Alexiei Dingli

Education is currently facing various challenges. Teachers acknowledge that each student is unique but teaching methods are targeted towards the whole class making it difficult for teachers to cater for the needs of individual students. The Education AI system provideS a personalized learning experience. The browser-based application is aimed at both teachers and students and was specifically designed to be used on their tablets. It generates personalized classwork and homework worksheets for students. They can immediately view the results of the completed worksheets and unlock trophies. Teachers may use the app to create their own exercises or utilize the preloaded curricula. The initial tests show that low performing students who used the app benefitted the most as there was a significant improvement of around 20%. From the questionnaire responses it resulted that almost 3/4 of students preferred working out the worksheets on our system rather than on paper. The teachers surveyed agreed that the app is easy to use and a useful resource. Though we are in the first stages of development of this system, the initial results are very promising.

Check our publications:

Contact us: Alexiei DingliLara Caruana Montaldo

Advances in technology have enabled the integration of intelligent systems throughout different facets of society. From personal use such as smartphones and IoT devices to larger systems used in hospitals, entertainment, and transport. Intelligent Transport Systems (ITS) are primarily information-based systems that aim to reproduce or predict traffic flow and develop mobility management aids. The problem with implementing ITS systems is the effort in terms of installing and operating the right hardware and technology and the expense that goes with it. In addition, most ITS models work with predictive data. Different use cases are evaluated depending on the location and the model which best fits the requirements is developed. Over time, the model is tweaked to reflect more accurately the traffic and mobility demands of the surrounding environment. The automatic extraction of vehicular and transport information is an important aspect of ITS systems. Being able to effectively manage different sources of data and correctly mapping them on a live representation of the traffic situation can aid not only in correcting issues with predictive analysis but also dynamically manage mobility by controlling traffic lights, tidal lanes, and other traffic-related control structures. Our work presents an approach to modelling live data provided by different sources over the internet. The approach is inexpensive to build and operate. It aims to improve upon current implementation by making information more accessible to both authorities and citizens. In addition, a use case is presented with the intent of establishing the groundwork for future studies in the area.

Check our publications:

Contact us: Alexiei DingliMark Bugeja

Children suffering from various illnesses and diseases such as cancer have to undergo numerous procedures on a day-to-day basis as part of their treatment. According to the World Health Organisation, childhood cancer makes up between 50 to 200 cases per million every year. Most medical interventions involved with these cases are painful and as a result instil uneasiness and anxiety in the child who has to undergo these procedures. Distractions are a simple way to alleviate this anxiety and also possibly reduce the actual pain of the intervention itself. Local branches of NGOs such as Dr Klown have popularized the use of "clown doctors" to entertain children suffering from these conditions. However, such NGOs require personnel on a 24/7 basis, whilst still having to spend limited amounts of time beside every patient. The objective of this project is to make use of an Adaptive Virtual Reality (VR) system which is autonomous and can distract the child whilst undergoing treatment. The effect of such a system (which has been documented in various projects) is to reduce the pain felt by the patient. Our project will go beyond the state of the art since we will make use of adaptation which will make the VR environment change according to the level of anxiety experienced by the patient. This will be measured using biometrics collected through non-invasive wearable devices.

Contact us: Alexiei DingliLuca Bondin

Collecting data about processes and performance has been a core area of interest for most businesses. Various technological advancements have improved this process through several tools, methods, and approaches which have been introduced to this field. In the Semiconductor industry, a massive volume of ingested data is being generated from the lifecycles of different products. In contrast, gathering data in real time from platforms such as the Internet of Things (IoT) has brought a number of challenges, for instance; the structured, semi-structured and unstructured data generated from the manufacturing process, bottlenecks when designing large analytical tools, and a need for high capacity storage to save and organize the collected data. Because of this, we are designing a Digital Twin (DT) system within the Semiconductor manufacturing area. A new approach is being piloted that is going to help an multinational semiconductor manufacturing company to automate and digitize their physical production process. Furthermore, this project aims to design a high-level application that allows engineers to enhance their product design, planning, manufacturing while incorporating predictive and prescriptive maintenance.

Check our publications:

Contact us: Alexiei DingliFoaad Haddod

In recent years, Machine Learning (ML) solutions have taken great strides in domains such as Image Recognition, Natural Language Processing, Speech Automation, and many more domains. However, despite performing their tasks well, many companies which utilize these systems as a backbone for their decision-making processes are wondering how their machines are arriving to certain conclusions. More importantly, they are questioning whether or not they should trust these systems. In order to address these concerns, we are working on Explainable AI (XAI), where a machine is able to demonstrate the reasoning behind its answers in a concise, human-interpretable format. XAI is critical to create transparency in our AI systems to allow these systems to be trusted.

Contact us: Alexiei DingliNatalia Mallia