Search

Food products are an essential source of human life, so they have a very important place, and it will be important to monitor and determine their quality in a short time. Our study deals with a very important and indispensable food product, which is the milk product, which is considered the main and important element in people’s lives, especially children, because it is the main source of their growth, building their bones, and strengthening. Because it is a highly perishable product, it must be monitored and its specifications must be monitored, because any gram of milk that is of low or poor quality may cause tons of milk to spoil, and also cause major financial losses. Therefore, a study was conducted to determine the quality of dairy product through machine learning algorithms (ML), which are support vector machine (SVM) algorithm, nearest neighbors (KNN) algorithm, decision tree (DT) algorithm and Bagging algorithm using milk dataset taken from data warehouse Kaggle. This data consists of 1059 samples and seven features. The proposed models were trained and tested with the aim of finding the best and most accurate model for detecting milk quality and were evaluated using the evaluation metrics: accuracy, precision, recall, f1_score and confusion matrix. According to the evaluation results three models: SVM, KNN, and DT outperformed Bagging algorithm, as they obtained the highest level for all metrics 100%. The SVM algorithm was the most efficient because its execution time was 0.146 seconds, which was less than the other models.

The rapid growth of Artificial Intelligence (AI) algorithms has created the opportunity to solve complex problems such as Internet of Things (IoT) botnet attacks. The severity of IoT botnet attacks is a critical challenge for improving the smart IoT environment. Therefore, there is an urgent need to design and implement an efficient detection model to deal with various IoT bot attacks and simultaneously handle issues related to the massive feature space. This paper introduces a wrapper feature selection technique by adapting the Artificial Rabbit Optimization (ARO) algorithm and the Decision Tree (DT) algorithm to detect various types of IoT botnet attacks. During the design of the suggested DT-ARO model, the N-BaIoT datasets were used as a testbed environment. The feature space optimization step was carried out using the ARO algorithm to select only the high-priority features for detecting the IoT botnet attacks. The binary vector technique was used to distinguish the optimal features. The detection engine was performed using the DT algorithm. The conducted experiments have demonstrated the ability of the suggested DT-ARO model to detect various types of IoT botnet attacks, where the accuracy rate was 99.89%. Meanwhile, effectively reducing the feature’s space. In addition, the accomplished results were compared with the latest typical approaches. The DT-ARO model was found to be competitive with these methods and even outperformed them in reducing the feature space. [ABSTRACT FROM AUTHOR]

Human Activity Recognition (HAR) based on Inertial Measurement Unit (IMU) has become increasingly important in health and fitness applications. These systems can continuously and cost-effectively monitor human activity, regardless of the surrounding environment. However, the current dominant trend in HAR uses black box flat classification (FC) methods, which lack interpretability and do not consider the natural hierarchical relationship between activity classes. Such systems often achieve greater accuracy at the cost of increased complexity and are not suitable for critical decision-making applications. This paper proposes an Adaptive Hierarchical Decision Tree (AHDT) HAR system that recognizes human activities based on IMU measurements along with quasi-stationary inclination feature extraction. The proposed system generates a global classifier that classifies human activities according to a tree taxonomy structure. This approach maintains interpretability while considering the fundamental signal data features embedded in the natural hierarchical representation of the activity classes. In addition to the commonly used flat classification accuracy measures, we applied modified hierarchical accuracy measures to assess the exclusive characteristics of hierarchical relationships between the classes. We used seven publicly available datasets to evaluate the proposed system and compared its performance with other tree-based classifiers, including Random Forest, Gradient Boosting, XGBoost, and AdaBoost classifiers. Our results demonstrated that the AHDT system significantly improves the recognition performance of fine-grained activities and offers a balance between lower complexity and higher interpretability. Overall, the proposed AHDT system provides an interpretable and practical approach to HAR that can be valuable in critical decision-making applications.

