Machine Learning Classification Algorithms Research Articles

An overview of machine learning classification techniques

Machine learning (ML) is a key component within the broader field of artificial intelligence (AI) that employs statistical methods to empower computers with the ability to learn and make decisions autonomously, without the need for explicit programming. It is founded on the concept that computers can acquire knowledge from data, identify patterns, and draw conclusions with minimal human intervention. The main categories of ML include supervised learning, unsupervised learning, semisupervised learning, and reinforcement learning. Supervised learning involves training models using labelled datasets and comprises two primary forms: classification and regression. Regression is used for continuous output, while classification is employed for categorical output. The objective of supervised learning is to optimize models that can predict class labels based on input features. Classification is a technique used to predict similar information based on the values of a categorical target or class variable. It is a valuable method for analyzing various types of statistical data. These algorithms have diverse applications, including image classification, predictive modeling, and data mining. This study aims to provide a quick reference guide to the most widely used basic classification methods in machine learning, with advantages and disadvantages. Of course, a single article cannot be a complete review of all supervised machine learning classification algorithms. It serves as a valuable resource for both academics and researchers, providing a guide for all newcomers to the field, thereby enriching their comprehension of classification methodologies.

Hybrid computing framework security in dynamic offloading for IoT-enabled smart home system.

In the distributed computing era, cloud computing has completely changed organizational operations by facilitating simple access to resources. However, the rapid development of the IoT has led to collaborative computing, which raises scalability and security challenges. To fully realize the potential of the Internet of Things (IoT) in smart home technologies, there is still a need for strong data security solutions, which are essential in dynamic offloading in conjunction with edge, fog, and cloud computing. This research on smart home challenges covers in-depth examinations of data security, privacy, processing speed, storage capacity restrictions, and analytics inside networked IoT devices. We introduce the Trusted IoT Big Data Analytics (TIBDA) framework as a comprehensive solution to reshape smart living. Our primary focus is mitigating pervasive data security and privacy issues. TIBDA incorporates robust trust mechanisms, prioritizing data privacy and reliability for secure processing and user information confidentiality within the smart home environment. We achieve this by employing a hybrid cryptosystem that combines Elliptic Curve Cryptography (ECC), Post Quantum Cryptography (PQC), and Blockchain technology (BCT) to protect user privacy and confidentiality. Additionally, we comprehensively compared four prominent Artificial Intelligence anomaly detection algorithms (Isolation Forest, Local Outlier Factor, One-Class SVM, and Elliptic Envelope). We utilized machine learning classification algorithms (random forest, k-nearest neighbors, support vector machines, linear discriminant analysis, and quadratic discriminant analysis) for detecting malicious and non-malicious activities in smart home systems. Furthermore, the main part of the research is with the help of an artificial neural network (ANN) dynamic algorithm; the TIBDA framework designs a hybrid computing system that integrates edge, fog, and cloud architecture and efficiently supports numerous users while processing data from IoT devices in real-time. The analysis shows that TIBDA outperforms these systems significantly across various metrics. In terms of response time, TIBDA demonstrated a reduction of 10-20% compared to the other systems under varying user loads, device counts, and transaction volumes. Regarding security, TIBDA's AUC values were consistently higher by 5-15%, indicating superior protection against threats. Additionally, TIBDA exhibited the highest trustworthiness with an uptime percentage 10-12% greater than its competitors. TIBDA's Isolation Forest algorithm achieved an accuracy of 99.30%, and the random forest algorithm achieved an accuracy of 94.70%, outperforming other methods by 8-11%. Furthermore, our ANN-based offloading decision-making model achieved a validation accuracy of 99% and reduced loss to 0.11, demonstrating significant improvements in resource utilization and system performance.

Machine Learning Based Developmental Capability Prediction: A Diagnosis to the Learning Capacity Disorder for Specially-Abled Children

Specially-abled people are recognized and acknowledged for their issues such as hyperactivity, learning disorder, proprioceptive sensory issues, problems in self-help skills and problems in various motor skills such as Gross Motor Skills (GMS), Fine Motor Skills (FMS) and Oral Motor Skills (OMS), This study sought to identify effective machine-learning-based classification models to predict developmental capability disorders and thereby addressing of the learning disorder issue at opportune time. We have used machine learning classification algorithms Decision Tree, Random Forest, K-nearest neighbors, and Logistic Regression for the developmental capability prediction of individuals. The generalized progress monitoring datasets were carried out by interpreting and visualizing gender, age and disability-specific developmental competence. We have collected dataset from an occupational therapist for the study. The results of the study show that the Random Forest algorithm has a high accuracy of 95% compared to other algorithms that we have implemented.

