How S&P is using deep web scraping, ensemble learning and Snowflake architecture to collect 5X more data on SMEs
1 min read
Summary
S&P Global Market Intelligence has built an AI platform that crawls data from over 200 million websites on companies, and uses machine learning (ML) and advanced algorithms to generate credit scores and risk assessments.
Called RiskGauge, the platform increases S&P’s coverage of SMEs by 5x, and offers institutional investors, banks and insurers, and wealth managers greater insights into company creditworthiness.
RiskGauge creates firmographic drivers using Snowflake’s data warehouse and Snowpark Container Services to crawl websites and process data on companies.
The platform’s machine learning models assess firms’ market risks, business credit reports, historical performances, and key developments to generate reports.
RiskGauge uses ensemble algorithms that combine predictions from various models to validate company information and create risk scores using a combination of financial, business and market risks.