Machine Learning Detection of Campaign Financing from FIRE Industries to Members of the 116th U.S. Congress

This paper uses a random forest machine learning classifier to detect relationships between funding of congressional candidates in the United States House of Representatives 116th congress from sources associated with the Finance, Insurance, and Real Estate (FIRE) industries, and votes cast by those members on bills and resolutions that have been lobbied by companies within FIRE industries. Patterns between representative voting and funding were investigated in two modes; detection of previous funding level using votes on relevant bills, and detection of future funding level using those votes. This paper shows detectable patterns in both modes.