Accern: AI Powered Credit Insights & Analytics
Improve your credit decisioning and risk processes by identifying adverse company events in the media.
Accern generates credit signals based on over a billion public news and social media content for companies, sectors and regions. A credit signal is identified for a company if an article contains mentions of negative news about credit events such as defaults, bankruptcy, covenant default, debt restructuring, rating downgrades, lawsuits, employment actions, analyst rating, downsizing, fraud events and many more, pertaining to a company.
– History from 2014
– Coverage of all US Companies and Sectors
– Over 150 Credit Events (i.e. Bankruptcy, Legal Actions, Analyst Ratings, etc.)
– Companies are mapped using Bloomberg FIGIs and Tickers
**coverage (e.g. private companies) and credit events can be extended if required**
Accern’s credit NLP model has been trained on massive amounts of distressed company news and uses Accern’s knowledge graph. The model identifies only highly relevant articles and passages that contain discussions of credit risk issues. For each passage, the model then computes a credit sentiment along with a number of other analytics (https://www.accern.com/adaptive-nlp).
A full data dictionary can be found (https://drive.google.com/file/d/1-S7TeQ48KxuGyNKcZT3yf6x-3F2IngU9/view)
Financial Services institutions (i.e. Banks, Insurance and Asset Management Firms) can use the dataset to help inform them of potential rating migrations, the risk of default and late payments from their service providers and counterparties by quickly being alerted of abnormal exposures based on adverse content around on a company such as default, bankruptcy, debt restructuring, lawsuits etc.
Identifying early warning signals
Detect a credit migration before it occurs by forecast the potential deterioration in the health of a company based on market data and Accern’s credit signals.
Improving their research efficiency
By filtering on the credit signals, an analyst eliminates 99% of the noise and improves their initial searches.