Analysis of Machine Learning Classifiers for Speaker Identification: A Study on SVM, Random Forest, KNN, and Decision Tree

Background and purposeRenal cell carcinoma (RCC) is a heterogeneous group of cancers. The collagen fiber content in the tumor microenvironment of renal cancer has an important role in tumor progression and prognosis. A radiomics model was developed from dual-energy CT iodine maps to assess collagen fiber content in the tumor microenvironment of ccRCC.MethodsA total of 87 patients with ccRCC admitted to our hospital were included in this retrospective study. Among them, 59 cases contained large amounts of collagen fibers and 28 cases contained a small amount of collagen fibers. We established a radiomics model using preoperative dual-energy CT scan Iodine map (IV) imaging to distinguish patients with multiple collagen fibers from those with few collagen fibers in the tumor microenvironment of ccRCC. We extracted features from dual-energy CT Iodine map images to evaluate the effects of six classifiers, namely k-nearest neighbor (KNN), support vector machine (SVM), extreme gradient boosting (XGBoost), random forest (RF), logistic regression (LR), and decision tree (DT). The effects of the models built based on the dynamic and venous phases are also compared. Model performance was evaluated using quintuple cross-validation and area under the receiver operating characteristic curve (AUC). In addition, a clinical model was developed to assess the clinical factors affecting collagen fiber content.ResultsCompared to KNN, SVM, and LR classifiers, RF, DT, and XGBoost classifiers trained with higher AUC values, with training sets of 0.997, 1.0, and 1.0, respectively. In the validation set, the highest AUC was found in the SVM classifier with a size of 0.722. In the comparative test of the active and intravenous phase models, the SVM classifier had the best effect with its validation set AUC of 0.698 and 0.741. In addition, there was a statistically significant effect of patient age and maximum tumor diameter on the collagen fiber content in the tumor microenvironment of kidney cancer.ConclusionRadionics features based on preoperative dual-energy CT IV can be used to predict the amount of collagen fibers in the tumor microenvironment of renal cancer. This study better informs clinical prognosis and patient management. Iodograms may add additional value to dual-energy CTs.

In recent years, there has been a noticeable trend toward targeted threats to information security, where companies are now leveraging vulnerabilities and risks associated with widely used services in order to generate financial gain. Additionally, they implement numerous precautions and consistently carry out their tasks. One item that requires precautionary measures is the network devices utilized. Network devices in computer networks possess the capability to log events. These logs enable the identification of security events on the network and facilitate the implementation of precautionary measures. Various security measures can be implemented to handle such data. One of these measures is Security Information and Event Management (SIEM). It is a system that gathers and analyzes data from networks and security devices. SIEM is a technique employed to consolidate critical information within a cohesive structure. It allows for the correlation of events from different security devices, thereby improving the monitoring capabilities of cybersecurity operations centers. This study extensively covers the critical infrastructure-SIEM relationship, current studies, critical infrastructure, cyber security policies, and SIEM. Our system design was developed using the UNSW\_NB15 dataset, a widely recognized dataset in cybersecurity due to its comprehensive and realistic representation of cyber threats. This dataset consists of data obtained from network traffic, various attack activities, and real-life modern normal scenarios, making it particularly relevant to our study. With the studies, a total of 10 different categories were analyzed, with the category consisting of nine types of attacks, namely Analysis, Backdoor, DoS, Exploits, Fuzzers, Generic, Reconnaissance, Shellcode, and Worms and Normal activities. The study is divided into two as the basic structure. The first step was carried out on Google Collaboratory, and then some experimental studies were carried out in Weka. Classifications were made using several methods, including Logistic Regression (LR), Extra Trees (XT), Support Vector Machines (SVM), Random Forest (RF), and Decision Trees (DT). These methods were chosen for their proven effectiveness in similar studies. In the application developed with Google Colabratory, we achieved 98.62\% in Random Forest, 99.10\% in Decision Trees, 98.87\% in Logistic Regression, 95.13\% success in Extra Trees and 99.12\% success in Support Vector Machines. As a result of the studies and experiments carried out in Weka, we achieved 92.05\% in Random Forest, 100\% in Decision Trees, 100\% in k-Nearest Neighbours, 100\% in J48, 99.19\% in Naive-Bayes and 99.35\% in BayesNet achievements.

