Al-Iraqia Journal for Scientific Engineering Research

This study provides a new approach for identifying water distribution ‎system leaks. It blends vibration signal processing with machine learning. The ‎system is based on vibration signals from accelerometers for pipeline observation ‎through non-invasive methods and real-time. Based on a Random Forest classifier, the ‎system is able to differentiate between different leak scenarios from no-leak cases with an accuracy of 97.3%. We validated the findings using a confusion matrix, which confirmed some cases of misclassification, indicating there is still much scope for improvement. We identified key statistical features such as RMS, kurtosis, and variance as being of prime ‎importance for leak identification using feature importance analysis. These features enable capturing the specific vibration patterns of diverse leaks, ‎allowing for accurate identification. This is an improvement over conventional ‎leak detection techniques, offering a more reliable and efficient method for ‎pipeline observation. The study also discusses how the procedure could make ‎water distribution systems sustainable and operationally efficient for ‎application in the real world.‎

Early and accurate detection of liver disease is critical to improving patient outcomes yet remains challenging due to class imbalance and noisy clinical data. In this study, we present a robust ensemble learning framework applied to the Indian Liver Patient Dataset, incorporating systematic data cleaning, normalization, and Synthetic Minority Over‑Sampling (SMOTE) to address missing values, outliers, and class skew. We then perform correlation-based feature reduction before training a stacking classifier that combines Random Forest, XGBoost, and ExtraTrees base learners with an ExtraTrees meta‑learner. Using stratified 10‑fold cross‑validation on the balanced cohort (n = 792), our ensemble achieves 91.6 % accuracy, 92 % F1‑score, and a high area under the ROC curve, outperforming individual models and prior published approaches. These results demonstrate the potential of heterogeneous ensembles for clinical decision support in hepatology and lay the groundwork for prospective validation in diverse patient populations.

Blood sugar issues are a major health issue worldwide, with their incidence growing rapidly and affecting human health, economic systems, and societal structures. If diabetes remains untreated and undiagnosed, it can cause blood sugar levels to vary significantly, potentially damaging essential organs like the kidneys, eyes, and arteries of the heart in critical cases. As a result, there is an increasing focus on the prevention and early detection of diabetes mellitus within the medical community. Utilizing machine learning algorithms to analyze appropriate datasets for early disease prediction could prove life-saving. The objective of this paper is to examine four algorithms that are proposed to enhance the diagnosis of diabetes. This research analyzes the effectiveness of various machine learning algorithms in processing datasets with minority classes. The evaluation was based on the classification report (including accuracy, precision, recall, and F1-score), the confusion matrix, and the ROC AUC. The Diabetes Prediction Dataset is used to evaluate four machine learning algorithms. The classifier that deserves a singular mention is the Artificial Neural Network (ANN), which achieves a 97% accuracy rate. This demonstrates its capability of classifying instances that are common and less common types. The Random Forest and Decision Tree models also perform well in terms of their ability to deliver strong performance, and the outcome shows some incremental differences, suggesting their ability to manage the dataset is quite high. However, the Support Vector Machine (SVM) model performs worse than all the above models at 96.36% and seems to struggle with the correct classification of less frequent instances. Therefore, it would be problematic to distinguish between classes that are prominent and those that are not. Notably, the ANN, Random Forest, and Decision Tree models effectively identify cases that are more likely to capture rare cases, an important aspect when dealing with datasets that have class imbalance.

Plug-in electric vehicles (PEVs) are a practical and environmentally friendly substitute for conventional automobiles. PEVs have great potential to reduce greenhouse gas emissions by utilizing electricity as their primary energy source, thereby mitigating the negative environmental effects of traditional transportation systems. However, due to the increased and frequently irregular demand for charging, the growing integration of PEVs into the electrical grid raises significant concerns regarding operational dependability and grid stability. In addition to increasing higher charging prices and perhaps causing infrastructure stress, random charging could place further strain on the distribution network. To cope with this issue, this paper proposes a controlled charging approach with centralized control architecture to regulate and schedule the charging process of PEVs powered by machine learning techniques such as neural networks and Naive Bayes, to minimize charging costs. Simulation results demonstrate the efficacy of this strategy, showing cost savings of around 50% and 36% in comparison to the random charging process.