Application of Big Data Analysis in Model Construction to Prevent Athlete Injury in Training

Abstract Sports injury is a hot issue in the field of exercise science and sports medicine and a practical problem that needs to be solved urgently. Based on big data analysis, this paper proposes an Ada-RF prediction model that integrates the fusion of the Adaboost algorithm and Random Forest algorithm by studying five machine learning algorithms: logistic regression, k-nearest neighbor, plain Bayes, Adaboost algorithm, and Random Forest and evaluates experiments through the model evaluation criteria. The influence factors of injury risk in adolescent male basketball players were explored in terms of demographic information, training load, subjective perceived health, and assessment of athletic quality. By screening the important factors as independent variables, the risk of athlete injury and illness as dependent variables were applied to the model constructed in this paper and the models built by five traditional machine learning classification algorithms. The prediction effects of multiple models are compared. The experimental results show that the accuracy, recall, specificity, F1 score, sensitivity, and AUC value of the Ada-RF model constructed in this paper are 0.869, 0.885, 0.755, 0.683, 0.754, and 0.789, respectively, which indicate that the Ada-RF integrated model outperforms the single prediction model, and can be used for the early prevention and early treatment of athletes’ injuries and illnesses. It shows that the Ada-RF integrated model outperforms the single prediction model and can provide scientific and accurate auxiliary information for the early prevention and treatment of injuries and diseases in athletes.

Use of Raman spectroscopy to study rat lung tissues for distinguishing asphyxia from sudden cardiac death.

Determining asphyxia as the cause of death is crucial but is based on an exclusive strategy because it lacks sensitive and specific morphological characteristics in forensic practice. In some cases where the deceased has underlying heart disease, differentiation between asphyxia and sudden cardiac death (SCD) as the primary cause of death can be challenging. Herein, Raman spectroscopy was employed to detect pulmonary biochemical differences to discriminate asphyxia from SCD in rat models. Thirty-two rats were used to build asphyxia and SCD models, with lung samples collected immediately or 24 h after death. Twenty Raman spectra were collected for each lung sample, and 640 spectra were obtained for further data preprocessing and analysis. The results showed that different biochemical alterations existed in the lung tissues of the rats that died from asphyxia and SCD and could be used to distinguish between the two causes of death. Moreover, we screened and used 8 of the 11 main differential spectral features that maintained their significant differences at 24 h after death to successfully determine the cause of death, even with decomposition and autolysis. Eventually, seven prevalent machine learning classification algorithms were employed to establish classification models, among which the support vector machine exhibited the best performance, with an area under the curve value of 0.9851 in external validation. This study shows the promise of Raman spectroscopy combined with machine learning algorithms to investigate differential biochemical alterations originating from different deaths to aid determining the cause of death in forensic practice.

Implementation of Machine Learning Classification Algorithm Based on Ensemble Learning for Detection of Vegetable Crops Disease

Implementation of Machine Learning Classification Algorithm Based on Ensemble Learning for Detection of Vegetable Crops Disease

Analysis of uncertainty of neural fingerprint-based models.

Machine learning has gained popularity for predicting molecular properties based on molecular structure. This study explores the uncertainty estimates of neural fingerprint-based models by comparing pure graph neural networks (GNN) to classical machine learning algorithms combined with neural fingerprints. We investigate the advantage of extracting the neural fingerprint from the GNN and integrating it into a method known for producing better-calibrated probability estimates. Comparisons are made using three classical machine learning methods and the Chemprop model, considering different molecular representations and calibration techniques. We utilize 19 datasets from Toxcast, reflecting real-world scenarios with balanced accuracies ranging from 0.6 to 0.8. Results demonstrate that neural fingerprints combined with classical machine learning methods exhibit a slight decrease in prediction performance compared to the native Chemprop model. However, these models provide significantly improved uncertainty estimates. Notably, uncertainty estimates of neural fingerprint-based methods remain relatively robust for molecules dissimilar to the training set. This suggests that methods like random forest with neural fingerprints can deliver strong prediction performance and reliable uncertainty estimates. When considering both performance and uncertainty, the calibrated Chemprop model and the combination of neural fingerprints with random forest or support vector classifier (SVC) yield comparable results. Surprisingly, the SVC method shows promising performance when combined with neural or count fingerprints. These findings are particularly relevant in real-world industrial projects where accurate predictions and reliable uncertainty estimates are crucial.

