Abstract
This paper introduces the Large Events Model (LEM) for soccer, a novel deep learning framework for generating and analyzing soccer matches. The framework can simulate games from a given game state, with its primary output being the ensuing probabilities and events from multiple simulations. These can provide insights into match dynamics and underlying mechanisms. We discuss the framework's design, features, and methodologies, including model optimization, data processing, and evaluation techniques. The models within this framework are developed to predict specific aspects of soccer events, such as event type, success likelihood, and further details. In an applied context, we showcase the estimation of xP+, a metric estimating a player's contribution to the team's points earned. This work ultimately enhances the field of sports event prediction and practical applications and emphasizes the potential for this kind of method.

Abstract
This study investigates the impact of class imbalance and its potential interplay with other factors on machine learning models for transportation mode classification, utilising two real-world GPS trajectory datasets. A Random Forest model serves as the baseline, demonstrating strong performance on the relatively balanced dataset but experiencing significant degradation on the imbalanced one. To mitigate this effect, we explore various state-of-the-art class imbalance learning techniques, finding only marginal improvements. Resampling the fairly balanced dataset to replicate the imbalanced distribution suggests that factors beyond class imbalance are at play. We hypothesise and provide preliminary evidence for class overlap as a potential contributing factor, underscoring the need for further investigation into the broader range of classification difficulty factors. Our findings highlight the importance of balanced class distributions and a deeper understanding of factors such as class overlap in developing robust and generalisable models for transportation mode detection. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Abstract
This work extends the energy-efficient job shop scheduling problem with transport resources by considering speed adjustable resources of two types, namely: the machines where the jobs are processed on and the vehicles that transport the jobs around the shop-floor. Therefore, the problem being considered involves determining, simultaneously, the processing speed of each production operation, the sequence of the production operations for each machine, the allocation of the transport tasks to vehicles, the travelling speed of each task for the empty and for the loaded legs, and the sequence of the transport tasks for each vehicle. Among the possible solutions, we are interested in those providing trade-offs between makespan and total energy consumption (Pareto solutions). To that end, we develop and solve a bi-objective mixed-integer linear programming model. In addition, due to problem complexity we also propose a multi-objective biased random key genetic algorithm that simultaneously evolves several populations. The computational experiments performed have show it to be effective and efficient, even in the presence of larger problem instances. Finally, we provide extensive time and energy trade-off analysis (Pareto front) to infer the advantages of considering speed adjustable machines and speed adjustable vehicles and provide general insights for the managers dealing with such a complex problem.

Abstract
Climate change has driven agriculture to alter farming methods for food production. This paper presents a new concept for monitoring, acquisition, management, analysis, and synthesis of ecological data, which captures the environmental determinants and direct gradients suited to a particular requirement for specific plant cultivation and sustainable agriculture. The purpose of this study is to investigate a smart seablite cultivation system. A novel digital agricultural method was developed and applied to digitised seablite cultivation. Machine learning was used to predict the future growth conditions of plants (seablites). The study identified the illustrative maps of seablite origins, a conceptual seablite smart farming model, essential factors for growing seablite, a digital circuit for cultivating seablite, and digital data of seablite growth phases comprised the digital data. The findings indicate that: (1) An indicator of soil salinity is a quantity of sodium chloride extracted from a seablite sample indicating its origin of environmental determinants. (2) Saline soil, saline water, pH, moisture, temperature, and sunlight are essential factors for seablite development. These factors are dependent on climate change and were measured using a smart seablite cultivation system. (3) Digital circuits of seablite cultivation provide a better understanding of the relationship between the essential factors for seablite growth and seablite growth phases. (4) Deep neural networks outperformed vector machines, with 86% accuracy at predicting future growth of seablites. Therefore, this finding showed that the essential seablite development factors can be manipulated as key controllers for agriculture in response to climate change and agriculture can be planned. Basic digitisation of specific plants aids plant migration. Digital agriculture is an important practice for agroecosystems.

Abstract
We model the financial markets as a game and make predictions using Markov chain estimators. We extract the possible patterns displayed by the financial markets, define a game where one of the players is the speculator, whose strategies depend on his/her risk-to-reward preferences, and the market is the other player, whose strategies are the previously observed patterns. Then, we estimate the market's mixed probabilities by defining Markov chains and utilizing its transition matrices. Afterwards, we use these probabilities to determine which is the optimal strategy for the speculator. Finally, we apply these models to real-time market data to determine its feasibility. From this, we obtained a model for the financial markets that has a good performance in terms of accuracy and profitability.

Abstract