Speech is a powerful means to expressing thoughts, emotions, and perspectives. However, accurately determining the emotions conveyed through speech remains a challenging task. Existing manual methods for analyzing speech to recognize emotions are prone to errors, limiting our understanding and response to individuals’ emotional states. To address diverse accents, an automated system capable of real-time emotion prediction from human speech is needed. This paper introduces a speech emotion recognition (SER) system that leverages supervised learning techniques to tackle cross-accent diversity. Distinctively, the system extracts a comprehensive set of nine speech features—Zero Crossing Rate, Mel Spectrum, Pitch, Root Mean Square values, Mel Frequency Cepstral Coefficients, chroma-stft, and three spectral features (Centroid, Contrast, and Roll-off) for refined speech signal processing and recognition. Seven machine learning models are employed, encompassing Random Forest, Logistic Regression, Decision Tree, Support Vector Machines, Gaussian Naive Bayes, K-Nearest Neighbors, ensemble learning, and four individual, hybrid deep learning models including Long short-term memory (LSTM) and 1-Dimensional Convolutional Neural Network (1D-CNN) with stratified cross-validation. Audio samples from diverse English regions are combined to train the models. The performance evaluation results of conventional machine learning and deep learning models indicate that the Random Forest-based feature selection model achieves the highest accuracy of up to 76% among the conventional machine learning models. Simultaneously, the 1D-CNN model with stratified cross-validation reaches up to 99% accuracy. The proposed framework enhances the cross-accent emotion recognition accuracy up to 86.3%, 89.87%, 90.27%, and 84.96% by margins of 14.71%, 10.15%, 9.6%, and 16.52% respectively.

Researchers have focused their efforts on investigating the integration of crumb rubber as a substitute for conventional aggregates and cement in concrete. Nevertheless, the manufacture of crumb rubber concrete (CRC) has been linked to the release of noxious pollutants, hence presenting potential environmental hazards. Rather than developing novel CRC formulations, the primary objective of this work is to construct an extensive database by leveraging prior research efforts. The study places particular emphasis on two crucial concrete properties: compressive strength (fc') and tensile strength (fts). The database includes a total of 456 data points for fc' and 358 data points for fts, focusing on nine essential characteristics that have a substantial impact on both attributes. The research employs several machine learning algorithms, including both individual and ensemble methods, to undertake a comprehensive analysis of the created databases for fc' and fts. In order to ascertain the correctness of the models, a comparative analysis of machine learning techniques, namely decision tree (DT) and random forest (RF), is conducted using statistical evaluation. Cross-validation approaches are used in order to address the possible issues of overfitting. Furthermore, the Shapley additive explanations (SHAP) approach is used to investigate the influence of input parameters and their interrelationships. The findings demonstrate that the RF methodology has superior performance compared to other ensemble techniques, as shown by its lower error rates and higher coefficient of determination (R2) of 0.87 and 0.85 for fc' and fts respectively. When comparing ensemble approaches, it can be seen that AdaBoost outperforms bagging by 6 % for both outcome models and individual decision tree learners by 17% and 21% for fc' and fts respectively in terms of performance. The average accuracy of AdaBoost algorithm for both the models is 84%. Significantly, the age and the inclusion of crumb rubber in CRC are identified as the primary criteria that have a substantial influence on the mechanical properties of this particular kind of concrete.

This study innovates livestock health management, utilizing a top-view depth camera for accurate cow lameness detection, classification, and precise segmentation through integration with a 3D depth camera and deep learning, distinguishing it from 2D systems. It underscores the importance of early lameness detection in cattle and focuses on extracting depth data from the cow's body, with a specific emphasis on the back region's maximum value. Precise cow detection and tracking are achieved through the Detectron2 framework and Intersection Over Union (IOU) techniques. Across a three-day testing period, with observations conducted twice daily with varying cow populations (ranging from 56 to 64 cows per day), the study consistently achieves an impressive average detection accuracy of 99.94%. Tracking accuracy remains at 99.92% over the same observation period. Subsequently, the research extracts the cow's depth region using binary mask images derived from detection results and original depth images. Feature extraction generates a feature vector based on maximum height measurements from the cow's backbone area. This feature vector is utilized for classification, evaluating three classifiers: Random Forest (RF), K-Nearest Neighbor (KNN), and Decision Tree (DT). The study highlights the potential of top-view depth video cameras for accurate cow lameness detection and classification, with significant implications for livestock health management. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Education & Science is the property of Republic of Iraq Ministry of Higher Education & Scientific Research (MOHESR) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This study innovates livestock health management, utilizing a top-view depth camera for accurate cow lameness detection, classification, and precise segmentation through integration with a 3D depth camera and deep learning, distinguishing it from 2D systems. It underscores the importance of early lameness detection in cattle and focuses on extracting depth data from the cow’s body, with a specific emphasis on the back region’s maximum value. Precise cow detection and tracking are achieved through the Detectron2 framework and Intersection Over Union (IOU) techniques. Across a three-day testing period, with observations conducted twice daily with varying cow populations (ranging from 56 to 64 cows per day), the study consistently achieves an impressive average detection accuracy of 99.94%. Tracking accuracy remains at 99.92% over the same observation period. Subsequently, the research extracts the cow’s depth region using binary mask images derived from detection results and original depth images. Feature extraction generates a feature vector based on maximum height measurements from the cow’s backbone area. This feature vector is utilized for classification, evaluating three classifiers: Random Forest (RF), K-Nearest Neighbor (KNN), and Decision Tree (DT). The study highlights the potential of top-view depth video cameras for accurate cow lameness detection and classification, with significant implications for livestock health management.

