Decision Tree Algorithm Research Articles

Nutritional intake of micronutrient and macronutrient and type 2 diabetes: machine learning schemes

BackgroundDiabetes mellitus, an endocrine system disease, is a common disease involving many patients worldwide. Many studies are performed to evaluate the correlation between micronutrients/macronutrients on diabetes but few of them have a high statistical population and a long follow-up period. We aimed to investigate the relationship between intake of macro/micronutrients and the incidence of type 2 diabetes (T2D) using logistic regression (LR) and a decision tree (DT) algorithm for machine learning.MethodOur research explores supervised machine learning models to identify T2D patients using the Mashhad Cohort Study dataset. The study population comprised 9704 individuals aged 35–65 years were enrolled regarding their T2D status, and those with T2D history. 15% of individuals are diabetic and 85% of them are non-diabetic. For ten years (until 2020), the participants in the study were monitored to determine the incidence of T2D. LR is a statistical model applied in dichotomous response variable modeling. All data were analyzed by SPSS (Version 22) and SAS JMP software.ResultNutritional intake in the T2D group showed that potassium, calcium, magnesium, zinc, iodine, carotene, vitamin D, tryptophan, and vitamin B12 had an inverse correlation with the incidence of diabetes (p < 0.05). While phosphate, iron, and chloride had a positive relationship with the risk of T2D (p < 0.05). Also, the T2D group significantly had higher carbohydrate and protein intake (p-value < 0.05).ConclusionMachine learning models can identify T2D risk using questionnaires and blood samples. These have implications for electronic health records that can be explored further.

Optimization of Intrusion Detection System with Machine Learning for Detecting Distributed Attacks on Server

This study develops an intrusion detection system optimized with machine learning techniques for efficient and effective detection of Distributed Denial-of-Service (DDoS) attacks. Using the Decision Tree algorithm, the system is designed to maximise accuracy in the identification and classification of DDoS attacks. The CIC-DDoS2019 dataset, which consists of various comprehensive simulated attack scenarios, is used as the basis for training and validation, providing the model with robust capabilities in recognizing DDoS attacks with high accuracy. This IDS successfully achieved a 100% detection rate, which is a significant result in the network security environment. The system is integrated into the existing network infrastructure, monitoring data flows in real-time and performing predictive analysis to detect early indications of attacks. Each attack detection immediately triggers a notification sent via a Telegram bot, ensuring that the security team can react quickly to isolate and address the attack. These notifications include details such as the source, type of attack, detection time, and involved protocol information, enabling more informed and strategic response actions. The use of Telegram bots for real-time communication not only enhances the speed of response to threats but also supports system scalability by facilitating adjustments and integration across various operational scenarios. The system's quick detection and response is a big step forward for machine learning-based intrusion detection systems (IDS). It provides opportunities for further research and practical applications that can adapt to various digital security scenarios.

The relationship between activities of daily living and speech impediments based on evidence from statistical and machine learning analyses

