One of the challenges with building solutions to mitigate harms caused by deepfakes is lack of a standard taxonomy for it. Its made worse by the fact that this field is changing at a rapid rate. This leads to people and organizations developing solutions in siloes that can't share results and resources with each other because of incompatible taxonomies or data structure. The scope of this study is to :
- Do a literature review of existing approaches to building ontologies, especially of emerging fields
- Support adding to the ontology through referencial sources (coming from research papers) and practical sources (coming from the working experience of practitioners like fact checkers, media forensic experts etc)
- How to enable participation of multiple stakeholders to keep the ontology updated
- Identifying the best practices to share the ontology such that it encourages reuse and contribution
One of the challenges with building solutions to mitigate harms caused by deepfakes is lack of a standard taxonomy for it. Its made worse by the fact that this field is changing at a rapid rate. This leads to people and organizations developing solutions in siloes that can't share results and resources with each other because of incompatible taxonomies or data structure. The scope of this study is to :