2020/04/29 by Ivan Slobozhan, Slobozhan, Ivan, Peter L. Ormosi +3
Business, Management and Accounting · Social Sciences · #Auditing, Earnings Management, Governance #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Economics and business #General Economics (econ.GN) #Judicial and Constitutional Studies #Machine Learning (cs.LG) #Political Influence and Corporate Strategies
paper · pdf · doi:10.48550/arxiv.2005.06386
openalex publication_date 2020/04/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Using lobbying data from OpenSecrets.org, we offer several experiments\napplying machine learning techniques to predict if a piece of legislation (US\nbill) has been subjected to lobbying activities or not. We also investigate the\ninfluence of the intensity of the lobbying activity on how discernible a\nlobbied bill is from one that was not subject to lobbying. We compare the\nperformance of a number of different models (logistic regression, random\nforest, CNN and LSTM) and text embedding representations (BOW, TF-IDF, GloVe,\nLaw2Vec). We report results of above 0.85% ROC AUC scores, and 78% accuracy.\nModel performance significantly improves (95% ROC AUC, and 88% accuracy) when\nbills with higher lobbying intensity are looked at. We also propose a method\nthat could be used for unlabelled data. Through this we show that there is a\nconsiderably large number of previously unlabelled US bills where our\npredictions suggest that some lobbying activity took place. We believe our\nmethod could potentially contribute to the enforcement of the US Lobbying\nDisclosure Act (LDA) by indicating the bills that were likely to have been\naffected by lobbying but were not filed as such.\n