IntroductionSpeech impediments (SIs) are increasingly prevalent among middle-aged and older adults, raising concerns within public health. Early detection of potential SI in this demographic is critical. This study investigates the potential of Activities of Daily Living (ADL) as a predictive marker for SI, utilizing data from the 2018 China Health and Retirement Longitudinal Study (CHARLS), which includes 10,136 individuals aged 45 and above. The Barthel Index (BI) was used to assess ADL, and the correlation between ADL and SI was examined through statistical analyses. Machine learning algorithms (Support Vector Machine, Decision Tree, and Logistic Regression) were employed to validate the findings and elucidate the underlying relationship between ADL and SI.BackgroundSI poses significant challenges to the health and quality of life of middle-aged and older adults, increasing the demands on community-based and home care services. In the context of global aging, it is crucial to investigate the factors contributing to SI. While the role of ADL as a potential biomarker for SI remains unclear, this study aims to provide new evidence supporting ADL as an early predictor of SI through statistical analysis and machine learning validation.MethodsData were derived from the 2018 CHARLS national baseline survey, comprising 10,136 participants aged 45 and above. ADL was evaluated using the BI, and SI was assessed based on the CHARLS records of “Speech impediments.” Statistical analyses, including independent sample t-tests, chi-square tests, Pearson and Spearman correlation tests, and hierarchical multiple linear regression, were conducted using SPSS 25.0. Machine learning algorithms, specifically Support Vector Machine (SVM), Decision Tree (DT), and Logistic Regression (LR), were implemented in Python 3.10.2.ResultsAnalysis of demographic characteristics revealed that the average BI score in the “With Speech impediments” group was 49.46, significantly lower than the average score of 85.11 in the “Without Speech impediments” group. Pearson correlation analysis indicated a significant negative correlation between ADL and SI (r = −0.205, p &amp;lt; 0.001). Hierarchical multiple linear regression confirmed the robustness of this negative correlation across three models (B = −0.001, β = −0.168, t = −16.16, 95% CI = −0.001 to −0.001, p = 0.000). Machine learning algorithms validated the statistical findings, confirming the predictive accuracy of ADL for SI, with the area under the curve (AUC) scores of SVM-AUC = 0.648, DT-AUC = 0.931, and LR-AUC = 0.666. The inclusion of BI in the models improved the overall predictive performance, highlighting its positive impact on SI prediction.ConclusionThe study employed various statistical methodologies to demonstrate a significant negative correlation between ADL and SI, a finding further corroborated by machine learning algorithms. Impairment in ADL increases the likelihood of SI occurrence, underscoring the importance of maintaining ADL in middle-aged and older populations to mitigate the risk of SI.

Transfer models and Standard models for predicting wear rates on the basis of friction noise: A Comparative Study

Abstract Under varying working conditions, friction between polymers and metals leads to surface wear of materials, accompanied by significant energy dissipation, part of which transforms into friction noise. Despite their high nonlinearity, friction noises share certain commonalities in their generation mechanisms. This study proposes a novel transfer mapping model, which, after modeling a specific pair, can predict the behavior of other pairs. We simplify the model through Pearson feature selection and employ decision tree-based algorithms (Decision Tree, Extreme Gradient Boosting, Categorical Boosting) to model the transfer mapping. By comparing the performance of standard models with transfer models, we identify the optimal approach for constructing the transfer model by using Categorical Boosting algorithm.

Evaluasi Kinerja Algoritma Klasifikasi dalam Studi Kasus Prediksi Kelulusan di Universitas XYZ

This study aims to evaluate the performance of three classification algorithms—Decision Tree, K-Nearest Neighbors (KNN), and Naive Bayes—in a case study of predicting student graduation outcomes at XYZ University. Utilizing a dataset that includes a diverse range of academic attributes of students, this research conducts an in-depth analysis of each algorithm's capability to predict students' graduation status. The findings reveal that the Decision Tree provides the best performance with an accuracy rate of 97%, while KNN also shows very good results with similar accuracy, and Naive Bayes records an accuracy of 91%. The advantage of the Decision Tree lies in its ease of interpretation and high decision transparency, making it highly suitable for educational system implementations. Meanwhile, KNN excels in handling complex datasets due to its adaptability to data variability, and Naive Bayes offers rapid analysis speeds ideal for processing large data volumes. These insights provide valuable guidance for educational institutions in developing more effective graduation prediction systems and supporting strategic decisions in academic management. The study also opens avenues for further research in exploring these algorithms' applications in broader educational contexts, as well as fine-tuning modeling techniques to enhance prediction accuracy and efficiency.

PENERAPAN ALGORITMA DECISION TREE UNTUK KLASIFIKASI KONSUMSI ENERGI LISTRIK RUMAH TANGGA DENGAN PENGGUNAAN RAPIDMINER

The research aims to explore and understand energy consumption patterns in households. By using the Decision Tree algorithm, to classify the level of electrical energy consumption. And data on household electrical energy consumption can be obtained from various sources, such as Household electricity meter. Survey or questionnaire filled out by homeowners regarding the use of electrical appliances. Based on the image above, the application of the Decision Tree algorithm in analyzing risk factors for The classification of household electrical energy consumption produces an accuracy value of 100.00%. From the displayed confusion matrix, we can see the distribution of predicted and actual values for various classes. For example, in the class "true 110 25," there are 17052 correct predictions. The evaluation results also show the precision and recall values for each class. The highest precision was achieved in the "true 2205" class with 100% recall, while the precision was found in the "true 122.5" class of 100.00%.

