Connecting the Business World
Imagine a detailed map of the entire business world, a map that not only shows the locations of companies and people but also reveals all the intricate connections between them. That's essentially what the Strise Knowledge Graph is: a dynamic, constantly evolving network designed to help us understand the complex web of relationships in the business landscape.
What is it, really?
At its heart, the Knowledge Graph is a vast database, but instead of a simple list, it's structured like a spiderweb. This structure allows us to visualize and analyze connections. Think of it as having two fundamental components:
- Entities: These are the key players in the business world – the
Companies
, Persons
, Locations
, Industries
, and Topics
. In our map analogy, these are like the dots, the specific points of interest.
- Relationships: These are the lines that connect the dots, showing how different entities are linked. This can range from a person's position on a board of directors, to a company's ownership of another, or their operations in a particular location.
What data is in it?
The Strise Knowledge Graph contains highly curated and structured data that provides a comprehensive view of the business world. Here’s a breakdown of the key data types it holds:
- Entities:
- Companies: Includes details like ownership structures, board members, financial information, and industry classifications.
- Persons: Covers roles such as beneficial owners (UBOs), board members, and other stakeholders, enriched with metadata like nationality and date of birth.
- Locations: Tracks operational addresses, headquarters, and geographic presence of companies.
- Industries: Categorizes companies by their industry for targeted risk analysis and insights.
- Topics: Tags content and events with relevant themes (e.g., adverse, high risk, money laundering, or other relevant AMS media topics).
- Relationships:
- Ownership: Maps direct and indirect ownership structures, including UBOs.
- Board Memberships: Connects individuals to their roles within companies.
- Regulatory Links: Highlights ties to sanctions, PEPs, and adverse media mentions.
- Geographic Connections: Links to high-risk or sanctioned regions or countries.
- Events and Updates:
- Media Mentions: Articles and reports about entities, enriched with relevance scoring and contextual insights.
- Regulatory Changes: Updates on sanctions, PEP statuses, and other compliance-related events.
- Corporate Changes: Includes updates from registries, such as changes in ownership, company structure, or financial standing.
<aside>
💡
Read more: Strise Data Position Paper
</aside>
How it works
The Strise Knowledge Graph combines real-time updates, advanced context understanding, and powerful algorithms to map and connect the business world. Here's how it functions step by step:
Data ingestion
The system collects vast amounts of structured and unstructured data from diverse sources, such as news articles, company filings, and regulatory lists. This data is normalized and standardized, ensuring compatibility with the graph’s structure.
Contextual understanding
To ensure accuracy, the system disambiguates terms using natural language processing (NLP). For example, it distinguishes between "Apple" the company and "apple" the fruit by analyzing the surrounding text and context.
Entity linking