A pipeline is the most efficient way to transport water from one place to another. Due to aging, corrosion, and external factors, the pipeline is prone to damage, which causes leaks. Many machine learning (ML) and deep learning (DL) methods are available to address this issue. This paper does an experimental study on available methods in ML and DL for leak detection for the collected data using an acousto-optic sensor. The experimental setup comprises of an acousto-optic sensor made of an erbium-doped fiber laser (EDFL), galvanized iron pipeline, a tank, a pump, and a data acquisition unit. The dimensions of the galvanized pipeline looped with the water tank are a length of 40 m, an inner diameter of 89 mm, and an outer diameter of 90 mm. The diameter of the simulated leak aperture is 5 mm. The methods analyzed in this study are k-means, k-medoids, Naive Bayes (NB), support vector machines (SVM), k-nearest neighbor (KNN), decision tree (DT), categorical boosting (CatBoost), random forest (RF), XGBoost, AdaBoost, and one-dimensional convolutional neural network (1DCNN). ML algorithms need a feature extraction technique because the data collected from the experiment is too large and contains redundant information. Feature extraction reduces the data size by extracting essential information. This paper extracts ten features from raw data. Among the ML algorithms, AdaBoost gives the highest prediction accuracy of 98.02%. This paper also implements eight models of 1DCNN, and Model 1 shows the best prediction accuracy of 98.16%, which is the highest compared with all the other classifiers in ML and DL for one-dimensional time series acousto-optic sensor data. [ABSTRACT FROM AUTHOR]

Chronic kidney disease (CKD) is a significant global health challenge that requires timely detection and accurate prognosis for effective treatment and management. The application of machine learning (ML) algorithms for CKD detection and prediction holds promising potential for improving patient outcomes. By incorporating key features which contribute to CKD, these algorithms enhance our ability to identify high-risk individuals and initiate timely interventions. This research highlights the importance of leveraging machine learning techniques to augment existing medical knowledge and improve the identification and management of kidney disease. In this paper, we explore the utilization of diverse ML algorithms, including gradient boost (GB), decision tree (DT), K-nearest neighbor (KNN), random forest (RF), histogram boost (HB), and XGBoost (XGB) to detect and predict chronic kidney disease (CKD). The aim is to improve early detection and prognosis, enhancing patient outcomes and reducing the burden on healthcare systems. We evaluated the performance of the ML algorithms using key metrics like accuracy, precision, recall, and F1 score. Additionally, we conducted feature significance analysis to identify the most influential characteristics in the detection and prediction of kidney disease. The dataset used for training and evaluation contained various clinical and demographic attributes of patients, including serum creatinine level, blood pressure, and age, among others. The proficiency analysis of the ML algorithms revealed consistent predictors across all models, with serum creatinine level, blood pressure, and age emerging as particularly effective in identifying individuals at risk of kidney disease. These findings align with established medical knowledge and emphasize the pivotal role of these attributes in early detection and prognosis. In conclusion, our study demonstrates the effectiveness of diverse machine learning algorithms in detecting and predicting kidney disease. The identification of influential predictors, such as serum creatinine level, blood pressure, and age, underscores their significance in early detection and prognosis. By leveraging machine learning techniques, we can enhance the accuracy and efficiency of kidney disease diagnosis and treatment, ultimately improving patient outcomes and healthcare system effectiveness. [ABSTRACT FROM AUTHOR]