Detection of cardiovascular disease using explainable artificial intelligence and gut microbiota data

Purpose:Gut microbiota are defined as the microbial population of the intestines. They include various types of bacteria which can influence and predict the existence or onset of some specific diseases. Therefore, it is a common practice in medicine to analyze the gut microbiota for diagnostic purposes by analyzing certain measurable biochemical features associated with the disease under investigation. However, the evaluation of all the data collected from the gut microbiota is a labor-intensive process. Artificial Intelligence (AI) may be a helpful tool to identify the hidden patterns in gut microbiota for the detection of disease and other classification problems. Methods:In this study, we propose a deep neural model based on a one-dimensional convolutional neural network (1D-CNN) to detect cardiovascular disease using bacterial taxonomy and OTU (Operational Taxonomic Unit) table data. The developed AI method is compared to classical machine learning algorithms, regression, boosting algorithms and a deep model, Tabular Network (TabNet), developed for tabular data and obtained outperforming classification results. Results:According to AUC (Area Under Curve) values, boosting and regression methods outperformed the classical machine learning methods. However, the highest value of 97.09 AUC was obtained with the developed 1D-CNN model by using bacterial taxonomy data even with less then expected number of samples. Using explainable AI, nine bacteria were identified which the models find important for classification. Conclusion:The proposed method is robust and well adapted to taxonomy data in tabular form. It can be easily adapted to detect other diseases by using taxonomy data. The study also investigated the effect on barcode sequence for the classification, but the result showed that barcode sequences do not contribute to the bacterial taxonomy data for the estimation of CVD disease.

Critical Review of Stack Ensemble Classifier for the Prediction of Young Adults’ Voting Patterns Based on Parents’ Political Affiliations

Aim/Purpose: This review paper aims to unveil some underlying machine-learning classification algorithms used for political election predictions and how stack ensembles have been explored. Additionally, it examines the types of datasets available to researchers and presents the results they have achieved. Background: Predicting the outcomes of presidential elections has always been a significant aspect of political systems in numerous countries. Analysts and researchers examining political elections rely on existing datasets from various sources, including tweets, Facebook posts, and so forth to forecast future elections. However, these data sources often struggle to establish a direct correlation between voters and their voting patterns, primarily due to the manual nature of the voting process. Numerous factors influence election outcomes, including ethnicity, voter incentives, and campaign messages. The voting patterns of successors in regions of countries remain uncertain, and the reasons behind such patterns remain ambiguous. Methodology: The study examined a collection of articles obtained from Google Scholar, through search, focusing on the use of ensemble classifiers and machine learning classifiers and their application in predicting political elections through machine learning algorithms. Some specific keywords for the search include “ensemble classifier,” “political election prediction,” and “machine learning”, “stack ensemble”. Contribution: The study provides a broad and deep review of political election predictions through the use of machine learning algorithms and summarizes the major source of the dataset in the said analysis. Findings: Single classifiers have featured greatly in political election predictions, though ensemble classifiers have been used and have proven potent use in the said field is rather low. Recommendation for Researchers: The efficacy of stack classification algorithms can play a significant role in machine learning classification when modelled tactfully and is efficient in handling labelled datasets. however, runtime becomes a hindrance when the dataset grows larger with the increased number of base classifiers forming the stack. Future Research: There is the need to ensure a more comprehensive analysis, alternative data sources rather than depending largely on tweets, and explore ensemble machine learning classifiers in predicting political elections. Also, ensemble classification algorithms have indeed demonstrated superior performance when carefully chosen and combined.

