https://www.um.edu.mt/library/oar/handle/123456789/127756 2026-04-05T10:02:36Z 2026-04-05T10:02:36Z https://www.um.edu.mt/library/oar/handle/123456789/142036 2025-12-09T11:05:38Z 2024-01-01T00:00:00Z

Title: Multi-vehicle ride-pooling system using reinforcement learning Abstract: Ride-pooling services have surged in popularity in recent years due to them being more efficient, convenient, and cost-effective than alternative traditional methods such as taxis. Leveraging mobile technologies, ride pooling services use the location of drivers and customers to assign a shared vehicle to passengers travelling in the same direction, reducing the number of vehicles on the road and, therefore, helping reduce traffic congestion. Such services require algorithms that dynamically match passengers with nearby drivers and optimise routes. Ride-pooling can be considered a variant of the Vehicle Routing Problem (VRP), which is a combinatorial NP-hard problem as it involves finding an efficient set of routes for a fleet of vehicles to serve customers while satisfying constraints such as customers’ time windows and vehicle capacity. Literature shows how the use of various metaheuristic algorithms to solve the VRP, such as tabu search (TS), can provide good solutions in terms of quality but suffer from scalability. With the recent advances in artificial intelligence, Reinforcement Learning (RL) is also being applied to capture the dynamic and stochastic nature of the VRP. In this research, we propose to use RL to solve the Multi-Vehicle Routing for Ride-Pooling Problem (MVRRPP) and generate solutions faster than the traditional metaheuristic methods. This algorithm aims to optimise passenger allocation to vehicles and vehicle routing while minimising the overall waiting time, travel time and total driving distance. We first implemented a baseline algorithm that uses TS with an initial solution consisting of equally distributed customers along the routes to solve the MVRRPP and establish its performance. We then model the MVRRPP as an RL problem and solve it using the REINFORCE algorithm with a dynamic attention model consisting of a dynamic encoder-decoder architecture. The performance of this model was compared with the results achieved using TS. Finally, we evaluate the effect of using an RL solution as input for the TS algorithm. The results of the three models showed that TS found higher-quality solutions than RL and TS with RL; however, its computational complexity resulted in longer computation times when solving large problem instances. Using RL involved a trade-off between solution quality and computation time, where it was quicker to find a solution even for problems on a large scale. On the other hand, performance using TS with RL showed minimal improvement except for reducing the distance travelled by vehicles, which suggests that the RL solution, used as the initial solution for TS, was in a region of the search space that had an inferior local optimum then the initial solution used in the TS without RL. The choice between TS and RL for solving the MVRRPP depends on the application’s requirements and the problem’s complexity. Description: M.Sc.(Melit.)

2024-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/137982 2025-08-06T08:10:08Z 2024-01-01T00:00:00Z

Title: Digital autonomous virtual educator DAVE Abstract: Every student deserves an individualised learning experience that is tailored to their specific needs in order to help them reach their full potential. The current educational system generalises students, making them vulnerable to falling behind. Despite efforts to provide more individualised attention, teachers simply have too much on their plates. This thesis develops the Digital Autonomous Virtual Educator (DAVE) project which provides a personalised digital tutor to each student. The DAVE project utilises AI technologies to analyse student academic data, adapt academic material according to individual needs, and provide clear and detailed explanations in real-time. Through DAVE, round-the-clock personalised learning assistance is provided to students, ensuring accuracy and safety in the process. Through the integration of advanced automated prompt engineering, responses are personalised, focusing on the specific needs of each student. Additionally, by developing a novel, multi-stage verification approach named FORT Verification, harmful and incorrect content is filtered out before it is sent to users, allowing them to take confidence in the support they are receiving. The DAVE project demonstrates its efficacy through real-world testing evaluating the system's performance against commercial AI systems. Students utilising DAVE achieved improved results on mathematics worksheets by a margin of 32% while also reporting higher user satisfaction levels of 12%. Students emphasised DAVE's helpfulness, clear explanations, and personalised support. These results demonstrate the feasibility, benefits, and potential of effectively integrating AI into educational systems. Description: M.Sc. ICT(Melit.)

