Case Study: CQ Roll Call achieves 98% bill passage prediction accuracy with Quantiphi's Congressional Bill Prediction solution

CQ Roll Call - Customer Case Study

CQ Roll Call worked with Quantiphi to help government relations customers make strategic decisions by predicting the probability that a legislative bill would pass the U.S. Congress. The challenge was to generate useful predictions from hundreds of variables across disparate sources, run the model daily on a large volume of bills, and track amendments that could affect the outcome.

Quantiphi built a congressional bill prediction solution using a logistic regression model with recursive feature selection, deployed on AWS services including Amazon EC2, Amazon S3, AWS Lambda, and Amazon API Gateway, and integrated it with the client’s website via API. The solution achieved 98% overall model accuracy across 18,203 bills, surfaced the top factors influencing bill passage, and enabled bill scoring through the client’s front end.

CQ Roll Call