Decision Tree Algorithm Research Articles

Ultimate load bearing of helical piles prediction and evaluation using machine learning-based algorithms

ABSTRACT This study aims to predict the Ultimate Load-Bearing (ULB) capacity of Helical Piles (HP) using six Machine Learning Algorithms (MLA) on an in situ-based 110 pile load tests including a wide range of pile properties: shaft diameters (73–406 mm), helix diameters (254–762 mm), helix spacing (300–1000 mm), number of helices (1–6), pile lengths (2.4–16 m), and helix thicknesses (6–12 mm). The measured axial ULB is analysed using the Brinch-Hansen 80% criterion and 5% criterion of the average diameter of the helices. Load-displacement curves were fitted using the Hyperbolic Function. MLA including Multi Linear Regression (MLR), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest, eXtreme Gradient Boosting, and Support Vector Regression were optimised to grid search for hyperparameters like neighbour count, tree depth, learning rate, and kernel type. Input parameters were categorised into Geometric and Soil Properties packages. Results indicate that the DT algorithm excelled in pullout loading, KNN in compression loading, and MLR for Brinch-Hansen 80% criterion estimations. The input parameters related to the soil surrounding the pile helices have the most impact on the ULB prediction of HP. This study enhances HP foundation design by enabling data-driven decisions for optimal pile selection and configuration.

Risk prediction of computer investment database information management system based on machine learning algorithms

In recent years, with the continuous development of the financial market, the risk prediction of computer investment database information management systems (IMS) has high practical value. At present, there are risk issues in the information management system, which may cause drawbacks to investment data processing. To address these issues, this article used Machine Learning (ML) algorithms to analyze the risk prediction of computer investment database IMS. This article introduced and utilized typical Self-Organizing Map (SOM) and Artificial Neural Network (ANN) combination algorithms, regression algorithms, and Gradient Boosting Decision Tree (GBDT) algorithms to compare and analyze the prediction accuracy of these three algorithms. This article found that the GBDT algorithm has the highest prediction accuracy. Through a large amount of experimental data, it has been proven that the average testing accuracy using regression algorithms was 3.5% higher than that using neural network algorithms. It was found that the average test accuracy using the GBDT algorithm was 7.2% higher than the average test accuracy using the regression algorithm. The study also explores the combination of physiological and behavioral data collected by wearable devices to provide more comprehensive risk assessment and decision support, which provides an important reference for the optimization of enterprise risk management. Through this innovative data source integration, this paper provides a new perspective for the application and development of machine learning algorithms in computer investment database IMS.

Human Perception Based on Textual Analysis

The complex process by which humans use their senses to clarify and understand the world around them is referred to as human perception. Analyzing human perception is important for comprehension of how humans think, feel, and act, which is helpful in a variety of contexts and ultimately promotes improved understanding, communication, and engagement. This study examines the field of text mining-based human perception analysis using a precisely chosen dataset of Twitter customer service discussions. Decision Trees, KNN, Naive Bayes, and GLM are four different algorithms that are methodically examined to determine which is the most effective method for understanding and predicting human perception from textual data. After an exhaustive analysis, the Decision Tree algorithm is shown to be the best performer, closely followed by Naive Bayes. The human perception analysis of text mining, including the methodology, findings, and implications, is described in depth.

Mining Hypertension Predictors using Decision Tree: Baseline Data of Kharameh Cohort Study