Mangroves, vital salt-tolerant coastal forests, confer numerous societal benefits. However, mapping and monitoring these dense, coastal ecosystems is challenging due to limited field access. Leveraging recent hyperspectral remote sensing advancements, this study addresses this challenge. Hyperspectral imaging's narrow contiguous bands are potent for characterizing vegetation structure. The research centers on evaluating hyperspectral remote sensing potential, identifying suitable vegetation indices (VI) for the study area, and employing a multistage Decision Tree (DT) by stacking Vis at each stage to extract distinctive mangrove features. Mangrove attributes like unique leaf greenness, high canopy moisture, coastal-induced temperature moderation, and distinct shortwave infrared absorption contrast them from terrestrial vegetation. Various spectral Vegetation Indices (VI) capture these features by utilizing hyperspectral image bands across wavelengths. The proposed study employs Decision Tree (DT) classifier to optimize feature extraction for mangrove mapping by employing VI-based decision functions to discriminate mangroves from non-mangrove classes. Focused on the Marine National Park (MNP) in Jamnagar, Gulf of Kutch, the study exploits AVIRIS-NG hyperspectral data. It employs the Atmospherically Resistant Vegetation Index (ARVI) for vegetation separation, the Normalized Difference Infrared Index (NDII) for elevated canopy moisture analysis, and Shortwave Infrared Absorption Depth (SIAD) to assess unique shortwave infrared absorption in mangroves. Moreover, it integrates the Mangrove Vegetation Index (MVI) and Normalized Difference Mangrove Index (NDMI) for mangrove recognition. Stacking these indices in a DT remarkably boosts classification accuracy, yielding a peak overall accuracy (OA) of 94.71% and a kappa value (k) of 0.86. Published data and reports corroborate these results. This milestone furnishes preliminary insights into mangrove presence, vital for remote sensing-driven mapping and monitoring within challenging field-access environments. Furthermore, these findings pave the way for species-level discrimination via spectral matching and Machine Learning (ML) algorithms. [ABSTRACT FROM AUTHOR]

The health sector around the world faces the continuous challenge of improving the services provided to patients. Therefore, digital transformation in health services plays a key role in integrating new technologies such as artificial intelligence. However, the health system in Peru has not yet taken the big step towards digitising its services, currently ranking 71st according to the World Health Organisation (WHO). This article proposes a learning health system for the management and monitoring of private health services in Peru based on the three key components of intelligent health care: (1) a health data platform (HDP); (2) intelligent technologies (IT); and (3) an intelligent health care suite (HIS). The solution consists of four layers: (1) data source, (2) data warehousing, (3) data analytics, and (4) visualization. In layer 1, all data sources are selected to create a database. The proposed learning health system is built, and the data storage is executed through the extract, transform and load (ETL) process in layer 2. In layer 3, the Kaggle dataset and the decision tree (DT) and random forest (RF) algorithms are used to predict the diagnosis of disease, resulting in the RF algorithm having the best performance. Finally, in layer 4, the intelligent health-care suite dashboards and interfaces are designed. The proposed system was applied in a clinic focused on preventing chronic kidney disease. A total of 100 patients and six kidney health experts participated. The results proved that the diagnosis of chronic kidney disease by the learning health system had a low error rate in positive diagnoses (err = 1.12%). Additionally, it was demonstrated that experts were "satisfied" with the dashboards and interfaces of the intelligent health-care suite as well as the quality of the learning health system. [ABSTRACT FROM AUTHOR]

Regarding the pandemic taking place in the world from the spread of the Coronavirus pandemic and viral mutations, the need has arisen to analyze the epidemic data in terms of numbers of infected and deaths, different geographical regions, and the dynamics of the spread of the virus. In China, the total number of reported infections is 224,659 on June 11, 2022. In this paper, the Gaussian Mixture Model and the decision tree method were used to classify and predict new cases of the virus. Although we focus mainly on the Chinese case, the model is general and adapted to any context without loss of validity of the qualitative results. The Chi-Squared (χ2) Automatic Interaction Detection (CHAID) was applied in creating the decision tree structure, the data has been classified into five classes, according to the BIC criterion. The best mixture model is the E (Equal variance) with five components. The considered data sets of the world health organization (WHO) were used from January 5, 2020, to 12, November 2021. We provide numerical results based on the Chinese case.

This study focused on demonstrating how to design riding behavior-based collision avoidance systems (RB-CAWS). Statistical and machine learning algorithms were adopted and analyzed previously collected data from the riding simulator-based experiments. The experiments involved 23 participants who completed a 20 km course under four driving conditions, including high and low arousal and positive and negative valence emotional states. Our analysis started by implementing structural equation modeling (SEM) to examine the relationship between riding behaviors measured by speed variability and lateral instability and how these factors influence riding performance. Results revealed that lateral instability strongly affects riding performance. The average lateral deviation from the center of the lane (DCavg) and rolling entropy (RE) were clarified as more influential riding behaviors. Next, using the decision tree (DT) algorithm, significant thresholds for DCavg and RE were determined as 0.302m and 0.019, respectively. The neural network (NN) algorithm was also adopted and defined 0.5 as the optimal threshold value at a combined risk level. We compared the models based on their collision prediction accuracy and found that the DT model performed better with an accuracy rate of 80% compared to the NN model's accuracy rate of 76%. Using the identified thresholds, we highlighted potential warning systems and provided a framework for developing an RB-CAWS that integrates SEM and DT analysis techniques. This study can benefit researchers of motorcyclists' riding behavior and designers of advanced rider assistance systems (ARAS).