Microorganisms are fundamental to ecosystems, human health, and various industrial processes. The rapid and accurate identification of microorganisms is crucial for diagnosing diseases, monitoring environmental changes, and advancing biotechnological applications. Traditional methods of microbial identification, including culturing and microscopy, are often labor-intensive and time-consuming, necessitating the development of computational techniques to automate and improve accuracy. This project presents a comprehensive approach to microbe prediction by leveraging advanced machine learning algorithms applied to a dataset containing genomic and morphological features of ten different microorganism species. The study investigates the performance of four classification algorithms — K-Nearest Neighbors (KNN), Naive Bayes, Random Forest, and Decision Trees — to classify microorganisms based on multiple quantitative features extracted from images and genetic data. The methodology involves extensive data preprocessing steps such as cleaning, normalization, polynomial feature expansion, and dimensionality reduction using Principal Component Analysis (PCA) to enhance model performance. Rigorous experimentation and comparative analysis reveal that ensemble methods, particularly Random Forest, provide superior classification accuracy, robustness, and generalizability across unseen data. In addition to model development, a user-friendly web interface built using Streamlit facilitates real-time microorganism prediction, making the system accessible to researchers and practitioners without deep technical expertise. The platform enables the input of feature values and instantly returns predicted microorganism classes along with relevant biological descriptions and preventive measures, thereby bridging the gap between computational predictions and practical microbiological insights. This project contributes to the emerging intersection of microbiology and machine learning by demonstrating an effective pipeline from raw data to deployable predictive tool. Its implications span environmental monitoring, medical diagnostics, and industrial microbiology, emphasizing the potential for machine learning to revolutionize microbial identification and understanding. Key Words – K-Nearest Neighbors (KNN), Naive Bayes, Random Forest, and Decision Trees, Stream lit user interface software.

Aim: The aim of the work is to evaluate the accuracy and precision in predicting heart disease using Support Vector Machine (SVM), Random forest (RF), Logistic Regression (LR), Decision Tree (DT) Classification algorithms. Materials and Methods: Classification algorithm is appealed on a heart dataset which consists of 180 records. A framework for heart disease prediction in the medical sector comparing Random forest, Logistic Regression, Decision Tree and SVM classifiers has been proposed and developed. The sample size was calculated as 55 in each group using G power 80%. Sample size was calculated using clincalc analysis, with alpha and beta values 0.05 and 0.5, 95% confidence, pretest power 80% and enrolment ratio 1. Results: The Novel Random Forest Algorithm (92.13%), Support Vector Machine (62.51%), Logistic Regression (84.89%), Decision Tree (86.25%) classifiers produce respectively. SVM, RF exists a statistically significant difference between the two groups (p=0.001,p=.004;p<0.05).LR, RF exists a statistically insignificant difference between the two groups (p=.103, P=.080;p>0.05) both with confidence interval 95%. Hence Random forest is better than SVM, RF, DT classifiers. Conclusion: The results show that the performance of RF is better when compared with SVM, LR and DT in terms of both precision and accuracy.

Spoken language identification is a field of research that is already being done by many people. There are many techniques proposed for doing speech processing, such as Support Vector Machines, Gaussian Mixture Models, Decision Trees, and others. This paper will use the system using the Mel-Frequency Cepstral Coefficient (MFCC) features of speech input signal, use Random Forest (RF), Gaussian Mixture Model (GMM), and K-Nearest Neighbor (KNN) as a classifier, use the 3s, 10s, and 30s as scoring method, and use dataset that consists of Javanese, Sundanese, and Minang languages which are traditional languages from Indonesia. K-Nearest Neighbor has 98.88% of accuracy for 30s of speech and followed by Random Forest that has 95.55% of accuracy for 30s of speech, GMM has 82.24% of accuracy.

Heart disease is a leading cause of death worldwide, and early detection is crucial for improving patient outcomes. Machine learning classifiers have shown promise in predicting heart disease using patient-specific factors such as demographic information, medical history, and lifestyle habits. In this paper, we aimed to evaluate the performance of machine learning classifiers in predicting heart disease using a combination of 5 different datasets such as Cleveland, Hungarian, Switzerland, Long Beach VA, and StatLog (Heart) Datasets available on IEEE data port. The two significant challenges are addressed in this work: 1) predicting heart failure using machine-learning models without eliminating any clinical features, which increases the risk of overfitting and can result in poor performance metrics, and 2) we propose a model that will provide remarkable accuracy regardless of the type of data, to offer model generalizability. Machine-learning algorithms such as logistic regression, decision trees, random forests, support vector machine, extreme gradient boosting, extra-tree, and K-nearest neighbor are applied for heart disease prediction. In addition, ensemble approaches such as majority voting, boosting, bagging, and stacking are employed. The performance of the classifiers is evaluated using performance metrics such as accuracy, precision, recall, and F1-score. The results showed that the ensemble approach of stacking outperformed individual models, with an accuracy of 93.67%.