Prediction Study of Vocational College Students’ English- Speaking Proficiency based on Decision Tree Algorithm

Prediction Study of Vocational College Students’ English- Speaking Proficiency based on Decision Tree Algorithm

The Impact of Sports Reforms in the New Era on the Physical and Mental Health Development of College Students

With the development of economy and technology, more and more students are addicted to electronic products, which seriously damages their physical and mental health, and seriously affects their daily life and learning. Sports are an important component of school education and a crucial means of promoting students' comprehensive development. The essential function of sports is to improve people's health, enhance their physical fitness, and promote their socialization process. In order to promote the physical and mental health development of college students, this article uses association rule algorithm to analyze the relationship between sports and the physical and mental health of college students. This article constructs a research on mental health intelligence based on decision tree algorithm, using scientific mental health evaluation tools to comprehensively and objectively reflect the user's mental health level.

MORF: Multi-view oblique random forest for hepatotoxicity prediction.

Hepatotoxicity prediction is significant in drug development, so the experts expect to get effective and reliable references. Based on the consideration that the hepatotoxicity data involve multiple types of features, this paper proposes a multi-view oblique random forest (MORF) for hepatotoxicity prediction by considering each type of feature as an independent view. The Householder transformation is employed to get the inclined cut hyperplane in each view. Two versions of the multi-view oblique decision tree (ODT) algorithms were designed by generating optimal nodes based on selecting proper hyperplanes from different views, named ODT-N and ODT-R. These two types of ODT algorithms serve as the base learners to construct two MORF. Experiments conducted on the hepatotoxicity data provide performance comparisons among different algorithms, and the results confirm that our algorithms can fully utilize the information in different views.

The benefits of physical literacy for human flourishing: A machine learning-based exploration of adolescents.

Physical literacy is a multidimensional concept considered fundamental for lifelong participation in physical activity. Although theories on the relationship between physical literacy and human flourishing have been proposed, no comprehensive study of this relationship in adolescents has been found. This study aimed to predict human flourishing variables (physical activity, physical health, mental health, and academic performance) that correlate highly with adolescents' physical literacy. A sample of 1004 primary and middle school students was recruited and six machine learning algorithms (decision tree, random forest, AdaBoost, CatBoost, LightGBM, and k-nearest neighbours) were used. Random forest predicted physical activity, physical health, and mental health in the overall sample, with an accuracy of 53 percent, 86 per cent, and 91.7 percent, respectively; AdaBoost predicted academic performance with an accuracy of 98 per cent. Overall sample and sex subgroup predictions were generally consistent, with "sense of self and self-control" and "self-expression and communication with others" as the most significant variables. Family-type subgroup analysis results varied greatly, suggesting that one-child families should focus on "knowledge and understanding," whereas multi-child families should focus on "self-expression and communication with others." Awareness of adolescents' underlying characteristics may yield greater benefits when intervening through physical literacy.

Development and Validation of a Novel Model to Discriminate Idiosyncratic Drug-Induced Liver Injury and Autoimmune Hepatitis.

Discriminating between idiosyncratic drug-induced liver injury (DILI) and autoimmune hepatitis (AIH) is critical yet challenging. We aim to develop and validate a machine learning (ML)-based model to aid in this differentiation. This multicenter cohort study utilised a development set from Beijing Friendship Hospital, with retrospective and prospective validation sets from 10 tertiary hospitals across various regions of China spanning January 2009 to May 2023. Different ML algorithms were tested using 24 routine laboratory parameters. The Shapley Additive exPlanations (SHAP) analysis was used to evaluate the contribution of each parameter in the ML model. A total of 2554 patients (1750 for DILI and 804 for AIH) were included. Using Gradient Boost Decision Tree algorithm, five key parameters-aspartate transaminase, globulin, prealbumin, creatinine and platelet count-were selected to construct the ML model. Consequently, a web-based tool named Beijing-AID (BJ-AID) was developed (http://43.143.153.225:5000/). The BJ-AID model demonstrated excellent discrimination performance, with an area under the receiver operating characteristic curve (AUROC) of 0.94 (95% CI, 0.902-0.975) in the development set, 0.91 (95% CI, 0.900-0.928) in all external validation sets and 0.93 (95% CI, 0.889-0.974) in a prospective validation set. Notably, the BJ-AID model also effectively discriminated atypical cases, including drug-induced autoimmune-like hepatitis and AIH with the history of drug consumption, achieving an AUROC = 0.85 (95% CI, 0.742-0.949). We successfully developed and validated a machine learning-based model, BJ-AID, which exhibits a strong discrimination performance. BJ-AID can assist practitioners and hepatologists in diagnosing both typical and atypical cases of DILI and AIH. ClinicalTrials.gov identifier: NCT05532345.

