Why OSR4Rights?
Open source evidence is of increasing importance to human rights investigations, in light of the fact that investigators are often denied access to affected regions. Not only does open source intelligence help to overcome some of the logistical barriers investigators face; it has also been heralded for its great democratizing potential, by hearing some of the voices that have traditionally been silenced in investigations. However, new issues arise with these types of investigations. The huge volume of evidence retrievable from social media can make it difficult for investigators to extract truly useful information. There are further issues of informational bias that can be attributed to algorithmic bias where keywords searches are used, or to misinformation posted online, intended to obfuscate or exaggerate human rights abuses.
Work Package 1
Mapping Open Source Research in Human Rights
We carried out a comprehensive overview of the impact of Open Source Evidence on human rights investigations, and the challenges and opportunities posed in dealing with open source material in practice. Combining interview data with a systematic review of reports produced by human rights organisations, commissions of inquiry, and international courts, we determined the extent to which information gathered through open source research could complement and address some of the informational gaps inherent to traditional investigative methods, or whether the same or new informational biases arise.
Work Package 2
Transformative Methods, Transformative Impact
We conducted several Case Studies using cutting-edge natural language processing and technological techniques, combined with legal analysis, to demonstrate how open source research methods can be leveraged in practice. In dedicated workshops, outreach activities, webinars and training sessions convened with and attended by stakeholder organisations, we demonstrated our methods and key findings.
Work Package 3
Cutting-Edge Technology
We created a Virtual Machine, the ‘Knowledge Hub Framework’, which allows end-user applications to programmatically access a range of microservices. These tools assist with evidence verification and discovery tasks in relation to open source information, including: identifying place names; assigning weightings to images based on similarities, and hate speech identification. To preserve the confidentiality of evidence and to avoid the need for users to upload their evidence to the cloud, the ‘Virtual Machine’ format allows users to download the Knowledge Hub Framework to their local machine and run the tools from there. Specific tools are also available as containerized ‘Docker’ applications. If you would like to test out any of these tools, please contact us.