About this document
Predicting Votes from Bill Text Analysis by grilifannie is a document available to read on EtoBox.
The document discusses the development of models to predict voting outcomes from legislative text using various machine learning techniques, including baseline classifiers and neural networks. It highlights the challenges of high-dimensional data and overfitting, and presents a neural network approach that achieves state-of-the-art accuracy by distilling semantic information from bill texts. The study emphasizes the importance of data cleaning and feature selection in improving model performance.
- Author
- grilifannie
- Language
- EN