PENGUNAAN DATA MINIG UNTUK MENGIDENTIFIKASI PELANGGAN BERESIKO TINGGI DALAM PENJUALAN MENGUNAKAN ALGORITMA DECITION TREE C4.5

In the competitive world of business, identifying high-risk customers is critical to minimizing churn rates and increasing profitability. This research uses data mining techniques using the C4.5 decision tree algorithm to classify customers based on their churn risk. The research stages include data collection, cleaning, data processing, as well as dividing the data into training and testing sets. The implementation of this algorithm was carried out using RapidMiner software, which facilitates customer clustering and predicting behavior based on historical attributes. The evaluation results show the model has an accuracy of 74.59%, with precision and recall indicating the model's ability to identify high-risk customers. Thus, the Decision Tree C4.5 algorithm is proven to be effective in supporting decision making for customer churn risk mitigation strategies.

Towards an interpretable breast cancer detection and diagnosis system.

According to the World Health Organization, breast cancer becomes fatal only if it spreads throughout the body. Therefore, regular screening is essential. Whilst mammography is the most frequently used technique, its interpretation can be challenging and time-consuming. For this reason, computer-aided detection and diagnosis systems are increasingly being used for second opinion. However, in order for doctors to trust such systems, they need to understand their decisions. We propose an automated and interpretable system for the detection and diagnosis of breast cancer, encompassing five steps. After a robust pre-processing and an unsupervised segmentation, we analyze four feature extraction techniques, both textural and shape-based, and three methods for feature selection. To facilitate interpretation, we employ the Decision Tree algorithm for benign/malignant classification and experiment with different methods to avoid overfitting: pre-pruning, post-pruning, and ensemble-based (Random Forest classifier). Our system reaches a maximum accuracy of 95% and 100% precision and specificity when tested on images from the mini-MIAS dataset, while also offering its users the possibility to analyze each of the steps.

PENERAPAN DECISION TREE C4.5 DALAM MEMPREDIKSI PREDIKAT TERBAIK DI MADRASAH TA'HILIYAH IBRAHIMY

To improve the evaluation process in assessing student progress, predicting the best grades plays a crucial role in enhancing the quality of education. By identifying the top-performing students, educational institutions can refine their teaching methods and create targeted strategies to foster better learning outcomes. This step is vital for ensuring that the learning process aligns with the institution's goals to produce highly skilled and knowledgeable students. In this research, we focused on utilizing the C4.5 algorithm, a widely recognized decision tree method in data mining, to predict student achievements. The C4.5 algorithm is known for its ability to classify and uncover hidden patterns within datasets, making it a powerful tool for educational data analysis. Through this approach, we aim to analyze the factors influencing student success and provide actionable insights for educators and administrators. The study was conducted on students from Madrasah Ta’hiliyah Ibrahimy, where we applied the decision tree algorithm to predict the best grades based on historical academic data. The experiment resulted in three distinct rules or patterns derived from the data, with an overall accuracy of 74.17%. These findings demonstrate the potential of data-driven approaches in supporting academic decision-making and guiding future interventions to further enhance student performance.