Comparative studies of human and animal corneal structures using Raman spectroscopy are limited. We compared the Raman spectra of human and animal corneal stroma to evaluate the suitability of animal corneas as experimental models and to assess the feasibility of Raman spectroscopy for characterizing biochemical differences. Raman microspectroscopy (700–4000[Formula: see text]cm[Formula: see text]) was performed on seven fresh porcine corneas, six fresh rabbit corneas, and 12 human corneal stromal lenticules obtained from patients who underwent small incision lenticule extraction. Four machine learning algorithms — Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Random Forest (RF), and Decision Tree (DT) — were used to classify the data. Ten characteristic peaks were identified in the human corneal stroma, including prominent peaks at 937[Formula: see text]cm[Formula: see text] (proline/valine/type I collagen), 1243[Formula: see text]cm[Formula: see text] (collagen amide III), 1448[Formula: see text]cm[Formula: see text] (collagen/phospholipids), 1663[Formula: see text]cm[Formula: see text] (DNA/proteins), 2940[Formula: see text]cm[Formula: see text] (lipids/proteins), and 3330[Formula: see text]cm[Formula: see text] (water). Eleven peaks were observed in the porcine and rabbit corneas. Using raw spectra, the SVM, KNN, and RF models achieved 100% classification accuracy (area under the receiver operating characteristic (ROC) curve, 1.00 for both test and validation sets). The DT model showed the lowest performance (approximately 33% accuracy for human data; [area under the ROC curve, 0.67 [Formula: see text]]). Human myopic corneal stroma differed from the animal models in regions linked to nucleic acids and water content. The porcine cornea more closely resembled the human cornea than the rabbit cornea, indicating it as a more appropriate analog. Raman spectroscopy is a useful tool for interspecies comparisons and elucidating the biochemical composition of diseased corneas.

Research efforts in the prediction of stroke and diabetes prioritize early detection in order to enhance patient outcomes. To achieve this, a variety of methodologies are integrated. Existing studies, on the other hand, are marred by imbalanced datasets, lack of diversity in their datasets, potential bias, and inadequate model comparisons; these flaws underscore the necessity for more comprehensive and inclusive research methodologies. This paper provides a thorough assessment of machine learning algorithms in the context of early detection and diagnosis of stroke and diabetes. The research employed widely used algorithms, including Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and XGBoost Classifier, to examine medical data and derive significant findings. The XGBoost Classifier demonstrated superior performance, with an outstanding accuracy, precision, recall, and F1-score of 87.5%. The comparative examination of the algorithms indicated that the Decision Tree, Random Forest, and XGBoost classifiers consistently exhibited strong performance across all measures. The models demonstrated impressive discrimination capabilities, with the XGBoost Classifier and Random Forest reaching accuracy rates of roughly 87.5% and 86.5% respectively. The Decision Tree Classifier exhibited notable performance, with an accuracy rate of 83%. The overall accuracy of the models was evident in the F1-score, a metric that incorporates recall and precision, where the XGBoost model exhibited a marginal improvement of 2% over the Random Forest and Decision Tree models, and 4.25 percent over the last two. The aforementioned results underscore the effectiveness of the XGBoost Classifier, which will be employed as a predictive model in this study, alongside the Random Forest and Decision Tree models, for the accurate identification of stroke and diabetes. Furthermore, combining datasets improves model performance by utilizing relative features. This integrated dataset improves the model's efficiency and creates a resilient and comprehensive prediction model, improving healthcare outcomes. The findings of this research make a valuable contribution to the advancement of AI-driven diagnostic systems, hence enhancing the quality of healthcare decision-making.

The leading cause of death worldwide today is heart disease (HD). The heart is recognised as the second‐most significant organ behind the brain. A successful outcome of treatment can be improved by an early diagnosis which can significantly reduce the chance of death in health care. In this paper, we proposed a method to predict heart disease. We used various machine learning algorithms (MLA), namely, logistic regression (LR), k‐nearest neighbor (KNN), support vector machine (SVM), Naive Bayes (NB), random forest (RF), and decision tree (DT). With the testing data set, we evaluated the model’s accuracy in heart disease prediction. When compared to the other five models, the random forest and k‐nearest neighbor approaches perform better. With a 99.04% accuracy rate, the k‐nearest neighbor algorithm and random forest provide the best match to the data as compared to other algorithms. Six feature selection algorithms were used for the performance evaluation matrix. MCC parameters for accuracy, precision, recall, and F measure are used to evaluate models.

