Introduction

The code for the analyses can be found in the subfolder scripts. It is structured as follows:

• 1_data_collection
• 2_data_preparation
• 3_exploratory_data_analysis

In adition, the code and procedures to create the word2vec models needed as input for running shico can be found in the subfolder w2v_models/delegates. The code is applied to the data from the delegates of congress, but can be easily adapted and equally applied to create word2vec models for all other datasets in this project.