Machine Learning is at the heart of the OakNorth Platform. Our credit analysts and developers have partnered together to train models on how credit analysis is performed. Examples include:
OCR/NLP to ingest financial documents
Detecting anomalies in financial statements
Finding and ranking Peers
Driver selection
Monitoring portfolios for potential problems
At OakNorth we create unique models that provide a granular level of analysis on each borrower. We look specifically at the borrower’s geography as well as its sector in that region to glean insights about its business. By combining borrower-provided data with our vast repository of external data, we are able to add depth to point-in-time analysis and monitoring.
In addition to traditional external data sources, the OakNorth Platform incorporates non-traditional data such as news stories, online reviews, and social media posts to provide an even richer picture of the landscape in which the borrower is operating. We are architecting a collection of data sets which power our models and give us insight into macroeconomic, geographical, and sector-specific trends.