Parking space management has become a critical challenge in urban areas due to increasing vehicle numbers and limited parking infrastructure. This paper presents a comprehensive study of machine learning (ML) models in IoT‐enabled environments focusing on proposing an ML‐based model for predicting available parking space. The study evaluates the performance of various models including K‐nearest neighbors (KNNs), support vector machines (SVMs), random forest (RF), decision tree (DT), logistic regression (LR), and Naïve Bayes (NB) based on “precision, recall, accuracy, and F1‐score performance metrics”. The results obtained by implementing ML models on the data with 65% and 85% threshold values are compared to draw meaningful conclusions regarding their performance in predicting parking space availability. Among the evaluated models, random forest (RF) demonstrates superior performance with high precision, recall, accuracy, and F1‐score values. It showcases its effectiveness in accurately predicting parking space availability in the IoT‐enabled environment. On the other hand, models such as K‐nearest neighbors (KNNs), decision tree (DT), logistic regression (LR), and Naïve Bayes (NB) show relatively lower performance in complex parking scenarios. The paper concludes that the use of advanced predictive models, particularly random forest, significantly enhances the accuracy and reliability of IoT‐enabled parking management systems and also reduces the waiting time of the vehicles, leading to more efficient resource utilization, reduced traffic congestion in real‐time scenarios, and better user satisfaction in the IoT‐enabled environment.

Recognition of car horns based on principal component analysis of MEL frequency Cepstral coefficients and support vector machine

<title>Abstract</title> Background The 2018 international classification for periodontitis enables individualized patient management through simultaneous staging (disease severity and complexity) and grading (progression rate). However, its clinical adoption is hindered by diagnostic inconsistency—particularly in grading and differentiation of advanced stages—due to the complexity of age-adjusted metrics such as radiographic bone loss–to–age ratio (RBL/age). While artificial intelligence (AI) shows promise in dental diagnostics, existing tools lack robust integration of multimodal clinical data and validation under real-world conditions. Methods We retrospectively collected data from 692 patients diagnosed with periodontitis at Hospital of Stomatology, Guangxi Medical University between June 2022 and June 2024. After data cleaning and feature selection, cases were labeled according to the 2018 international classification criteria. Machine learning models—including k-nearest neighbors (KNN), Random Forest (RF), and Decision Tree (DT)—were trained and optimized via hyperparameter tuning. Model performance was evaluated on internal test sets and further validated on a temporally held-out external cohort (n = 208). Results For staging, KNN, RF, and DT achieved 96.99% accuracy; DT showed the highest recall (94.74%) and F1-score (0.9231) for Stage II, while RF excelled in Stage III (recall and F1-score: 0.9747). For grading, KNN and RF reached 96.40% accuracy, with F1-scores &gt; 0.97 for Grade C. Feature importance analysis identified clinical attachment loss (CAL) and probing depth (PD) as top predictors for staging, and RBL/age as the dominant feature for grading. In temporal external validation, the model achieved 93.75% accuracy for full diagnosis (extent + stage + grade). Conclusions The proposed model demonstrates high accuracy, interpretability, and generalizability, offering a promising decision-support tool for standardized periodontitis classification.

Female Genital mutilation (FGM) has been associated with numerous negative health effects like sexual dysfunction, chronic pain, infections, infertility, and an increased risk of maternal mortality. To date, no recognized studies in sub-Saharan Africa (SSA) have combined machine learning (ML) models with nationally representative Demographic and Health Survey (DHS) data to classify FGM. Therefore, this study aims to identify the most effective ML model for classifying FGM in SSA. ML offers the significant advantage of handling extremely large and complex datasets with numerous variables and interactions, providing more accurate classifications than conventional statistical methods. This study used secondary data from the DHS collected between 2015 and 2023 in 11 SSA countries. The DHS employs a cross-sectional study design. A total of 62,249 women who had at least one daughter were included in the analysis. Seven classification algorithms, including Logistic Regression, Decision Tree, Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), XGBoost (XGB), and Naive Bayes (NB), were used to classify FGM. The RF classifier yielded the highest performance, achieving 0.85 accuracy, 0.83 precision, 0.88 recall, 0.85 F1 score, and 0.93 AUC. Random Forest (Brier score = 0.1319) and Decision Tree (0.1328) showed best calibration. The SHAP plots provide a comprehensive interpretation of the features influencing the model's classification of whether a mother reports that at least one daughter has undergone FGM. The bar plot shows the global importance of each feature, with opinion on FGM (+ 0.14), country of residence (+ 0.13), respondent's circumcision status (+ 0.11), and religious justification for FGM (+ 0.07) contributing the most on average to the classification of FGM. This study applied seven machine ML techniques to classify FGM in SSA, with the Random Forest classifier demonstrating the best performance. SHAP interpretability findings indicated that maternal opinion on FGM, country of residence, religious justifications for FGM, and the respondent's circumcision status contributed most, on average, to the classification of FGM. These results suggest that education and advocacy efforts aimed at mothers particularly those who have had FGM themselves or who believe that the practice is justified by religion should be given top priority in interventions. Implementing community engagement initiatives that challenge accepted conventions and ideas around FGM may be particularly beneficial, especially when these practices are rooted in religious or cultural justifications.