Food Sales Prediction Using MLP, RANSAC, and Bagging

Many datasets about food sales, these datasets contain different features depending on the data present. Also, the way these features are correlated differs from one dataset to another. The researchers used several artificial intelligence algorithms and applied them to food sales datasets. Despite the necessary pre-processing and cleaning of the datasets, some of the algorithms used in these studies did not give the desired results. Therefore, this study proposes a model based on two objectives, the first objective is to make a comparison between three different food sales datasets. the second objective is to apply three various Artificial Intelligence algorithms to obtain the best algorithm that gives the highest prediction accuracy with the specified dataset. Some studies used classical machine learning algorithms, some used deep learning algorithms, and others used ensemble techniques. To achieve a comprehensive comparison, one algorithm was chosen from each of the above. To measure the correlation between features used a tool available from the Seaborn library in Python. This tool is called a “Heatmap”. For comparison, used three datasets on which we performed the necessary preprocessing operations, after applying three algorithms, these algorithms are Multilayer perceptron, RANSAC, and Bagging regression. Then used several metrics to measure the accuracy of the algorithm applied to the specified dataset. Finally, identified the best dataset that gives excellent prediction results with these algorithms. The results showed that the first dataset gave ideal accuracy by using the Bagging regression algorithm, unlike the second dataset with medium correlation and the third dataset with weak correlation. This study lays the foundation for subsequent studies and saves them time in terms of choosing the datasets.

A Comparative Study Evaluated the Performance of Two-class Classification Algorithms in Machine Learning

Worldwide, heart attacks, also called myocardial infarctions, are a leading cause of death. Early detection and accurate prediction of heart attacks are crucial for effective medical intervention and patient care. In recent years, machine learning techniques have shown great promise in aiding the diagnosis and prediction of heart attacks. The Organization for World Health (WHO) reports that around 17 million individuals worldwide pass away from cardiovascular diseases (CVD), notably heart attacks and strokes, each year. In this study, 1026 patients, both men and women, are almost equally affected by CVDs. While heart attacks and strokes remain among the leading causes of mortality worldwide, the use of machine learning for predicting heart disease could potentially prevent premature deaths. A comparative study evaluated the performance of five well-known two-class classification algorithms: two-class boosted decision trees, two-class decision forests, two-class locally deep SVMs, two-class neural networks, and two-class logistic regression. Among these algorithms, the Two-Class Boosted Decision Tree method demonstrated outstanding prediction ability, achieving a 100% accuracy rating. Its exceptional recall and precision rates highlight its effectiveness in handling challenging classifications. To facilitate the development and deployment of machine learning models, Azure Machine Learning offers a range of tools and services. By leveraging Azure Machine Learning's capabilities, researchers and healthcare professionals can analyze large datasets containing patient information and medical records to identify patterns and risk factors associated with heart attacks.

Culture and heritage tourism sentiment classification through cross-industry standard process for data mining

This study investigates the efficacy of machine learning algorithms in sentiment classification within the context of Culture and Heritage Tourism content analysis. This study adopts the CRISP-DM method, a comprehensive methodology encompassing distinct stages, including business understanding, data understanding, modeling, evaluation, and deployment. The k-nearest Neighbors, Decision Tree, Naive Bayes Classifier, and Support Vector Machine models are used. The performance of each model is scrutinized through confusion matrix analysis, encompassing metrics such as accuracy, precision, recall, and F-measure. Additionally, the impact of the Synthetic Minority Over-sampling Technique (SMOTE) implementation on addressing data imbalance is assessed. Leveraging data from the national geographic channel's YouTube platform, with a focus on ma'nene content, results reveal SVM's consistent superiority, particularly with SMOTE integration, showcasing elevated accuracy (77.89%), precision (72.60%), recall (89.62%), and F-measure (80.21%) values. These findings underscore the importance of algorithm selection and data preprocessing methods in enhancing sentiment classification accuracy for culture and heritage tourism content, thus contributing quantifiable insights to the tourism research domain.

Retracted: Brand Digital Marketing under Intranet Security Control Based on the Machine Learning Classification Algorithm

Retracted: Brand Digital Marketing under Intranet Security Control Based on the Machine Learning Classification Algorithm

Artificial Intelligence on Diagnostic Aid of Leprosy: A Systematic Literature Review.

Leprosy is a neglected tropical disease that can cause physical injury and mental disability. Diagnosis is primarily clinical, but can be inconclusive due to the absence of initial symptoms and similarity to other dermatological diseases. Artificial intelligence (AI) techniques have been used in dermatology, assisting clinical procedures and diagnostics. In particular, AI-supported solutions have been proposed in the literature to aid in the diagnosis of leprosy, and this Systematic Literature Review (SLR) aims to characterize the state of the art. This SLR followed the preferred reporting items for systematic reviews and meta-analyses (PRISMA) framework and was conducted in the following databases: ACM Digital Library, IEEE Digital Library, ISI Web of Science, Scopus, and PubMed. Potentially relevant research articles were retrieved. The researchers applied criteria to select the studies, assess their quality, and perform the data extraction process. Moreover, 1659 studies were retrieved, of which 21 were included in the review after selection. Most of the studies used images of skin lesions, classical machine learning algorithms, and multi-class classification tasks to develop models to diagnose dermatological diseases. Most of the reviewed articles did not target leprosy as the study's primary objective but rather the classification of different skin diseases (among them, leprosy). Although AI-supported leprosy diagnosis is constantly evolving, research in this area is still in its early stage, then studies are required to make AI solutions mature enough to be transformed into clinical practice. Expanding research efforts on leprosy diagnosis, coupled with the advocacy of open science in leveraging AI for diagnostic support, can yield robust and influential outcomes.

