R2CASS 2026 Workshop: Interactive Reproducibility Demo (Python)

This repository is designed for the "Hands-On Reproducibility Session" at the 2nd International Workshop on Social Science Meets Web Data: Reproducible and Reusable Computational Approaches (R2CASS), held in conjunction with the ACM Web Conference 2026 (WWW 2026).

It serves as the primary interactive demonstration on how to make computational social science research reproducible using Python and MyBinder.org.

Our goal is to demonstrate that with proper configuration, anyone can launch and run your analysis in a pre-configured environment, directly from a web browser, without any local setup headaches.

🚀 Launch and Explore on Binder

The easiest way to explore this project is to launch it directly on MyBinder.org. This will open an interactive JupyterLab environment in your web browser, pre-loaded with all the necessary Python libraries and the dataset.

Click the "Launch Binder" badge above to get started!

Project Overview

This demo analyzes a subset of the LIAR dataset, which contains politician statements and their fact-checking truth ratings from PolitiFact. This type of data is highly relevant to research on misinformation, political communication, and public discourse.

Dataset Citation & Source

The original LIAR dataset was created by:

William Yang Wang. 2017. "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 422–426, Vancouver, Canada. Association for Computational Linguistics.

Our subset of the dataset was downloaded from Kaggle: LIAR Dataset for Fake News Detection, where it is listed under the Apache 2.0 License.

Repository Structure

The repository is organized into distinct folders for clarity and best practices in reproducible research:

• data/: Contains the raw input dataset (dataset.tsv) used for the analysis.
• scripts/: Holds the Jupyter Notebook (analyze_claims.ipynb) containing the Python code for data loading, analysis, and visualization.
• binder/: Contains configuration files (requirements.txt, runtime.txt) that tell MyBinder.org how to build the computational environment (e.g., specific Python version and library dependencies).
• LICENSE: Specifies the open-source license for this project's code (MIT License).
• README.md: This file, providing an overview and instructions.

Analysis Performed (scripts/analyze_claims.ipynb)

The Jupyter Notebook demonstrates the following steps:

1. Data Loading: Loading the dataset.tsv file and initial inspection.
2. Truth Rating Distribution: Analyzing the frequency of different truth ratings (e.g., "true", "false", "pants-on-fire").
3. Truth Rating Visualization: Creating a bar chart to visualize the distribution of truth ratings.
4. Party Affiliation Analysis: Examining the distribution of claims by political party affiliation.
5. Party Affiliation Visualization: Creating a bar chart for claims by party.
6. Stacked Bar Chart: Truth by Party: A deeper dive into the relationship between party affiliation and truth ratings using a stacked bar chart.

How to Use (Once Launched on Binder)

1. After launching the Binder environment, you will see a JupyterLab interface in your browser.
2. In the file browser on the left, navigate to the scripts/ folder.
3. Open analyze_claims.ipynb.
4. You can then run the cells sequentially (using Shift + Enter or the "Run" button in the toolbar) to execute the analysis step-by-step and see the output and generated plots directly in the notebook.
5. The plots (truth_ratings_distribution.png, party_affiliation_distribution.png, party_vs_truth_stacked.png) will also be saved in the scripts/ folder within the Binder environment, which you can download or view.

For R Users

A similar reproducible demo, built with R, is available in the companion repository: R2CASS-Reproducibility-R-Demo

License

This project's code is licensed under the MIT License. The LIAR dataset itself is distributed under the Apache 2.0 License.