Engineered cementitious composite (ECC) is a special material that, when continuously hydrated, can considerably aid in self-healing. It is necessary to look at the capacity of ECC for self-healing – as measured by crack-widthafter (CWA) – when process of healing has completed, gauge the severity of the cracking, and foresee the extent of the healing. However, modeling and forecasting capacity of ECC for self-healing is a challenging task. Prediction of self-healing is a notably uncommon application of machine learning (ML), which has been applied to forecast a range of concrete properties. To estimate the capacity of ECC for self-healing, this study used three different the ensemble ML algorithms namely AdaBoost regressor (AR), decision tree (DT), and bagging regressor (BR). In addition, k-fold cross-validation method is utilized to assess the model effectiveness. With an R2 value of 0.974, the BR model was more successful in predicting outcomes when compared to the DT and AR models. Improved model performance was shown for ensemble models with smaller MAE (AR = 3.40, and BR = 1.89), MSE (AR = 27.09, and BR = 10.40), and RMSE (AR = 5.21, and BR = 3.23) values and larger R2 (AR = 0.933, and BR = 0.974) values as compared to DT (MAE = 4.29, MSE = 43.28, RMSE = 6.58, R2 = 0.894). Eventually this study will lead to savings in time, effort, and money, and the use of ML approaches to predict CWA will advance the field of civil engineering. It is also advised to investigate the crack-healing properties of additional cementitious materials such rice husk ash, wheat straw ash, and pumice powder subject to modeling their crack-healing properties using ML approaches.

The world has seen the advent of numerous illnesses that cannot be medically recognized, such as Autism Spectrum Disorder (ASD). It affects several behavioral domains, including social and linguistic competence and stereotyped and repetitive actions. This illness is a serious neurodevelopmental disorder. Since many other mental illnesses have strikingly similar symptoms to those of ASD, diagnosing ASD can be difficult and time-consuming. Early diagnosis based on different health and physiological characteristics seems feasible with the rising usage of machine learning-based models in predicting many human diseases. This study aims to create a classification model that can predict the likelihood of ASD with the greatest degree of precision. To investigate the potential for predicting and analyzing ASD traits in the Toddler, Child, Adolescent, and adult age groups, we used several supervised Machine Learning (ML) models. These include Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbor (K-NN), Nave Bayes (NB), Logistic Regression (LR), and Random Forest (RF). Four publicly available, distinctive non-clinical ASD screening datasets from Kaggle and the UCI machine learning library are used to test these models. The first dataset includes 1054 instances and 19 toddler-related features. The remaining ones consist of 21 traits and 292, 104, and 704 cases involving children, adolescents, and adults, respectively. After implementing different ML approaches over the pre-processing datasets, the results showed that the DT, LR, and RF classifiers are the dominant models. These dominated models achieve the highest prediction accuracy, among other studied models, of about 100% for all the utilized datasets.

Distributed denial of service (DDoS) attacks are one of the most common global challenges faced by service providers on the web. It leads to network disturbances, interruption of communication and significant damage to services. Researchers seek to develop intelligent algorithms to detect and prevent DDoS attacks. The present study proposes an efficient DDoS attack detection model. This model relies mainly on dimensionality reduction and machine learning algorithms. The principal component analysis (PCA) and the linear discriminant analysis (LDA) techniques perform the dimensionality reduction in individual and hybrid modes to process and improve the data. Subsequently, DDoS attack detection is performed based on random forest (RF) and decision tree (DT) algorithms. The model is implemented and tested on the CICDDoS2019 dataset using different data dimensionality reduction test scenarios. The results show that using dimensionality reduction techniques along with the ML algorithms with a dataset containing high-dimensional data significantly improves the classification results. The best accuracy result of 99.97% is obtained when the model operates in a hybrid mode based on a combination of PCA, LDA and RF algorithms, and the data reduction parameter equals 40.

Vertebrate fossils/remains became recently significant in various study fields for determining the ecological biodiversity. However, with the great abundance of fossils/remains and their classes, there is a difficulty in identifying and detecting these classes. Hence, in this paper, an accurate machine learning classification technique is presented to differentiate automatically some types of 3D vertebrate remain images. A computed tomography (CT) scanner is utilized to construct a dataset of 3D images of some vertebrate remains found in Egypt. Adaptive enhancement and segmentation processes are applied to the dataset. The different selected geometric features are then extracted. Thus, the extracted features are classified using suitable machine learning classifiers (SVM, KNN, DTs). The automatic detection for the remains class, according to the extracted features, is obtained using the confusion matrix for the training and testing data points and the receiver operating characteristic (ROC) curve. The results confirmed an accurate technique with high performance.