Introduction: Hypertension is a serious chronic disease and an important risk factor for many health problems. This study aimed to investigate the factors associated with hypertension using a decision-tree algorithm. Methods: Methods: This cross-sectional study was conducted through the census in Kharameh City between 2014 and 2017. The study included 2510 hypertensive and 7840 non-hypertensive individuals. To create the decision tree, 70% of the cases were randomly allocated to the training dataset. In comparison, the remaining 30% were used as the testing dataset for the performance evaluation of the decision tree. Two models were assessed. In the first model (model I), 15 variables including age, gender, body mass index (BMI), years of education, occupation status, marital status, family history of hypertension, physical activity, total energy, number of meals, salt, oil type, drug use, alcohol use, and smoke entered into the model. in the second model (model II) 16 variables including age, gender, BMI and Blood factors as Hematocrit (HCT), Mean corpuscular hemoglobin concentration (MCHC), Platelet Count (PLT), Fasting blood sugar (FBS), Blood Urea Nitrogen (BUN), creatinine (Cr), triglycerides (TG), cholesterol (CHOL), Alkaline phosphatase (ALP), High-density lipoprotein (HDL), Gamma-glutamyl transpeptidase (GGT), low-density lipoproteins (LDL) and Urinary specific gravity (SG) were considered. A confusion matrix was used to measure the performance of the decision tree. Additionally, accuracy, sensitivity, specificity, and the receiver operating characteristics (ROC) curve were determined to compare the models. Results: For the model I, the accuracy, sensitivity, specificity and AUC value were 79.2%(77.8-80.6), 82.4%(80.1-84.5), 78.24%(76.4-80), and 0.80%(0.79-0.82), respectively. For model II, the corresponding values were 79.5%(78.2-80.8), 81.0%(78.3-83.6)79.0%(77.5-80.5)and 0.80%(0.79-0.81), respectively. Confusion matrix of model I showed that of the 1188 cases with hypertension in the training data set, 979 cases were classified correctly and, for model II of the 2812 non-hypertension cases, 2222 cases were classified correctly. Conclusion: We have suggested a decision tree model to identify the risk factors associated with hypertension. This model can be useful for early screening and improving preventive and curative health services in health promotion

A Decision Tree-Driven IoT systems for improved pre-natal diagnostic accuracy

Prenatal diagnostics are vital for the woman as well as her unborn baby. The diagnostics help in the early identification of the possibility of complication and the initial measures that help to ameliorate the mother and the fetus health status are taken. Over the year’s various techniques have been employed in diagnosing genetic disorders before birth that lack effectiveness in terms of cost, time, and places to access ultra-modern health facilities. To overcome these problems, this paper puts forward a diagnostic model that integrates Internet of Things innovation with a Machine Learning approach which is the Decision Tree Algorithms. First, it implies the application of IOT devices in the collection of vital information like heart rate, blood pressure, glucose levels, and fetal movement. The data is structured in the form of a dataset and transmitted to a Big Data storage for warehousing and processing. Secondly, the DTA is employed to analyze the data and look for patterns and possibilities of future health complications. The DTA operates in that it divides the dataset into subsets considering specific features and formulates a tree-like model of decisions. At every node, the algorithm chooses the attribute which has the highest information gain, to partition the data into different classes. This process goes on until it reaches a decision node through which, it can decide probable health problems from the input data. To increase the reliability of the developed model this study fine-tunes the model by using a large database of pre-natal health records. The system is capable of collecting data in real-time and flagging data that needs attention in the case of any abnormality to the health professional. The above methodology was tested on a 1000-record database of pre-natal health records where the proposal achieved 95% possibility of potential health problems as against 85% by classical statistical analysis. Furthermore, the system scaled down the number of false positive cases by 20 percent and false negatives by 15 percent thus the efficacy of the system.

Assessing the risk of high-grade squamous intraepithelial lesions (HSIL+) in women with LSIL biopsies: a machine learning-based study

ObjectiveThis study aims to analyze factors associated with the missed diagnosis of high-grade squamous intraepithelial lesions (HSIL+) in patients initially diagnosed with low-grade squamous intraepithelial lesions (LSIL) through colposcopic biopsy and to develop a predictive model for assessing the risk of missed HSIL+.MethodsWe conducted a retrospective analysis of 505 patients who underwent loop electrical excision procedure (LEEP) following an LSIL diagnosis by colposcopic biopsy. Logistic regression was used to identify demographic and pathological parameters associated with missed diagnoses of HSIL+. Additionally, several machine learning methods were employed to construct and assess the performance of the risk prediction models.ResultsThe overall rate of missed diagnoses for HSIL+ was 15.2%. Independent risk factors identified were HPV16/18 infection (OR 2.071; 95% CI 1.039–4.127; p = 0.039), TCT ≥ ASC-H (OR 4.147; 95% CI 1.392–12.355; p = 0.011), TZ3 (OR 1.966; 95% CI 1.003–3.853; p = 0.049) and Colposcopic impression G2 (OR 3.627; 95% CI 1.350–9.743; p = 0.011). Among the models tested, the Decision Tree algorithm demonstrated superior performance with an accuracy of 94.7%, sensitivity of 80.0%, specificity of 96.9%, and an area under the curve (AUC) of 0.936 in the validation set.ConclusionKey independent risk factors for the missed diagnosis of HSIL in patients with LSIL include HPV16/18 infection, TCT ≥ ASC-H, TZ3, and colposcopic impression G2. The Decision Tree model offers a cost-effective, reliable, and clinically valuable tool for accurately predicting the risk of missed diagnosis of HSIL+, facilitating early intervention and management.

