trump-speech-analysis

Statistical patterns in political rhetoric: The quantitative analysis of Donald Trump's 2024 election campaign speeches

https://github.com/brownepres/trump-speech-analysis

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (7.6%) to scientific vocabulary

Keywords

nlp nlp-machine-learning political-analysis political-science random-forest random-forest-regression sentiment-analysis
Last synced: 6 months ago · JSON representation ·

Repository

Statistical patterns in political rhetoric: The quantitative analysis of Donald Trump's 2024 election campaign speeches

Basic Info
  • Host: GitHub
  • Owner: brownepres
  • Language: Jupyter Notebook
  • Default Branch: main
  • Homepage:
  • Size: 4.23 MB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
nlp nlp-machine-learning political-analysis political-science random-forest random-forest-regression sentiment-analysis
Created about 1 year ago · Last pushed 10 months ago
Metadata Files
Readme Citation

README.md

Statistical patterns in political rhetoric: The quantitative analysis of Donald Trump's 2024 election campaign speeches

Overview

This is the GitHub repository or Barnabás Epres' research paper titled " Statistical patterns in political rhetoric: The quantitative analysis of Donald Trump's 2024 election campaign speeches " for the 2025 Scientific Students' Associations Conference (TDK). In the folders the reader can locate all code files and online resources used during the data collection, analysis and modelling.

Abstract

My paper for the Scientific Students’ Associations Conference focuses on the quantitative analysis of political speeches, narrowing it down to Donald Trump’s campaign speeches given thorough the 2024 US election rally. This study aims to analyze how the modern voters resonate with old-fashioned lengthy live political speeches amid the over-saturated space of political communication happening over social media. Public speeches naturally influence a great deal of voters and their preferences, however what I was interested in was whether the statistical properties of certain speeches correlate with the popularity, and vice versa. The research is concatenated of two main parts. In the first half descriptive statistical and text mining techniques were used to understand the characteristics of the candidate’s speeches, for example calculating the most important and unique words, the similarity of the speeches or placing them on a republican – democrat political scale. After having gathered the data, I trained and analyzed random forest regression models for better understanding the connection between the speeches and the calculated popularity of Donald Trump. Random forest regression was chosen for its robustness and interpretability. However, the results revealed, that in most cases there is only a slight connection between the popularity and the statistical characteristics of the speeches, except for when certain properties reached outlier, extreme values. This suggests that voters resonated with his rally performances almost exclusively when something “extreme” was said, which could be the result of polarization and the amplificatory effect of social media.

Owner

  • Login: brownepres
  • Kind: user

Citation (CITATION.cff)

cff-version: 1.2.0
authors:
- family-names: "Epres"
  given-names: "Barnabas"
title: "Quantitative analysis of Donald Trump's speeches"
version: 1.0.0
date-released: 2025-04-19
url: "https://github.com/brownepres/trump-speech-analysis"

GitHub Events

Total
  • Watch event: 1
  • Push event: 38
  • Create event: 2
Last Year
  • Watch event: 1
  • Push event: 38
  • Create event: 2