Peruvian-presidents-speeches

Final project for SOCI 40133: Computational Content Analysis, Winter 2020 at UChicago

About

Project name: An exploratory content analysis of the annual presidential speeches of Peru, 1956 - 2019

Authors: Carla Cristina Solis Uehara (csuehara) and Luis Eduardo San Martin Alegria (lsanmartin)

This analysis was conducted by the authors for the Computational Content Analysis course at the University of Chicago. The study analyzes a corpora of texts from the annual presidential speeches of Peru and applies a number of computational content analysis techniques to obtain political and historical insights and compare them with the main theories of ideas from political scientists studying Peru. The authors benefited greatly from the resources of the Linux servers at the University of Chicago Computer Science department.

Final presentation Final report

Setup

This project assumes the use of Python 3.5.2+. In terms of specific packages, we used the following versions:

• numpy 1.18.1+
• pandas 0.24.2+
• sklearn 0.19.2+
• requests 2.23.0+
• spacy 2.2.3
• nltk 3.4.5
• gensim 3.8.1
• altair 4.0.1
• wordcloud 1.5.0
• seaborn 0.9.0
• selenium 3.141.0

Directories

data: contains all the data we used in our analysis documents: stores the final presentation and final report of this project notebooks: stores all the notebooks we produced, where we completed the data extraction, preparation and analysis outputs: contains the visualization generated by our analysis scripts: folder where we stored all the utility functions we called from our notebooks

Data processing

Go to notebooks/data_preparation to find the notebooks that fetch and prepare the data.

Go to notebook/analysis to find the notebook that conduct every data analysis we did.