Mastering Object Healing Techniques for UI Automation: A Comprehensive Guide

Object healing plays a crucial role in the realm of UI tests, serving as a fundamental mechanism that enhances the resilience and stability of automated testing frameworks. As applications undergo constant changes, whether through updates in user interfaces or the addition of new features, the elements within those interfaces often change in ways that can disrupt automated tests. Object healing techniques address these disruptions by allowing test scripts to adapt to changes in UI elements, ensuring that tests remain valid and effective. This adaptability is particularly significant in dynamic environments, where the cost of maintaining outdated tests can be substantial. This paper provides a comprehensive overview of the importance of object healing in UI tests. It explores the fundamentals of object recognition and examines various techniques for effective object recognition. Additionally, the paper delves into self-healing techniques using Healenium, a tool specifically designed to support UI applications built on React. Furthermore, it discusses the optimization of machine learning algorithms to enhance object healing in UI tests, and outlines best practices for object healing in continuous integration environments. The paper concludes with recommendations and insights into future trends in this rapidly evolving field. Keywords: UI automation, Element recognization, self-healing techniques, Healenium, Object healing, decision tree algorithm

Framework Construction and Dynamic Characteristics of Spring Low-Temperature Disasters Affecting Winter Wheat in the Huang-Huai-Hai Region, China

The accurate and sub-daily identification of agricultural low-temperature disasters (LTDs) facilitates the understanding of their dynamic evolution, the evaluation of the characteristics of disaster events, and informs effective strategies aimed at disaster prevention and mitigation. In order to ensure the timely, precise, and comprehensive capture of disaster processes, we have developed a dynamic evaluation framework for winter wheat spring LTD in the Huang-Huai-Hai (HHH) region, driven by meteorological data. This framework consists of two primary components: a disaster classification module and a dynamic simulation-assessment module. Through disaster mechanisms and comprehensive statistical analysis, we have established the input features and structural framework of the classification module using a decision tree algorithm. The dynamic simulation evaluation module is based on our newly developed index for the cumulative hourly intensity of low-temperature stress (CHI) and its grade indicators. This index integrates the interaction between cold stress (low-temperature intensity, cooling amplitude, and duration) and mitigating conditions (air humidity) during the evolution process of LTD. Based on CHI, we found that as the intensity of low temperatures and the amplitude of cooling rise, along with an extended duration of stress and a reduction in relative humidity, the severity of spring LTDs in winter wheat get worse. The overall validation accuracy of the evaluation framework is 92.6%. High validation accuracy indicates that our newly established framework demonstrates significant efficacy in identifying LTDs and assessing grade. Through the analysis of the characteristics of the disaster process, spring LTDs affecting winter wheat are mainly mild, with frost identified as the primary category of LTD. The duration of freeze injury typically exceeds 24 h, while the duration of frost damage and cold damage is less than 24 h. From 1980 to 2022 in the HHH region, the frequency of spring freeze injury and frost damage on winter wheat showed an overall decreasing trend, with a particularly significant decrease in frost damage occurrences. Conversely, cold damage occurrences are on the rise. In addition, the duration of individual disaster events for the three categories of spring LTDs is decreasing, while both the average intensity and extremity of these events show increasing trends. This study has important practical value for the sub-daily scale evaluation of the spring LTD affecting winter wheat in the HHH region and serves as an effective guide for agricultural disaster prevention and mitigation, as well as for the formulation of planting strategies.

