This paper explores the application of machine learning to enhance boiler efficiency and cost management at a Uranium Mine in Africa. The current steam control system relies on a feedforward loop, which adjusts based on slurry flow into the leach tank, and a feedback loop, which regulates steam to a setpoint. However, this method is inefficient, as it does not account for slurry temperature variations, leading to unstable control and suboptimal steam usage. To address these limitations, this study applies the Extra Trees algorithm to predict steam demand more accurately. The data-driven approach achieves a 6.6% reduction in steam consumption and a 2% decrease in heavy fuel oil (HFO) usage, resulting in cost savings and improved sustainability. Based on multiple evaluation metrics, the Extra Trees model proved to be the most accurate and consistent algorithm, achieving a 96.67% R-squared score and a Root Mean Square Error (RMSE) of 1131.37 kg, indicating minimal deviation between actual and predicted values. The findings highlight the shortcomings of traditional control strategies under fluctuating conditions and demonstrate how advanced feature engineering enhances predictive accuracy. By integrating machine learning into operational workflows, this research provides actionable insights to improve boiler performance, process stability, and overall efficiency.

Thunderstorms are weather disturbances that can cause lightning, stormy winds, dense clouds, tornadoes, and heavy rain. Thunderstorms can cause extensive damage to people's lives, property, and economies, as well as livestock and national infrastructure. Early warning of thunderstorms can save people's lives and property. Previous thunderstorm prediction research did not develop a system for daily thunderstorm prediction with high accuracy for Bangladeshi citizens by assessing a wide range of meteorological variables. To address this issue, this work develops a daily high accuracy based localized thunderstorm event prediction system that analyzes various meteorological factors, dates, and specific location information. This dataset was analyzed using a variety of machine learning models, including traditional statistical models like ARMA, ARIMA, and SARIMA, as well as XGBoost ensemble methods and some deep learning models such as ANN, LSTM, and GRU. The results show that advanced neural network models, particularly GRU and LSTM, outperform others in terms of RMSE, R2, MAE, and MAPE. The GRU model outperformed all other schemes, with an RMSE of 0.794, R2 of 0.998, MAE of 0.476, and MAPE of 3.544%. The mobile application provides users with accurate, localized thunderstorm forecasts, allowing for better safety, event planning, and environmental preparedness. User feedback-based mobile app assessment confirms that more than 55% of users are highly satisfied with the thunderstorm assistance app’s features and usefulness.

Understanding the prevalence of genetic dwarfism and developing detection techniques are major difficulties. Genetic dwarfism is defined by below-average stature resulting from genetic alterations. In addition to advances in detection through machine learning algorithms, this abstract investigates the analytical interpretation and comparison of genetic dwarfism statistics. In the first section, we explore the epidemiological context of genetic dwarfism, including prevalence rates, frequencies of genetic mutations, and the range of clinical presentations in various groups. The figures emphasize the intricacy of genetic variants that lead to dwarfism and emphasize the necessity for rigorous analytical methods. Improving detection and diagnostic precision through the use of machine learning algorithms appears to be a potential approach. Machine learning algorithms are trained to identify minor patterns suggestive of genetic dwarfism by utilizing datasets that include genetic profiles, medical histories, and phenotypic features. Effective methods for determining genetic markers and forecasting clinical outcomes related to dwarfism include supervised learning algorithms (e.g., decision trees, support vector machines) and deep learning architectures e.g., Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), Recurrent Neural Networks (RNNs), Autoencoders, Capsule Networks (CapsNets), Graph Convolutional Networks (GCNs), and Long Short-Term Memory (LSTM) networks). A side-by-side comparison highlights the benefits and drawbacks of machine learning techniques over conventional diagnostic techniques. Large-scale genetic data procshines but subtle pattern detection are areas where machine learning  
shines but deciphering intricate genetic connections and guaranteeing model interpretability in clinical settings continue to be difficult tasks. Moreover, the interdisciplinary aspect of tackling genetic dwarfism with modern computational tools is highlighted by ethical problems pertaining to data privacy, informed consent, and equitable access to genetic testing. Ultimately, this abstract summarizes the state of the art on genetic dwarfism statistics and machine learning applications, promoting ongoing multidisciplinary cooperation to maximize the effectiveness of therapeutic approaches and diagnosis for people with genetic dwarfism.

