Tech4SocialGood aims to develop, pilot test, and study the efficacity of a teacher training program to promote both engagement as well as competence on the subject of cyberbullying. The grant holder will work on the technological components of a multi-modal learning platform (a serious game) that will be developed to that end. This will entail both research and the implementation of state-of-the art techniques from interactive digital storytelling, intelligent virtual agents and affective computing, that will support the creation of the serious game.

The workplan consists in developing the techniques and tools for the efficient monitoring and fault injection for PM applications.
The work also involves refining the existing application models that will guide the monitoring and fault injection tools.

Exploration of sparse computing methods for high-order epistasis detection, with a special focus on the evaluation of efficiency and modelling of different sparse data representation formats and operations on the-state-of-the-art computing devices.

The main goal of this work will consist in the design and development of the solutions to address the project’s challenges of text visualization, navigation and annotation.

Pattern recognition and statistical analysis of timestamped event data, including the proposal of null models and statistical tests to understand meaningful deviations. 
Development of a software application for the recognition of anomalies from timestamped event data. 
The grant holder will develop contributions with potential translation and applicability to the challenges within Task 4 of the ILU project.

Serverless is an increasingly popular cloud programming model that offers automatic scalability with high elasticity, per-request billing, and zero infrastructure management. 
Serverless functions, however, run inside containerized runtimes such as Java Virtual Machines, JavaScript Engines, Python Interpreters, requiring a separate container (or VM) and runtime for each invocation. This leads to high resource duplication and high latencies.
This project aims at taking advantage of GraalVM, a next-generation JVM capable of running many languages and compiling code ahead-of-time, to solve many of the shortcomings of current Serverless platforms.