A fact based analysis of decision trees for improving reliability in cloud computing

The popularity of cloud computing (CC) has increased significantly in recent years due to its cost-effectiveness and simplified resource allocation. Owing to the exponential rise of cloud computing in the past decade, many corporations and businesses have moved to the cloud to ensure accessibility, scalability, and transparency. The proposed research involves comparing the accuracy and fault prediction of five machine learning algorithms: AdaBoostM1, Bagging, Decision Tree (J48), Deep Learning (Dl4jMLP), and Naive Bayes Tree (NB Tree). The results from secondary data analysis indicate that the Central Processing Unit CPU-Mem Multi classifier has the highest accuracy percentage and the least amount of fault prediction. This holds for the Decision Tree (J48) classifier with an accuracy rate of 89.71% for 80/20, 90.28% for 70/30, and 92.82% for 10-fold cross-validation. Additionally, the Hard Disk Drive HDD-Mono classifier has an accuracy rate of 90.35% for 80/20, 92.35% for 70/30, and 90.49% for 10-fold cross-validation. The AdaBoostM1 classifier was found to have the highest accuracy percentage and the least amount of fault prediction for the HDD Multi classifier with an accuracy rate of 93.63% for 80/20, 90.09% for 70/30, and 88.92% for 10-fold cross-validation. Finally, the CPU-Mem Mono classifier has an accuracy rate of 77.87% for 80/20, 77.01% for 70/30, and 77.06% for 10-fold cross-validation. Based on the primary data results, the Naive Bayes Tree (NB Tree) classifier is found to have the highest accuracy rate with less fault prediction of 97.05% for 80/20, 96.09% for 70/30, and 96.78% for 10 folds cross-validation. However, the algorithm complexity is not good, taking 1.01 seconds. On the other hand, the Decision Tree (J48) has the second-highest accuracy rate of 96.78%, 95.95%, and 96.78% for 80/20, 70/30, and 10-fold cross-validation, respectively. J48 also has less fault prediction but with a good algorithm complexity of 0.11 seconds. The difference in accuracy and less fault prediction between NB Tree and J48 is only 0.9%, but the difference in time complexity is 9 seconds. Based on the results, we have decided to make modifications to the Decision Tree (J48) algorithm. This method has been proposed as it offers the highest accuracy and less fault prediction errors, with 97.05% accuracy for the 80/20 split, 96.42% for the 70/30 split, and 97.07% for the 10-fold cross-validation.

Regulation of the physicochemical properties of nutrient solution in hydroponic system based on the CatBoost model

Too high or low nutrient solution concentrations adversely affect hydroponic vegetables’ growth and yield performance. In addition, the temperature of the solution and the dissolved oxygen (DO) content of the water are critical factors. A real-time regulation model for nutrient solutions based on the CatBoost model was developed to address the issues of low automation and accuracy of nutrient solution configuration and management. The pH, electric conductivity (EC) and DO of the nutrient solution, and the corresponding water temperature of 96 groups of hydroponic lettuce nutrient solutions (KNO3, MgSO4, Ca(NO3)2, KH2PO4, NH4H2PO4, and microfertilizer) were measured using a gradient combination test designed according to the formula of hydroponic lettuce nutrient solution. A prediction model of the nutrient solution index was constructed using a gradient-enhanced decision tree algorithm and the accuracies of the random forest, XGBoost, and CatBoost algorithms were compared. The results show that the root mean square error (RMSE), mean absolute error (MAE) and coefficient of determination (R2) of the simulated and measured values of the EC prediction model established by the CatBoost model were 248.28 µS/cm, 203.19 µS/cm and 0.946, respectively. At the same time, it was found that the hydroponic variable fertilization control system, composed of four modules–data acquisition, data decision, field control, and manual control–could predict the EC and pH value changes of the nutrient solution and timely add compounds for control. The application of nutrients based on the developed system was good and the weight of the cream lettuce (Flandria RZ) grown using the system was more than 200 g, which meets the requirement for high-quality cream lettuce production. Therefore, the CatBoost model system can be used to improve the hydroponic production system of vegetables by managing nutrients effectively.