Enhancing roadway scenario classification through vision transformer features and decision tree algorithms

Abstract Drivers face significant visibility challenges throughout the day and night due to varying environmental and lighting conditions. During the day, heavy rain and dense fog can severely reduce visibility, making it difficult to identify road markings and navigate safely. At night, the absence of natural light exacerbates visibility issues, and the intense glare from high-beam headlights further impairs focus of the driver which affects his reaction times. These challenges significantly increase the risk of accidents. To address these issues, modern vehicles are equipped with advanced systems such as front-mounted cameras and displays to improve road visibility. However, maintaining display clarity under adverse conditions remains a challenge for them. This study proposes a novel method for classifying roadway scenarios using Vision Transformer (ViT) features combined with decision-tree-based classifiers and applying tailored image enhancement algorithms for each scenario. The method effectively distinguishes between six road scenarios, including adverse weather and low-light conditions. Using Random Forest as the classification model, the proposed approach achieved a maximum accuracy of 96.60%, outperforming state-of-the-art methods. By accurately classifying scenarios and applying appropriate enhancement techniques, this approach improves road visibility and enhances driver safety.

Cloud Access Security Brokers: An easy-implement Method for Accessing Cloud Services Securely

In this changing digital age, cloud services have become very common. But the main challenge is to provide secure access to cloud services for retailers and users and also for providers. Read here The Important Role of Cloud Access Security Brokers (CASBs). Either on-premise or cloud, CASBs take the place of hardened enforcement points of security in policy that are capable of bringing corporate security policy together, layering it on top of cloud resource access. These essential functions include authentication, single sign-on, authorization, credential mapping, device profiling, encryption, tokenization, many more. But in general, the most common challenges for CASBS provide these in the cloud: threat detection, access control policy enforcement, risk assessment, data protection and compliance. The literature survey presented in this paper focuses on these threats and vulnerabilities, and clearly highlight the requirement of improved protective processes in Cloud computing. Complementing CASBs with machine learning (ML) is at the heart of our proposed solution. If you are only looking at how well ML algorithms worked in detecting real-time threats, or in automating access control policies, comprehensive risk assessments, classifying sensitive data and monitoring compliance Machine learning techniques like Decision Tree and Random Forest algorithms have been applied with the initiative taken by us on CSE-CICIDS database, a real-world dataset that is described with the characteristics for cloud utilization behaviors along with limited numbers of security violation incidents occurred. The Random Forest Algorithm is performing considerably better than others, it has perfect precision, recall and F1-scores, it is able to predict all the records in the data set correctly.

Decision tree-based learning and laboratory data mining: an efficient approach to amebiasis testing

BackgroundAmebiasis represents a significant global health concern. This is especially evident in developing countries, where infections are more common. The primary diagnostic method in laboratories involves the microscopy of stool samples. However, this approach can sometimes result in the misinterpretation of amebiasis as other gastroenteritis (GE) conditions. The goal of the work is to produce a machine learning (ML) model that uses laboratory findings and demographic information to automatically predict amebiasis.MethodData extracted from Jordanian electronic medical records (EMR) between 2020 and 2022 comprised 763 amebic cases and 314 nonamebic cases. Patient demographics, clinical signs, microscopic diagnoses, and leukocyte counts were used to train eight decision tree algorithms and compare their accuracy of predictions. Feature ranking and correlation methods were implemented to enhance the accuracy of classifying amebiasis from other conditions.ResultsThe primary dependent variables distinguishing amebiasis include the percentage of neutrophils, mucus presence, and the counts of red blood cells (RBCs) and white blood cells (WBCs) in stool samples. Prediction accuracy and precision ranged from 92% to 94.6% when employing decision tree classifiers including decision tree (DT), random forest (RF), XGBoost, AdaBoost, and gradient boosting (GB). However, the optimized RF model demonstrated an area under the curve (AUC) of 98% for detecting amebiasis from laboratory data, utilizing only 300 estimators with a max depth of 20. This study highlights that amebiasis is a significant health concern in Jordan, responsible for 17.22% of all gastroenteritis episodes in this study. Male sex and age were associated with higher incidence of amebiasis (P = 0.014), with over 25% of cases occurring in infants and toddlers.ConclusionsThe application of ML to EMR can accurately predict amebiasis. This finding significantly contributes to the emerging use of ML as a decision support system in parasitic disease diagnosis.Graphical

Multi-band single bio-molecule detection through inverse-designed graphene-folded spherical particle dimers