2024-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/136327 2025-06-10T13:21:29Z 2024-01-01T00:00:00Z

Title: Innovative pain management techniques Abstract: Pain is a prevalent issue in our everyday lives that can significantly affect our well-being. Subjective in nature, various elements affect an individual’s perception of pain. The expanding use of pain analgesia presents numerous issues which can aggravate one’s well-being. Non-trivial pain mitigating methodologies, which aim to alleviate pain without standard analgesia, are therefore required and being explored. Notably, distraction therapy, mindfulness, and meditation are all commonly employed methodologies effectively utilised for pain relief. In the domain of digital health, technology is integrated to improve the quality of medical care. Numerous technologies can be employed for pain mitigation. Virtual Reality (VR) is used in a plethora of research about pain relief due to its immersive and distraction capabilities. Given that pain is multifaceted and complex, numerous variables must be considered in its study. A standard methodology is distinguishing pain as either chronic or acute. Apart from this, other factors such as individual demographics, medical condition, and pain intensity are all elements which might affect one’s perception of pain. This work aims to further the research available in the domain of pain mitigation through the use of VR distraction therapy. Throughout this research, a personalised VR pain mitigation framework is proposed that harnesses the potential of user modelling, progressive content generation, and various VR experiences. The work utilises numerous user profiles consisting of well-known pain-characterising parameters mapped to several environmental elements the users prefer. Classifiers were trained on these profiles and used to generate a personalised environment. Progressive content generation techniques are utilised to create an environment at run-time entirely based on the user’s profile. The solution builds an endless meditative space at run-time, consisting of vegetation, different environments, and animals, amongst other immersive elements. A survey tool was devised to evaluate the framework built, which captured the participants’ characteristics and experience. An experiment was also run consisting of six anonymous participants. During the study, participants were exposed to both a standard and a personalised version of the solution for five-minute sessions. The results show that the average perceived pain intensity decreased by 68.9% from before the experiment session to after the last, significantly reducing perceived pain. Even though the standard environment effectively reduced pain intensity, the personalised environment further introduced a significant 27.5% reduction in perceived pain, displaying the efficacy such a solution has for pain mitigation. Description: M.Sc.(Melit.)

2024-01-01T00:00:00Z https://www.um.edu.mt/library/oar/handle/123456789/135523 2025-05-16T13:48:41Z 2024-01-01T00:00:00Z

Title: Artificial intelligence for team sports Abstract: Football is one of the world’s most popular sports, with a massive fan base and a yearly revenue of billions of euros. Therefore, accurately predicting the outcomes of football matches has become a crucial task within the field of sports. It has always been a chal‐ lenging task to predict the outcome of a football match, not only for fans but also for experts like bookmakers. There are multiple factors that can significantly influence the result, including the team’s form throughout a season, weather conditions, and playing style. In this dissertation, we aim to provide a comprehensive overview of the differ‐ ent methods employed to predict football match outcomes through the implementation of machine learning algorithms, while also leveraging historical data. Machine learning models have proven to be highly effective in predicting the outcome of football matches since they take into account a wide range of factors. Furthermore, these models use historical data to uncover patterns and trends that can subsequently be used to make predictions. The goal of this dissertation is to predict the full‐time result of a football match. A prediction can be classified into three possible outcomes: win, draw, or loss. The first step in predicting the outcome of a match is to collect and preprocess the data. The data collected focuses on the English Premier League, which is widely recognised as one of the most popular leagues in the world. The data is sourced from Football‐Data, an open‐source platform. In total, four machine learning algorithms are employed, Lo‐ gistic Regression, Random Forest, Extreme Gradient Boosting, and Support Vector Ma‐ chine. These algorithms are trained using an 80:20 ratio split. Initially, a baseline model is defined, employing manual feature selection and default parameters. The accuracies achieved of the models ranged between 49.5% and 55.5%, with the Logistic Regression model performing the best. Then, we conducted an optimisation procedure to fine‐tune the parameters of the achieved models. This resulted in a 55% accuracy for the Sup‐ port Vector Machine model. In the next experiment, we introduced feature selection and dimensionality reduction techniques, such as Forward Feature Selection, and Prin‐ cipal Component Analysis, whilst also keeping the default parameters for each model. The accuracies achieved ranged between 86% and 90%, with the top performer being the Random Forest model. Furthermore, another experiment is performed by combin‐ ing these techniques with an exhaustive grid search to identify the optimal parameters for each model. The Extreme Gradient Boosting model achieved the best accuracy of 94%. Furthermore, besides accuracy, other evaluation metrics are considered to gain a more detailed understanding of the predictive performance of each model. We con‐ cluded that implementing appropriate techniques and selecting optimal parameters can significantly enhance predictive power. Description: M.Sc.(Melit.)

2024-01-01T00:00:00Z