A preordonance-based decision tree method and its parallel implementation in the framework of Map-Reduce

In supervised classification, decision trees are one of the most popular learning algorithms that are employed in many practical applications because of their simplicity, adaptability, and other perks. The development of effective and efficient decision trees remains a major focus in machine learning. Therefore, the scientific literature provides various node splitting measurements that can be utilized to produce different decision trees, including Information Gain, Gain Ratio, Average Gain, and Gini Index. This research paper presents a new node splitting metric that is based on preordonance theory. The primary benefit of the new split criterion is its ability to deal with categorical or numerical attributes directly without discretization. Consequently, the Preordonance-based decision tree” (P-Tree) approach, a powerful technique that generates decision trees using the suggested node splitting measure, is developed. Both multiclass classification problems and imbalanced data sets can be handled by the P-Tree decision tree strategy. Moreover, the over-partitioning problem is addressed by the P-Tree methodology, which introduces a threshold ϵ as a stopping condition. If the percentage of instances in a node falls below the predetermined threshold, the expansion of the tree will be halted. The performance of the P-Tree procedure is evaluated on fourteen benchmark data sets with different sizes and contrasted with that of five already existing decision tree methods using a variety of evaluation metrics. The results of the experiments demonstrate that the P-Tree model performs admirably across all of the tested data sets and that it is comparable to the other five decision tree algorithms overall. On the other hand, an ensemble technique called “ensemble P-Tree” offers a reliable remedy to mitigate the instability that is frequently associated with tree-based algorithms. This ensemble method leverages the strengths of the P-Tree approach to enhance predictive performance through collective decision-making. The ensemble P-Tree strategy is comprehensively evaluated by comparing its performance to that of two top-performing ensemble decision tree methodologies. The experimental findings highlight its exceptional performance and competitiveness against other decision tree procedures. Despite the excellent performance of the P-Tree approach, there are still some obstacles that prevent it from handling larger data sets, such as memory restrictions, time complexity, or data complexity. However, parallel computing is effective in resolving this kind of problem. Hence, the MR-P-Tree decision tree technique, a parallel implementation of the P-Tree strategy in the Map-Reduce framework, is further designed. The three parallel procedures MR-SA-S, MR-SP-S, and MR-S-DS for choosing the optimal splitting attributes, choosing the optimal splitting points, and dividing the training data set in parallel, respectively, are the primary basis of the MR-P-Tree methodology. Furthermore, several experimental studies are carried out on ten additional data sets to illustrate the viability of the MR-P-Tree technique and its strong parallel performance.

Unveiling DNA Sequences: A Comparison of Machine Learning and Deep Learning Techniques for Prediction

<div align="center"><table width="590" border="1" cellspacing="0" cellpadding="0"><tbody><tr><td valign="top" width="387"><p>DNA is the biological macromolecule unit that carries the information of all protein, amino acid sequences. With the help of this protein sequence, we explore the mutated gene and disease-causing mutated genomic pattern. Currently, the progression of genomic innovation is the source of DNA arrangement information developing at a dangerous rate—external factors have stimulated the volume of research into DNA genomes. Initially, the development process of DNA sequencing is accomplished with the support of the Database, data structures, and sequence similarity. The method is capable of extracting a particular property in DNA. We employ the deep learning algorithm to pull out protein sequences' features. The DNA sequence is classified based on the in-build protein structures extracted into the Fasta file. Therefore, the DNA sequence of E.coli with 106 data sets and 57 nucleotides is tested experimentally. Finally, we compared the results with the existing decision tree algorithm, KNN- Classification, random forest, and neural networks. The deep learning algorithm yields higher efficiency of 98% compared to other machine learning algorithms. This highlights the potential of deep learning in genomics research and its ability to yield superior results in classifying DNA sequences.</p></td></tr></tbody></table></div>

Breast cancer identification using machine learning and hyperparameter optimization

Breast cancer identification can be analyzed through genomic analysis using gene expression data, one type of which is mRNA. This involves analyzing gene expression patterns of breast tissue samples to distinguish breast cancer from healthy tissue or to differentiate subtypes of different breast cancers. This research developed the right computational model for breast cancer classification using machine learning and hyperparameter optimization algorithms. The primary objective of this research is to utilize various machine learning algorithms to classify breast cancer based on gene expression and enhance the models developed in previous studies. This paper provides an extensive literature review of prior breast cancer classification research and offers new theoretical perspectives. This research used a problem-solving approach with conventional machine learning techniques, most notably the decision tree. It also evaluates other machine learning algorithms for comparison, including k-nearest neighbor, naïve bayes, random forest, extra tree classifier, and support vector machine. The evaluation process used classification reports that provide insight into the precision, recall, F1-score, and accuracy of each machine learning model. The evaluation results show that the performance of the decision tree algorithm model is superior and impressive, achieving 99.73% accuracy and a score of 1 for precision, recall, and F1-score.

