Rogério de Lemos, Virginia Franqueira, Tracee Green, Marek Grzes, Robin Ayling (University of Kent)
The overall vision of this project is to investigate the role and provision of synthetic dataset for developing, evaluating, and comparing techniques for protecting young people from online harm. Since there is no similar initiative in generating and using synthetic dataset in developing age assurance techniques, the aim of this project is to evaluate the challenges associated with the generation of synthetic datasets. For that, we will be using simulations for capturing the dynamic aspects of a social networks where online harm takes place. In these dynamic environments, mitigations and changes in behaviour due to mitigations, for example, should also be incorporated into the data. Moreover, the challenges are not restricted to the generation of synthetic datasets, there is also the need to evaluate their quality, and to make recommendations for improving how these datasets are generated.