear, and polynomial kernel functions and voting mechanism then kicked in for classification. The results were compared with that of the traditional EMD and EEMD through simulation, showing that the proposed scheme can eliminate mode mixing effectively and improve the single-trial EEG signal classification accuracy significantly, suggesting the probability of designing a more efficient EEG control system based on the proposed scheme.

Deep-learning and conventional radiomics to predict IDH genotyping status based on magnetic resonance imaging data in adult diffuse glioma.

In adult diffuse glioma, preoperative detection of isocitrate dehydrogenase (IDH) status helps clinicians develop surgical strategies and evaluate patient prognosis. Here, we aim to identify an optimal machine-learning model for prediction of IDH genotyping by combining deep-learning (DL) signatures and conventional radiomics (CR) features as model predictors. In this study, a total of 486 patients with adult diffuse gliomas were retrospectively collected from our medical center (n=268) and the public database (TCGA, n=218). All included patients were randomly divided into the training and validation sets by using nested 10-fold cross-validation. A total of 6,736 CR features were extracted from four MRI modalities in each patient, namely T1WI, T1CE, T2WI, and FLAIR. The LASSO algorithm was performed for CR feature selection. In each MRI modality, we applied a CNN+LSTM-based neural network to extract DL features and integrate these features into a DL signature after the fully connected layer with sigmoid activation. Eight classic machine-learning models were analyzed and compared in terms of their prediction performance and stability in IDH genotyping by combining the LASSO-selected CR features and integrated DL signatures as model predictors. In the validation sets, the prediction performance was evaluated by using accuracy and the area under the curve (AUC) of the receiver operating characteristics, while the model stability was analyzed by using the relative standard deviation of the AUC (RSDAUC). Subgroup analyses of DL signatures and CR features were also individually conducted to explore their independent prediction values. Logistic regression (LR) achieved favorable prediction performance (AUC: 0.920 ± 0.043, accuracy: 0.843 ± 0.044), whereas support vector machine with the linear kernel (l-SVM) displayed low prediction performance (AUC: 0.812 ± 0.052, accuracy: 0.821 ± 0.050). With regard to stability, LR also showed high robustness against data perturbation (RSDAUC: 4.7%). Subgroup analyses showed that DL signatures outperformed CR features (DL, AUC: 0.915 ± 0.054, accuracy: 0.835 ± 0.061, RSDAUC: 5.9%; CR, AUC: 0.830 ± 0.066, accuracy: 0.771 ± 0.051, RSDAUC: 8.0%), while DL and DL+CR achieved similar prediction results. In IDH genotyping, LR is a promising machine-learning classification model. Compared with CR features, DL signatures exhibit markedly superior prediction values and discriminative capability.

Support Vector Machine Application for Classification of Tempe Fermentation Maturity with Information Gain Selection Feature

Tempe is one of the ingredients of traditional Indonesian cuisine. In making tempe, a soybean fermentation process is needed which is generally still carried out in an open environment so that the maturity time becomes slow and erratic. Therefore, in the tempe fermentation process, a detector is needed to find out optimal maturity in tempe. This detection effort makes it possible to use image processing by utilizing various feature extractions through the classification process. This research utilizes a variety of image features, namely texture features using the GLCM method and various color features, namely RGB, HSV, LAB, CMYK, YUV, HIS, HCL, LCH. However, with so many features, it causes a high computational load, so that in this study the Information Gain approach was used to select features. Furthermore, the classification process is carried out using the Support Vector Machine (SVM) method with variations of linear, polynomial, gaussian and sigmoid kernels. Tempe objects in the fermentation process are divided into unripe, ripe and rotten classes with a total of 410 images as a dataset. The test results (SVM+IG) on the Sigmoid kernel produce the fastest time accuracy with a computational result of 2.18 seconds on a 30:70 split ratio, the longest split ratio is 80:20 which is 2.50 seconds on a Linear kernel and produces the highest accuracy of 96 ,74%. Furthermore, in the SVM test without using Information Gain on the gaussian kernel, it produced the fastest time accuracy of 2.28 seconds, and the longest at a split ratio of 40:60, namely 3.00 seconds in the polynomial kernel. Thus the result of using SVM+IG is that the average level of accuracy when using (SVM+IG) is faster than the SVM process without IG which obtains slower computation time. Based on the description above, this study aims to apply the SVM method to classify tempe fermented images with feature selection using Information Gain.

Fracture toughness evaluation of ground granulated blast furnace slag concrete using experimental study and machine learning techniques

The relationship between mix design and fracture properties of ground granulated blast furnace slag (GGBFS) concrete is complex and influenced by many factors, such as mixture composition. This study evaluates the fracture toughness under mode-I, mode-II, and mixed-mode, performed on centrally straight-notched Brazilian discs. An experimental program generated an extensive database of the fracture toughness of GGBFS mortar, consisting of 143 specimens with different mix designs under different loading angle conditions. Then, a combination of support vector regression (SVR) algorithm and genetic algorithm (GA) as hyperparameter tuner was conducted to develop models for fracture toughness. The SVR algorithm was trained with Radial Basis Function (RBF), polynomial, sigmoid, and linear kernel functions. Based on the Taylor diagram, the models demonstrated robust performance in predicting fracture toughness. The performance metrics indicate that the RBF function is the most reliable model, with R2-score 0.95 and 0.94 for KIC and KIIC, respectively. The importance score of input features was determined with the permutation importance function. The angle between the crack and loading direction is the most influential parameter in the models.

Comparative Study of Various Classification and Regression Techniques on Movie Datasets Using R

The entertainment industry has been exponentially growing since its inception and is primarily well-received. Creating content, in this case, Movies has been the bread and butter of the industry and has proven extremely lucrative over the years. In this comparative study, we have picked out various Machine Learning models, viz. Classification Tree, Regression Tree, AdaBoost, SVM (Linear Kernel, Polynomial Kernel, Radial Kernel), XGBoost, Random Forest, Simple Bagging, and Naive Bayes. The Classification models are used to predict the rate of a certain movie winning the Oscar whereas the Regression model implementation estimates the average earning of a film considering the various attributes.

Identification of gene-level methylation for disease prediction.

DNA methylation is an epigenetic alteration that plays a fundamental part in governing gene regulatory processes. The DNA methylation mechanism affixes methyl groups to distinct cytosine residues, influencing chromatin architectures. Multiple studies have demonstrated that DNA methylation's regulatory effect on genes is linked to the beginning and progression of several disorders. Researchers have recently uncovered thousands of phenotype-related methylation sites through the epigenome-wide association study (EWAS). However, combining the methylation levels of several sites within a gene and determining the gene-level DNA methylation remains challenging. In this study, we proposed the supervised UMAP Assisted Gene-level Methylation method (sUAGM) for disease prediction based on supervised UMAP (Uniform Manifold Approximation and Projection), a manifold learning-based method for reducing dimensionality. The methylation values at the gene level generated using the proposed method are evaluated by employing various feature selection and classification algorithms on three distinct DNA methylation datasets derived from blood samples. The performance has been assessed employing classification accuracy, F-1 score, Mathews Correlation Coefficient (MCC), Kappa, Classification Success Index (CSI) and Jaccard Index. The Support Vector Machine with the linear kernel (SVML) classifier with Recursive Feature Elimination (RFE) performs best across all three datasets. From comparative analysis, our method outperformed existing gene-level and site-level approaches by achieving 100% accuracy and F1-score with fewer genes. The functional analysis of the top 28 genes selected from the Parkinson's disease dataset revealed a significant association with the disease.