Mechanism Isomorphism Identification Based on Decision Tree Algorithm and Hybrid Particle Swarm Optimization Algorithm

Mechanism isomorphism identification is a typical quadratic assignment problem similar to traveling salesman and job-shop scheduling. For the complex mechanism with more components, common methods of isomorphism identification may fail due to low solving efficiency and reliability. Based on the decision tree algorithm and hybrid particle swarm optimization (HPSO) algorithm, the global-local search method is proposed to identify isomorphism of mechanisms. More precisely, based on the intrinsic relationship between links and vertices in the mechanism, the decision tree algorithm globally searches the characteristic path with mapping properties of different mechanisms. On this basis, HPSO algorithm combines genetic algorithm with particle swarm optimization algorithm to find the exact global optimal solution instead of local optimal solution. Some complex cases such as 14-link kinematic chains, 18-vertex topological graphs, and 8-vertex planetary gear trains are used to evaluate the efficiency and reliability of the proposed method. Results show that the proposed method can accurately identify isomorphism of mechanisms in a relatively short time. It can improve the solving efficiency of isomorphism identification in structural synthesis.

Classification of Skin Diseases using Decision Tree Algorithm on an Imbalanced Dataset

Skin infections caused by pathogens such as bacteria and fungi are common and can lead to serious health complications if not properly managed. Accurate classification of these infections is crucial for effective treatment and management. This study focuses on classifying two skin diseases, Chickenpox and Shingles, using a Decision Tree algorithm applied to an imbalanced dataset sourced from Kaggle. The dataset, which is imbalanced by nature, was split into training (80%) and testing (20%) subsets. Pre-processing involved segmentation using Thresholding to isolate regions of interest and feature extraction using Hu Moments to capture shape characteristics of the lesions. The dataset was scaled to ensure that all features had a mean of 0 and variance of 1. The classifier's performance was evaluated using 5-fold cross-validation, yielding a mean accuracy of 66.06%, with precision, recall, and F1-scores indicating moderate performance. The study highlights the challenges posed by imbalanced datasets and the limitations of the Decision Tree algorithm in this context. The results underscore the importance of proper pre-processing and feature extraction but also suggest the need for more advanced classification techniques and data balancing methods. This research contributes to the field by providing a detailed methodology and comprehensive evaluation metrics, offering insights into the application of machine learning for medical image classification. Future work should focus on improving classifier performance through data augmentation, advanced feature extraction, and exploring other machine learning models better suited for imbalanced datasets.

Empowering Communities Through Resilient River Management

This paper focuses on empowering the society with various rescue strategies in the event of a natural disaster namely floods. This paper deals with applying the state of the art technologies such as machine learning in order to combat floods. The key aspects of a flood management system and the existing systems were then identified. The proposed system was then developed. Statistical, hydrological and Machine learning algorithms such as Random forest, support vector machines, Naïve Bayes, decision tree algorithms were further applied on the data parameters captured. to predict the overflow probability in different regions of the state. A comparison was then undertaken to determine the most suited algorithm to perform better river management. The paper then concluded highlighting the need for applying machine learning techniques for flood management.

Research on parameter regionalization based on decision tree algorithm: a case study of 16 catchments in northeast China

ABSTRACT In order to improve the accuracy of hydrological forecasting in ungauged catchments, this study explores a new machine learning (ML) approach and applies it to the regionalization of hydrological model parameters in ungauged catchments. Taking 16 catchments located in Northeast China as the research objects, based on the Spatiotemporal Variable Source-mixed Runoff (SVSMR) Model, the splitting conditions of Classification and Regression Tree (CART) are used to quantify the rules for parameter transplantation of the model, and suitable donor basins are identified for the target basin. The results show that the SVSMR model has good applicability in the research area, with an average NSE (Nash-Sutcliffe efficiency coefficient) of 0.85 for the 16 basins and an average EQP (error of peak flow) of 5.98%. The criteria for similar watershed discrimination formulated using CART can improve the effect of hydrological model parameter transplantation, with suitable donor watersheds found for 87.5% of the target watersheds.

