Regulators impose increasingly stricter regimes on financial institutions and other enterprises and practices, so those can mitigate risks associated with money laundering, terror financing, various types of fraud, and other compliance issues. Enterprises are forced to monitor a variety of data sources and then connect the dots in relation to their customers and partners.
For example, data gathering during the Know-Your-Customer (KYC) process and the screening of customers against structured economic sanctions watchlists and other bad-actor databases have been around for years. However, frequent screening against unstructured information including negative news (“adverse media”) to identify customers’ risk factors is emerging as a regulatory expectation. Overall, the gap is growing between what regulators expect and what is feasible using manual, resource-intensive solutions for compiling and screening compliance-relevant data.
Blackswan’s Approach to Compliance
BlackSwan has emerged as a technological leader in the compliance space with the aim of bringing all relevant intelligence into a single, customer-centric viewpoint. While the compliance lifecycle starts with KYC when each customer is onboarded, a holistic approach requires that transactional and other relevant activity data are routinely scanned and that a customer’s profile is constantly enriched with open-source data to identify emerging compliance risks.
At the heart of BlackSwan’s approach to compliance is a knowledge graph that builds, in real-time, a comprehensive representation of all relevant entities and the relationships between them. As such, the vital data obtained during KYC is viewed as seed data, which is enriched using a variety of social media, global news and other unstructured sources to ensure all relevant details are included in each entity’s profile. As an entity acts or new information becomes available (e.g., adverse news), the knowledge graph evolves over time to reflect the most up-to-date state of intelligence.
In order to efficiently and accurately process the universe of available data, BlackSwan uses state-of-the-art Artificial Intelligence (AI) algorithms to filter through millions of sources in seconds and append the structured information within the graph. Going beyond simply keyword tagging, multi-level semantic Natural Language Processing (NLP) and Deep Learning (DL) methods process each source and extract the actors and context in a way that mimics analyst review. Graph computation is instantaneous, implying the information is available the moment it is needed.
BlackSwan provides a holistic representation of all intelligence that is relevant for compliance:
- Machine Learning: State-of-the-art Machine Learning (ML) algorithms identify the most relevant negative news items and can assign that information to a particular compliance domain (e.g., fraud, AML, etc.).
- Natural Language Processing & Deep Learning: Multi-level semantic Natural Language Processing (NLP) and Deep Learning (DL) algorithms break each article down to the sentence-level, resolving the nature of each entity and their respective actions.
ELEMENT™ of Compliance
BlackSwan’s complete suite of capabilities are made available via ELEMENT™ of Compliance, our software application that sits atop the knowledge graph and integrates with existing data infrastructure and applications to provide a comprehensive solution. ELEMENT™ of Compliance is an industry- and function-specific solution that BlackSwan has created with ELEMENT™, our Enterprise A.I. Operating System product.
ELEMENT’s Artificial Intelligence (AI) capabilities are available throughout the entirety of the compliance process to better identify risk and enable efficiencies. These methods can, for example, enable better detection of complex risk typologies that are not easily expressed in rule-based logic, or identify alerts that are deemed low risk and enable bulk decisioning. The benefits are measured in more accurate findings and conserved resources:
- Know Your Customer (KYC) module includes an adaptive elicitation procedure that ensures all required data inputs are captured during onboarding and represented within the knowledge graph, while the extensive variety of information fetchers ensure the information is always current and complete.
- Entity matching module handles regulatory requirements of comparing targeted entities (e.g. people, organizations, vehicles), and their related entities (e.g. stakeholders, customers, suppliers), matching these entities against a dynamically maintained indexes and watchlists(sanctions, PEP, etc.) and a variety of structured and unstructured sources, while addressing the challenges of resolving and disambiguating these entities. There are special demands placed on compliance applications when supporting economic sanctions, given the potential for substantial penalties from overlooking a sanctions list entity.
- Regulators are expecting financial institutions to examine all data available, including unstructured documents and Web-based resources, including news outlets and open sources, when making judgements. ELEMENT™ of Compliance’s features for economic sanctions support continuously and comprehensively updating its knowledge graph of all entities potentially related to a customer or transaction. Continuous Reputation Monitoring, is allowing to scan and interpret countless data sources (e.g. international news outlets), resolving mentioned entities, comparing them to the persisted knowledge networks, in order to assess their compliance impact
- The Transactional Intelligence layer provides context for alerted behaviors, and can integrate and ingest alerts from existing transactional monitoring tools, or a user can use the Transactional Monitoring module within ELEMENT™ to create transactional alerts that align with and surpass industry expectations.
- While reviewing a particular customer, an analyst can access visualizations and analyses related to temporal flow of funds, peer-group comparison, and other insights that go well beyond the simple description of the alerted behaviors.
Financial Crime (Fraud)
By creating a contextual understanding of the business environment in real-time, this application discovers sophisticated patterns of potential fraud across entities, time and space. It’s the next generation of Fraud Intelligence.
Interpreting events of temporal nature and monetary value, such as financial and commercial transactions, based on the event profile, related entities, and other related events.
The first autonomous, continuously-evolving index of companies with comprehensive, customisable profiles and analysis.
A complete framework to support the identification, preservation, collection, processing, review, analysis and production of digital data in order to sustain mission critical business processes.