Accurate building energy forecasting is essential for efficient energy management and operational optimization in intelligent building systems. This study presents an explainable machine learning framework for high-resolution healthcare building energy demand forecasting. Random Forest, XGBoost, Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) models were evaluated alongside a naïve persistence baseline using engineered temporal and meteorological features. Random Forest achieved the best predictive performance, with MAE = 12.99, RMSE = 29.32, R² = 0.9930, and SMAPE = 8.17%, followed closely by XGBoost. In contrast, the recurrent neural models produced comparatively higher prediction errors. The naïve persistence model achieved a competitive SMAPE of 12.48%, highlighting strong short-term temporal dependence in healthcare energy demand. SHapley Additive exPlanations (SHAP) analysis identified recent energy demand and building size as the dominant predictors, while meteorological, seasonal, and calendar variables had smaller contributions. Ablation studies, cross-building validation, statistical significance testing, and residual diagnostics further supported model robustness. Overall, the findings demonstrate that explainable ensemble learning provides an accurate, interpretable, and scalable approach to healthcare building energy forecasting. Read more

Denial-of-Service (DoS) attacks has been a critical challenge in cybersecurity, disrupting the availability of network services and causing significant operational and economic losses. To ascertain the most suitable approaches to mitigate to dilemma, this study compares the effectiveness of some selected machine learning models in identifying denial-of-service (DoS) attacks. Two ensemble learning models, Random Forest (RF) and Extreme Gradient Boosting (XGB), showed remarkable accuracy and dependability. RF performed almost perfectly on criteria including accuracy (99%), precision (99%), and recall (99%). Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) stood out as Deep Learning models, capturing complex patterns with CNN achieving an accuracy of 98% and a perfect AUC score of 1.00. The models utilized the Recursive Feature Elimination (RFE) to select significant features and ensured proper data balancing techniques for robust model training and evaluation, minimizing overfitting and enhancing generalization. The results highlight RF and CNN as the best-performing models, with RF offering interpretability and computational efficiency, while CNN excels in handling unstructured and complex datasets. This study underscores the need for context-driven model selection and suggests exploring hybrid approaches that integrate the strengths of ML and DL for improved DoS attack detection. Future work should aim to enhance scalability and adaptability for real-world cybersecurity applications. Read more

This study investigates the impact of inflation on foreign travel dynamics in Nigeria, focusing on the Headline Inflation Rate (HIR), Core Inflation Rate (CIR), and Food Inflation Rate (FIR) as predictors of the number of passengers travelling abroad (PPF) and the percentage of aircraft travelling internationally (PAF). Secondary data was employed in this study. The Central Bank of Nigeria Statistical Bulletin 2021 and the Federal Airport Authority of Nigeria (FAAN) records from 2015–2020 were the sources of secondary data. Correlation analysis, the linear regression model, the random forest regression model, and the gradient boosting regression model are among the statistical tools used. Correlation analysis revealed significant relationships among variables, with a strong positive correlation between PPF and PAF (0.95384) and between CIR and FIR (0.75894). In contrast, HIR exhibited weak negative correlations with PPF (-0.3024) and PAF (-0.24953). Linear regression models indicated statistical significance (F-statistic = 4.102, p = 0.0106), with HIR and FIR negatively impacting PPF (p = 0.0048, p = 0.0146, respectively) and CIR positively influencing it (p = 0.0419). However, these models explained only 18% of the variability in outcomes (adjusted R² = 0.1362). Machine learning models, particularly Random Forest, demonstrated superior predictive performance, explaining 51.24% and 55.23% of the variance in PPF and PAF, respectively, with the lowest RMSE values. Gradient Boosting also outperformed linear regression. HIR was the most influential predictor for PPF, while FIR dominated for PAF. These findings highlight the nuanced effects of inflation on travel dynamics and underscore the advantages of machine learning in policy modelling. Future research should explore additional factors, such as exchange rates and consumer confidence, to enhance understanding. Read more

This research explores the classification of startup risks to achieve high investment returns using Random Forest Regression. The study aims to identify and predict potential risks faced by startups, thereby aiding investors in making informed decisions. We analyzed a dataset comprising various features such as funding levels, market size, expenses, team experience, product development stage, customer satisfaction scores, and revenue streams. We employed a Random Forest Regression model to evaluate the predictive power of these features. The model's performance was assessed using several metrics: Mean Squared Error (MSE), R-squared, Mean Absolute Error (MAE), Mean Squared Logarithmic Error (MSLE), and Explained Variance Score.The model demonstrated robust predictive capabilities, with an MSE of 0.255, R-squared of 0.9515, MAE of 0.782, MSLE of 0.219, and an Explained Variance Score of 0.915. These results indicate that the model effectively captures the variance in startup risks and predicts them with high accuracy. Feature importance analysis revealed that expenses and funding levels were the most critical factors influencing startup risk classification. The distribution of risks identified 12.4% Strategic Risks, 12.6% Financial Risks, 13.1% Operational Risks, 13.7% Market Risks, and 48.2% of activities with no significant risks.Based on our findings, we recommend that investors focus on key features as outlined in this research when assessing startup risks. By employing the insights provided by our model, investors can better identify high-potential startups, optimize resource allocation, and improve their investment strategies.The Random Forest Regression model offers a reliable tool for predicting and classifying startup risks, providing valuable insights that can enhance investment decision-making and ultimately lead to higher returns. Read more

The increasing use of credit cards in various transactions has resulted in an upsurge in fraudulent activities. This has caused significant financial losses for both individuals and businesses. This research attempted to focus on developing an efficient credit card fraud detection system using machine learning algorithms. Specifically, the Random Forest, Logistic Regression, K-nearest neighbours, Decision Trees, and naive Bayes algorithms were used to analyze the dataset and predict fraudulent activities. The dataset was preprocessed, and feature engineering techniques were applied to improve the performance of the models. Experimental results show that the Random Forest algorithm outperformed other models with an accuracy rate of 99.95%, precision of 0.85%, and recall of 0.85%. These findings indicate the potential of using machine learning algorithms in detecting credit card fraud, and the proposed system could be implemented in financial institutions and payment processing companies to improve their fraud detection systems Read more

Comparison of machine learning models is carried out in order to determine which models are best to deploy as a system. However, for the purpose of our research we carried out a comparative analysis on Random Forest classifier, Decision Tree classifier and Extra Tree classifier for weather prediction systems as we focused on seeking the classifier with the highest performance metrics. Based on the metrics, accuracy score, the best model for the system was determined. We carried out training, testing and validation of the three different models on the same dataset from the Kaggle dataset. We were able to implement Random Forest Classifier, Decision Tree Classifier and Extra Tree Classifier from Scikit-Learn to make weather prediction and using matplotlib to visualize the accuracy score of the implemented models.  The Random Forest Classifier was chosen as the best able to achieve the highest at 66% accuracy. Read more