Evaluation and prediction of the physical properties and quality of Jatobá-do-Cerrado seeds processed and stored in different conditions using machine learning models

The conservation of seed quality throughout storage depends on established conditions, monitoring, sampling and laboratory analysis, which are subject to errors and require technical and financial resources. Thus, machine learning techniques can help optimize processes and obtain more accurate results for decision-making regarding the processing and conservation of stored seeds. Therefore, the aim was to assess and predict the physical properties (moisture content, seed mass, length, thickness, width, volume, apparent specific mass, projected area, sphericity, mean diameter, circular area, circularity, drag coefficient), and physicochemical quality (crude protein, ash content, and acidity index) of Jatobá-do-Cerrado seeds under different processing conditions with pulp, without pulp (scarification), without pulp (fermented), and storage conditions at 10 and 23 °C over six months. Data were analyzed on Weka software (Waikato Environment for Knowledge Analysis) version 3.9.5. testing the following models: Pearson correlation, Artificial Neural Networks, decision tree algorithms RepTree and M5P, Random Forest, and Linear Regression. Processing cerrado jatobá seeds by fermentation and storage at 10 °C minimized physical changes and preserved the physicochemical quality of the seeds in polyethylene plastic, glass container, tetrapack, and polyethylene container, over six months. The combination of processing, temperature, and packaging variables for Artificial Neural Networks, RepTree, Random Forest, and M5P algorithms outperformed linear regression, providing higher accuracy rates. Artificial Neural Network and Random Forest models were the best predicting the effects of treatments on changes in physical properties and physicochemical quality of jatobá seed.

Lightweight Machine Learning-Based DOS Attack Detection in Wireless Sensor Networks Using Decision Tree and Information Gain

Abstract—Wireless Sensor Networks (WSNs) are rapidly expanding across various domains due to their unique characteristics and performance capabilities. However, these networks are highly vulnerable to a range of security threats, particularly Denial-of-Service (DoS) attacks, which are among the most common in WSNs. This paper explores the vulnerabilities of WSNs, focusing on DoS threats, and reviews current techniques for their detection. It introduces a lightweight machine learning-based approach using a decision tree (DT) algorithm with the Information Gain (IG) feature selection method for efficient DoS detection. Tested on an enhanced WSN-DS dataset, the proposed method demonstrated high accuracy and minimal processing time compared to other classifiers, such as XGBoost, and RF. This efficiency makes the proposed method well-suited for real- time DoS attack detection in resource-constrained WSNs. Keywords—Machine Learning,Decision Tree (DT),Information Gain (IG),DoS attack detection

Distinguishing among standing postures with machine learning-based classification algorithms.

The purpose of our study was to evaluate the accuracy with which classification algorithms could distinguish among standing postures based on center-of-pressure (CoP) trajectories. We performed a secondary analysis of published data from three studies: Study A) assessment of balance control on firm or foam surfaces with eyes-open or closed, Study B) quantification of postural sway in forward-backward and side-to-side directions during four standing-balance tasks that differed in difficulty, and Study C) an evaluation of the impact of two modes of transcutaneous electrical nerve stimulation on balance control in older adults. Three classification algorithms (decision tree, random forest, and k-nearest neighbor) were used to classify standing postures based on the extracted features from CoP trajectories in both the time and time-frequency domains. Such classifications enable the identification of differences and similarities in control strategy. Our results, especially those involving time-frequency features, demonstrated that distinct CoP trajectories could be identified from the extracted features in all conditions and postures in each study. Although the overall classification accuracy was similar using time-frequency features (~ 86%) for the three studies, there were substantial differences in accuracy across conditions and postures in Studies A and B but not in Study C. Nonetheless, the models were far superior to the published results with conventional metrics in distinguishing between the conditions and postures. Moreover, a Shapley Additive exPlanation analysis was able to identify the most important features that contributed to the classification performance of the models.