ClauseMatch, an award-winning regulatory technology company that enables financial institutions and other regulated organisations to comply with their internal corporate governance documentation, today announced the results of its 12-month collaboration with the Financial Services Regulatory Authority (FSRA) of Abu Dhabi Global Market (ADGM) – introducing the auto-generated knowledge graphs and AI-powered tagging of regulatory texts.
The company has been working with the regulator in the Middle East as part of the RegTech initiatives to develop new possibilities for digitalising regulation. As a result of the project, the team presented actual regulation in a form of a structured knowledge graph as a step on the way to digital regulation.
The aim of the project was to reimagine the regulatory framework by taking the content of regulatory requirements, training AI models according to ADGM taxonomy, extracting entities and linking the content on a granular paragraph-level based on themes to create a quarriable knowledge graph filled with semantic triples. Within the knowledge graph detecting document’s obligations intersection by query becomes possible. Clicking on the bubbles in a graph brings you exactly to the place where a certain topic in regulation is covered. This is the first step before such graphs are created for internal documentation such as policies, procedures, controls to map them with the requirements in a visualised dynamic form.
For various concepts, more than a thousand pages from ADGM regulations have been labeled by ClauseMatch. For most of the AI models, the accuracy score is higher than 90%. Trained AI Models learned to understand financial concepts and then when applied to the whole ADGM corpus detected hundreds of thousands of occurrences for thousands of various entities from various concepts, many of which were not even presented at the training stage. Yet, the models were still able to detect these unseen cases successfully.
Enabled by artificial intelligence and advanced NLP (Natural Language Processing) algorithms, the knowledge graphs are designed to represent regulatory data in a structured and visual format. They are an essential mechanism to transform regulation into a machine-readable format.
Knowledge graphs are clearly showing where the future of regulation lies. Graph representations enable us to infer new relationships, gain a deeper understanding and realise patterns within the regulation that we would not have spotted otherwise. It is incredible as it’s the beginning of a very interesting journey. We’re in fact witnessing the beginning of Regulation 2.0. After having Software-as-a-service, cloud-as-a-service, banking-as-a-service, we’re now moving into Regulation-as-a-service, which will truly usher in the new developments.
Anastasia Dokuchaeva, Head of Partnerships at ClauseMatch: “Regulations provide a blueprint of duties and responsibilities expected from the individuals and businesses. And in today’s world, when we are tackling many issues from economic stability, sustainability, pandemics, financial crime, information & cybersecurity, and remote working to name a few, these same regulations are rapidly changing and incredibly hard to keep up with manually.
That’s why technology is so exciting. It’s been an honour to work alongside ADGM and see our data science team pioneering the way to identify, structure and visualise regulatory requirements and interdependencies. It is a humbling experience to be a part of the journey into the new era of Know-Your-Data and Regulation-as-a-service.”