Prediction Using SVM and DT

Diabetes is recognized as a severe and persistent illness that leads to an elevation in blood sugar levels. Untreated and undetected diabetes can give rise to numerous complications. The laborious process of identification typically involves a patient visiting a diagnostic centre and consulting with a doctor. However, the advancement of machine learning approaches has addressed this critical issue. The objective of this study is to develop a model capable of predicting the likelihood of diabetes in patients with the utmost accuracy. To achieve this goal, three machine learning classification algorithms—Decision Tree, SVM, and Naive Bayes—are employed in this experiment to identify diabetes at an early stage. The experiments are conducted on the Pima Indians Diabetes Database (PIDD), sourced from the UCI machine learning repository. The performance of each algorithm is assessed using various metrics such as Precision, Accuracy, F-Measure, and Recall. Accuracy is gauged based on both correctly and incorrectly classified instances. The results indicate that Naive Bayes outperforms the other algorithms, achieving the highest accuracy of 76.30%. These findings are corroborated through a meticulous examination of Receiver Operating Characteristic (ROC) curves.

A novel machine learning algorithm for finger movement classification from surface electromyogram signals using welch power estimation

Electromyogram (EMG) signal monitoring is an effective method for controlling the movements of a prosthetic limb. The classification of the EMG pattern of various finger motions in upper-arm amputees has drawn much attention in recent years to develop algorithms that provide adequate accuracy. However, due to the complexity of EMG data, movement detection is a challenging task. Therefore, an effective model is needed that can accurately process, analyze, and classify various hand and finger movements. This paper proposes a novel algorithm for processing and classifying 15 finger movements from surface EMG signals based on Welch power estimation from frequency analysis. Five time-domain features are extracted and trained with a machine learning classifier to classify 15 single fingers and combined finger gestures from eight healthy subjects. The experimental results show 92.30 % classification accuracy considering data from eight channels which was improved to 94.15 % after selecting two channels as dominating. For performance evaluation, 10-fold cross-validation is used during classification. We demonstrate an average accuracy of 92.35 % with 25 % test data.

Sequence-Type Classification of Brain MRI for Acute Stroke Using a Self-Supervised Machine Learning Algorithm.

We propose a self-supervised machine learning (ML) algorithm for sequence-type classification of brain MRI using a supervisory signal from DICOM metadata (i.e., a rule-based virtual label). A total of 1787 brain MRI datasets were constructed, including 1531 from hospitals and 256 from multi-center trial datasets. The ground truth (GT) was generated by two experienced image analysts and checked by a radiologist. An ML framework called ImageSort-net was developed using various features related to MRI acquisition parameters and used for training virtual labels and ML algorithms derived from rule-based labeling systems that act as labels for supervised learning. For the performance evaluation of ImageSort-net (MLvirtual), we compare and analyze the performances of models trained with human expert labels (MLhumans), using as a test set blank data that the rule-based labeling system failed to infer from each dataset. The performance of ImageSort-net (MLvirtual) was comparable to that of MLhuman (98.5% and 99%, respectively) in terms of overall accuracy when trained with hospital datasets. When trained with a relatively small multi-center trial dataset, the overall accuracy was relatively lower than that of MLhuman (95.6% and 99.4%, respectively). After integrating the two datasets and re-training them, MLvirtual showed higher accuracy than MLvirtual trained only on multi-center datasets (95.6% and 99.7%, respectively). Additionally, the multi-center dataset inference performances after the re-training of MLvirtual and MLhumans were identical (99.7%). Training of ML algorithms based on rule-based virtual labels achieved high accuracy for sequence-type classification of brain MRI and enabled us to build a sustainable self-learning system.

Using Natural Language Processing to Identify Stigmatizing Language in Labor and Birth Clinical Notes.