This paper presents an ensemble model in the determination of manifestation of emotion intensities from audio-dataset. An emotion denotes the mental state of the human mind or/and thought processes that represents a recognizable pattern of an entity like emotion arousal having a good similarity with its manifestation of vocal, facial or/and bodily signals. In this paper, we propose a stacking, late fusion approach where the best experimental outcome from two base models build from Random Forests and Extreme Gradient Boost are combined using simple majority voting. RAVDESS audio datasets, a public gender balanced dataset built by Ryerson University of Canada for the purpose of emotion study was used. 80% of the dataset was used for training while 20% was used for testing. Two features, MFCC and Chroma were introduced to the base models in a series of experimental setups and the outcome evaluated using confusion matrix, precision, recall and F1-Score. It was then compared to two state-of-the-art works done on KBES and RAVDESS datasets. This approach yielded an overall classification accuracy of 93%. 

Multi-Focus Image Fusion (MFIF) plays an important role in the field of computer vision. It aims to merge multiple images that possess different focus depths, resulting in a single image with a focused appearance. Though deep learning based methods have demonstrated development in the MFIF field, they vary significantly with regard to fusion quality and robustness to different focus changes. This paper presents the performance analysis of three deep learning-based MFIF methods specifically ECNN (Ensemble based Convolutional Neural Network), DRPL (Deep Regression Pair Learning) and SESF-Fuse. These techniques have been selected due to their publicly availability of training and testing source code, facilitating a thorough and reproducible analysis along with their diverse architectural approaches to MFIF. For training, three datasets were used ILSVRC2012, COCO2017, and DIV2K. The performance of the techniques was evaluated on two publicly available MFIF datasets: Lytro and RealMFF datasets using four objective evaluation metrics viz. Mutual Information, Gradient based metric, Piella metric and Chen-Varshney metric. Extensive experiments were conducted both qualitatively and quantitatively to analyze the effectiveness of each technique in terms of preserving details, artifacts reduction, consistency at the boundary region, texture fidelity etc. which jointly determine the feasibility of these methods for real-world applications. Ultimately, the findings illuminate the strengths and limitations of these deep learning approaches, providing valuable insights for future research and development in methodologies for MFIF.

Non-intrusive load monitoring (NILM) aims to estimate the operational states and power consumption of individual household appliances, providing real-time insights into energy usage for effective energy management and improved demand side response strategies. This study addresses the challenge of accurate energy disaggregation of household energy consumption data into individual appliances’ consumption, an important requirement for effective energy management in smart homes. Traditional energy monitoring systems provide only aggregate data, limiting the ability to optimize energy consumption. To overcome these difficulties, this study proposes a Convolutional Neural Network (CNN)-based model for Non-Intrusive Load Monitoring (NILM) that disaggregates total energy usage into appliance-specific consumption for five key appliances: kettle, microwave, fridge, dishwasher, and washing machine. Unlike the previous approaches, our model integrates a hybrid dataset from UK-DALE and REFIT, leveraging data fusion techniques to enhance generalization. The CNN architecture employed uses five convolutional layers for effective feature extraction, capturing temporal dependencies in appliance usage patterns and thus results in an improved MAE and SAE when compared to similar published results. The preprocessing and hybridization stage involves such processes as missing data imputation, appliance state labelling, feature normalization and merging of the datasets. The developed model achieved an overall accuracy of 98.3% and an F1-score of 81.7% in seen scenarios, while in unseen environments, it attained 96.5% accuracy and an F1-score of 58.1% when tested on the UK-DALE dataset. The seen scenario refers to testing using UK-DALE House 1 and REFIT House 2 data of the validation dataset, whereas the unseen scenario involves entirely new house data not used during training and validation.  It is shown that post-processing techniques reduce errors, highlighting its effectiveness, which help to enhance the model's predictive accuracy. This study contributes to the advancement of NILM technologies by combining datasets, offering a robust and scalable solution for individual appliace energy monitoring, with significant implications on energy conservation and smart home efficiency.