The present research employs graphene-coated spherical nano-particles as the basis of optical dimers for multi-frequency refractive index sensing applications. Under parallelly polarized incoming waves, dual operating bands are attained owing to the presence of localized surface plasmon resonances (LSPRs) on graphene shells. Thus, enormous local near-field enrichment is detected at the gap middle as a result of plasmonic hybridization and strong coupling. The potential usage of the proposed dimer as a dual-band high-performance refractive index sensor, with the typical sensitivity of S1 = 2.8143 × 104 nm/RIU and figure of merit of FOM1 = 213.2860 RIU-1 in the first band, and sensitivity of S2 = 1.8070 × 104 nm/RIU and figure of merit of FOM2 = 305.1521 RIU-1 in the second band, is illustrated. Importantly, the enormous near-field enhancement is maintained for particles with different radii, making the element suitable for single bio-molecule recognition of various types by tuning the quality of the graphene layer. The spectral tuning is correspondingly viable after production by imposing a suitable Fermi level on the graphene shells. Finally, machine learning forward and reverse problems based on the random forest (RF), decision tree (DT), and eXtreme Gradient Boosting (XGBoost) algorithms are implemented to propose a method respectively for analysis of the sensor's performance and to design an appropriate sensor for each desired molecule.

Improving flood-prone areas mapping using geospatial artificial intelligence (GeoAI): A non-parametric algorithm enhanced by math-based metaheuristic algorithms.

Flooding presents substantial dangers to human lives and infrastructure, underscoring the need to map flood-prone areas to implement effective mitigation measures precisely. Although machine learning algorithms have made great strides, their accuracy in flood susceptibility mapping (FSM) remains limited due to data dependence, interpretability, and explainability issues, overfitting, generalization difficulties, and hyperparameter tuning. This study suggests combining the Decision Tree (DT) algorithm with advanced, math-based metaheuristic optimization algorithms to address these limitations. In this study, the FSM was prepared using a combination of the DT non-parametric algorithm and three math-based metaheuristic optimization algorithms, including the gradient-based optimizer (GBO), chaos game optimization (CGO), and arithmetic optimization algorithm (AOA). Four models were developed and compared for modeling and mapping: the basic DT, DT-AOA, DT-GBO, and DT-CGO. The DT-CGO model stood out on the training set, achieving the lowest root mean square error (RMSE) of 0.17. It was closely followed by DT-AOA (0.19), DT-GBO (0.2), and DT (0.201). In terms of mean absolute error (MAE), the DT-CGO model exhibited the lowest value of 0.06, followed by DT-AOA (0.07), DT-GBO (0.079), and the basic DT model (0.08). When considering the coefficient of determination (R2), the DT-CGO model achieved the highest value of 0.871, followed by DT-AOA (0.853), DT-GBO (0.84), and DT (0.838). The enhanced models performed superior over the test set's basic DT model. The area under the curve (AUC) values validated the enhanced models' efficacy, with the DT-CGO model achieving the highest AUC of 0.978, followed by DT-AOA (0.974), DT-GBO (0.967), and DT (0.958). The findings emphasize the importance of accurate flood-prone area identification and provide valuable insights for policymakers to develop robust flood mitigation strategies.

A machine learning based algorithm accurately stages liver disease by quantification of arteries

A major histologic feature of cirrhosis is the loss of liver architecture with collapse of tissue and vascular changes per unit. We developed qVessel to quantify the arterial density (AD) in liver biopsies with chronic disease of varied etiology and stage. 46 needle liver biopsy samples with chronic hepatitis B (CHB), 48 with primary biliary cholangitis (PBC) and 43 with metabolic dysfunction-associated steatotic liver disease (MASLD) were collected at the Shuguang Hospital. The METAVIR system was used to assess stage. The second harmonic generation (SHG)/two-photon images were generated from unstained slides. Collagen proportionate area (CPA) using SHG. AD was counted using qVessel (previously trained on manually labeled vessels by stained slides (CD34/a-SMA/CK19) and developed by a decision tree algorithm). As liver fibrosis progressed from F1 to F4, we observed that both AD and CPA gradually increases among the three etiologies (P < 0.05). However, at each stage of liver fibrosis, there was no significant difference in AD or CPA between CHB and PBC compared to MASLD (P > 0.05). AD and CPA performed similar diagnostic efficacy in liver cirrhosis (P > 0.05). Using the qVessel algorithm, we discovered a significant correlation between AD, CPA and METAVIR stages in all three etiologies. This suggests that AD could underpin a novel staging system.