Stigma and bias related to race and other minoritized statuses may underlie disparities in pregnancy and birth outcomes. One emerging method to identify bias is the study of stigmatizing language in the electronic health record. The objective of our study was to develop automated natural language processing (NLP) methods to identify two types of stigmatizing language: marginalizing language and its complement, power/privilege language, accurately and automatically in labor and birth notes. We analyzed notes for all birthing people > 20 weeks' gestation admitted for labor and birth at two hospitals during 2017. We then employed text preprocessing techniques, specifically using TF-IDF values as inputs, and tested machine learning classification algorithms to identify stigmatizing and power/privilege language in clinical notes. The algorithms assessed included Decision Trees, Random Forest, and Support Vector Machines. Additionally, we applied a feature importance evaluation method (InfoGain) to discern words that are highly correlated with these language categories. For marginalizing language, Decision Trees yielded the best classification with an F-score of 0.73. For power/privilege language, Support Vector Machines performed optimally, achieving an F-score of 0.91. These results demonstrate the effectiveness of the selected machine learning methods in classifying language categories in clinical notes. We identified well-performing machine learning methods to automatically detect stigmatizing language in clinical notes. To our knowledge, this is the first study to use NLP performance metrics to evaluate the performance of machine learning methods in discerning stigmatizing language. Future studies should delve deeper into refining and evaluating NLP methods, incorporating the latest algorithms rooted in deep learning.

ADJUSTING THE PARAMETERS OF MACHINE LEARNING ALGORITHMS TO IMPROVE THE SPEED AND ACCURACY OF TRAFFIC CLASSIFICATION

Background. Telecommunications developments lead to new mobile network technologies and especially 5G, which has only recently been launched, sixth generation of which is already under active development. The development of new technologies influence on both types of mobile traffic (V2V, IoT) and leads to the significant increase in the volume of existing traffic types. Currently, existing methods of traffic processing are not adapted to such changes, which may lead to a deterioration in the quality of service.
 Objective. The purpose of the paper is to analyze the effectiveness of machine learning algorithms to solve the task of traffic classification in mobile networks in real time.
 Methods. The method of solving the problem of increasing the efficiency of information processing is the introduction of new algorithms for traffic classification and prioritization. In this regard, the paper presents the urgent task of analyzing the effectiveness of machine learning algorithms to solve the task of traffic classification in mobile networks in real time.
 Results. Comparison indicated the best accuracy of the ANN algorithm that was achieved with the number of hidden layers of the network equal to 200. Also, the research results showed that different applications have different recognition accuracy, which does not depend on the total number of packets in the dataset.
 Conclusions. This proceeding solves the urgent problem of increasing the efficiency of the mobile communication system through the use of machine learning algorithms for traffic classification. In this regard, it can be concluded that the most promising is the application of algorithms based on ANN. In future the aspect of anomaly detection based on traffic classification and traffic pattern preparation should be investigated, as this process allows detecting attacks to network infrastructure and increase mobile network security.

Near-Infrared Spectroscopy with Supervised Machine Learning as a Screening Tool for Neutropenia.

The use of non-invasive tools in conjunction with artificial intelligence (AI) to detect diseases has the potential to revolutionize healthcare. Near-infrared spectroscopy (NIR) is a technology that can be used to analyze biological samples in a non-invasive manner. This study evaluated the use of NIR spectroscopy in the fingertip to detect neutropenia in solid-tumor oncologic patients. A total of 75 patients were enrolled in the study. Fingertip NIR spectra and complete blood counts were collected from each patient. The NIR spectra were pre-processed using Savitzky-Golay smoothing and outlier detection. The pre-processed data were split into training/validation and test sets using the Kennard-Stone method. A toolbox of supervised machine learning classification algorithms was applied to the training/validation set using a stratified 5-fold cross-validation regimen. The algorithms included linear discriminant analysis (LDA), logistic regression (LR), random forest (RF), multilayer perceptron (MLP), and support vector machines (SVMs). The SVM model performed best in the validation step, with 85% sensitivity, 89% negative predictive value (NPV), and 64% accuracy. The SVM model showed 67% sensitivity, 82% NPV, and 57% accuracy on the test set. These results suggest that NIR spectroscopy in the fingertip, combined with machine learning methods, can be used to detect neutropenia in solid-tumor oncology patients in a non-invasive and timely manner. This approach could help reduce exposure to invasive tests and prevent neutropenic patients from inadvertently undergoing chemotherapy.