Journal of Artificial Intelligence and Metaheuristics

Weather forecasting is a major discipline that plays an important role in fields such as agriculture, transport, and emergency management, and it largely depends on accurate forecasts. Concerning this problem, this work aimed to analyze the effectiveness of recurrent neural networks, particularly the Long Short-Term Memory (LSTM), for estimating rainfall depending on precipitation, maximum temperature, minimum temperature, and wind speed. We will therefore use a large database containing recorded weather data obtained over several years to calibrate accurate predictive models designed to distinguish between drizzle, rain, sun, snow, and fog. The main idea of the work is to teach LSTM models that are capable of revealing temporal relations and patterns in sequential data, which makes them suitable for time-series forecasting such as weather prediction. The data is preprocessed effectively to clean it and make it suitable for analysis. To accurately compare the performance of one model against the others, we have divided the data into training, validation, and testing sets. The performance of the proposed LSTM model is evaluated with the Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and the coefficient of determination (R2) to measure forecasting accuracy. The findings show improved predictive performance: the best-performing LSTM model has an MSE of 8.74, RMSE of 2.96, MAE of 2.35, and R2 of 0.83. These metrics indicate a relationship between the predicted and actual weather conditions, demonstrating the model’s efficiency. The evaluation also shows how hyperparameter optimization, feature selection, and normalization affect model performance and how precise management of weather parameters can result in better forecasts. This research has practical as well as theoretical value because it can improve the dependability of weather forecasts. More precise weather data can benefit crop growing, road networks, transportation systems, and emergency and rescue operations preparing for adverse conditions and disasters. Consequently, this study contributes to research on weather patterns and unexpected rainfall by demonstrating the potential of LSTM networks to analyze key meteorological characteristics for rainfall prediction. Possible future directions include expanding the feature set, using continuous rather than weekly data, and considering a mixed-input approach to increase prediction accuracy. This strategy seeks to improve meteorological forecasting and timely resource allocation and management under climate volatility.

In this paper, we propose the Ocotillo Optimization Algorithm (OcOA), a novel desert-inspired metaheuristic designed to solve complex optimization problems. Inspired by the adaptive strategies of desert plants, OcOA aims to achieve a balance between exploration and exploitation in high-dimensional and multimodal search spaces. The algorithm dynamically adjusts its behavior based on feedback from prior iterations, optimizing both search breadth and solution refinement. To evaluate its effectiveness, OcOA was tested against several well-known algorithms on a range of benchmark functions, including unimodal and multimodal functions from the CEC 2005 suite such as Sphere, Rosenbrock, Ackley, and Rastrigin. The results demonstrate that OcOA outperforms competing approaches in terms of accuracy, convergence speed, and computational efficiency. Additionally, its adaptability was validated through feature selection tasks, highlighting its robustness in handling both continuous and discrete optimization challenges.

The Football Optimization Algorithm (FbOA) is introduced as a novel population-based metaheuristic optimization technique inspired by the dynamic strategies of a football team. Designed to address complex optimization problems characterized by high dimensionality, nonlinearity, and multiple local optima, FbOA draws on the strategic balance between exploration and exploitation observed in football gameplay. The algorithm mimics players’ tactical positioning and movement, incorporating short passes, long passes, and positional adjustments to explore and exploit the solution space effectively. This study comprehensively evaluates the performance of FbOA using benchmark functions from the CEC 2005 test suite with 30-dimensional and 100-dimensional optimization problems. The results demonstrate that FbOA outperforms several state-of-the-art metaheuristic algorithms regarding convergence speed, accuracy, and robustness. The findings suggest that FbOA offers a promising alternative for solving various optimization challenges across multiple fields.

This is a significant problem in diagnosing zoonotic opportunistic ‘emerging’ diseases like Monkeypox, which require not only better diagnostics but also efficient, effective, and affordable diagnostics. This paper considers the possibilities of machine learning (ML), deep learning (DL), and optimization algorithms for diagnosing and predicting Monkeypox. The presently employed strategies can be enhanced because clinical and imaging data can be harnessed to drive these technologies for early detection and subsequent containment activities. Generally, in a review, the authors offer information on how the diagnostic processes using ML and DL result in enhanced accuracy, specificity, and sensitivity of models, thus reducing design reliabilities. Furthermore, outbreak data is subjected to predictive modeling analysis to establish patterns useful in helping risk managers and policymakers prepare to manage future outbreaks. This system poses a new diagnostic model for Monkeypox and other zoonotic diseases by incorporating these complex computational tools into the present healthcare systems. This advancement not only strengthens the diagnostic arsenal of zoonotic diseases but also expands the possibilities for the interception and prevention of such diseases in the future at the world level.

Accurate forecasting of renewable energy source (RES) generation is increasingly important for effective grid operation and energy management as RES are integrated into electrical grids. Machine learning (ML) and deep learning (DL) algorithms can learn complex relationships from data and provide forecasts, addressing limitations of traditional approaches. This review examines the state of the art and future directions in ML- and DL-based renewable energy forecasting. It surveys the approaches and models used to predict renewable generation and highlights challenges including output uncertainty and variability, data accessibility, and model interpretability. Developing accurate and dependable forecasting models is essential to support the transition to sustainable energy and the integration of RES into electrical grids.