Accurate histopathological image classification plays a crucial role in cancer detection and diagnosis. In automated cancer detection methods, extraction of histological features of malignant and benign tissues is a challenging task. This paper presents a modified approach on octave convolution to extract high and low-frequency features which help to provide a comprehensive representation of histopathological images. Proposed octave convolution model is used to perform histopathological image classification using three different optimization strategies. Firstly, an optimal alpha value of 0.5 is used to give equal importance to both high-frequency and low-frequency feature maps. This balanced approach ensures that the model effectively considers critical high-frequency features as well as low-frequency features of cancerous tissues. Secondly, high-frequency and low-frequency feature maps are extracted and down sampled into half the spatial dimension size to reduce the computational cost compared to standard CNN. Thirdly, training and validation was conducted using ReLU, PReLU, LeakyReLU, ELU, GELU and Swish activation functions. From the experiment, it was concluded that PReLU is the best activation function for capturing intricate patterns inherent in cancer-related histopathological images. Combining all these optimization strategies, the proposed method proved to provide a classification accuracy of 93% and also to reduce the computational cost by 50%. Performance validation against pre-trained models, CNN variants and vision transformer-based models has also been conducted, which proved superior performance of the proposed model. 

Multi-objective optimization problems are crucial in real-world scenarios, where multiple solutions exist rather than a single one. Traditional methods like PERT/CPM often struggle to address such problems effectively. Meta- heuristic techniques, such as genetic algorithms and non-dominated sorting genetic algorithms (NSGA-II), are well- suited for finding true Pareto-optimal solutions. This paper introduces an enhanced NSGA-II algorithm, which utilizes Sobol sequences for initial population generation, ensuring uniform search space coverage and faster convergence. The proposed algorithm is validated using benchmark problems from the ZDT test suite and compared with state-of-the- art algorithms. Additionally, real-world optimization problems in project management, particularly the time-cost trade- off (TCT) problem, are solved using the enhanced NSGA-II. The performance evaluation includes key metrics such as standard deviation, providing a comprehensive assessment of the algorithm’s efficiency. Experimental results confirm that the proposed method outperforms traditional NSGA-II and other meta-heuristic algorithms in maintaining a well- distributed Pareto front while ensuring computational efficiency.

This study investigates logistic regression, linear support vector machine, multinomial Naive Bayes, and Bernoulli Naive Bayes for classifying Libyan dialect utterances gathered from Twitter. The dataset used is the QADI corpus, which consists of 540,000 sentences across 18 Arabic dialects. Preprocessing challenges include handling inconsistent orthographic variations and non-standard spellings typical of the Libyan dialect. The chi-square analysis revealed that certain features, such as email mentions and emotion indicators, were not significantly associated with dialect classification and were thus excluded from further analysis. Two main experiments were conducted: (1) evaluating the significance of meta-features extracted from the corpus using the chi-square test and (2) assessing classifier performance using different word and character n-gram representations. The classification experiments showed that Multinomial Naive Bayes (MNB) achieved the highest accuracy of 85.89% and an F1-score of 0.85741 when using a (1,2) word n-gram and (1,5) character n-gram representation. In contrast, Logistic Regression and Linear SVM exhibited slightly lower performance, with maximum accuracies of 84.41% and 84.73%, respectively. Additional evaluation metrics, including log loss, Cohen’s kappa, and Matthew’s correlation coefficient, further supported the effectiveness of MNB in this task. The results indicate that carefully selected n-gram representations and classification models play a crucial role in improving the accuracy of Libyan dialect identification. This study provides empirical benchmarks and insights for future research in Arabic dialect NLP applications.

One of the biggest causes of cancer-related fatalities among women is still Cervical cancer, especially in low and middle-income nations where access to broad screening and early detection may be limited. Cervical cancer is curable if detected in its early stages, but asymptomatic progression frequently results in late diagnosis, which makes treatment more difficult and lowers survival chances. Even though they work well, current screening methods including liquid-based cytology and Pap smears have drawbacks in terms of consistency, sensitivity, and specificity. Recent developments in Deep Learning and Artificial Intelligence have shown promise for greatly improving Cervical cancer detection and diagnosis. In this work, we have introduced CervixCan-Net, a novel Deep Learning based model created for the precise classification of Cervical cancer from histopathology images. Our approach offers a solid and dependable classification solution by addressing common problems like overfitting and computational inefficiency. CervixCan-Net performs better than many state-of-the-art models according to a comparison investigation. CervixCan-Net, with an impressive test accuracy of 99.83%, provides a scalable, automated Cervical cancer classification solution that has great promise for improving patient outcomes and diagnostic accuracy.