Gesture recognition is a topic in computer science and language technology that aims to interpret human gestures with computer programs and many different algorithms. It can be seen as the way computers can understand human body language. Today, the main interaction tools between computers and humans are still the keyboard and mouse. Gesture recognition can be used as a tool for communication with the machine and interaction without any mechanical device such as a keyboard or mouse. In this paper, we present the results of a comparison of eight different machine learning (ML) classifiers in the task of human hand gesture recognition and classification to explore how to efficiently implement one or more tested ML algorithms on an 8-bit AVR microcontroller for on-line human gesture recognition with the intention to gesturally control the mobile robot. The 8-bit AVR microcontrollers are still widely used in the industry, but due to their lack of computational power and limited memory, it is a challenging task to efficiently implement ML algorithms on them for on-line classification. Gestures were recorded by using inertial sensors, gyroscopes, and accelerometers placed at the wrist and index finger. One thousand and eight hundred (1800) hand gestures were recorded and labelled. Six important features were defined for the identification of nine different hand gestures using eight different machine learning classifiers: Decision Tree (DT), Random Forests (RF), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) with linear kernel, Naïve Bayes classifier (NB), K-Nearest Neighbours (KNN), and Stochastic Gradient Descent (SGD). All tested algorithms were ranged according to Precision, Recall, and F1-score (abb.: P-R-F1). The best algorithms were SVM (P-R-F1: 0.9865, 0.9861, and 0.0863), and RF (P-R-F1: 0.9863, 0.9861, and 0.0862), but their main disadvantage is their unusability for on-line implementations in 8-bit AVR microcontrollers, as proven in the paper. The next best algorithms have had only slightly poorer performance than SVM and RF: KNN (P-R-F1: 0.9835, 0.9833, and 0.9834) and LR (P-R-F1: 0.9810, 0.9810, and 0.9810). Regarding the implementation on 8-bit microcontrollers, KNN has proven to be inadequate, like SVM and RF. However, the analysis for LR has proved that this classifier could be efficiently implemented on targeted microcontrollers. Having in mind its high F1-score (comparable to SVM, RF, and KNN), this leads to the conclusion that the LR is the most suitable classifier among tested for on-line applications in resource-constrained environments, such as embedded devices based on 8-bit AVR microcontrollers, due to its lower computational complexity in comparison with other tested algorithms.

Single-cell RNA-seq overcomes the shortcomings of conventional transcriptome sequencing technology and could provide a powerful tool for distinguishing the transcriptome characteristics of various cell types in biological tissues, and comprehensively revealing the heterogeneity of gene expression between cells. Many Intelligent Computing methods have been presented to infer gene regulatory network (GRN) with single-cell RNA-seq data. In this paper, we investigate the performances of seven classifiers including support vector machine (SVM), random forest (RF), Naive Bayesian (NB), GBDT, logical regression (LR), decision tree (DT) and K-Nearest Neighbor (KNN) for solving the binary classification problems of GRN inference with single-cell RNA-seq data (Single_cell_GRN). In SVM, three different kernel functions (linear, polynomial and radial basis function) are utilized, respectively. Three real single-cell RNA-seq datasets from mouse and human are utilized. The experiment results prove that in most cases supervised learning methods (SVM, RF, NB, GBDT, LR, DT and KNN) perform better than unsupervised learning method (GENIE3) in terms of AUC. SVM, RF and KNN have the better performances than other four classifiers. In SVM, linear and polynomial kernels are more fit to model single-cell RNA-seq data.