Unveiling Ruby: Insights from Stack Overflow and Developer Survey

This repository contains code and data for analyzing Ruby-related topics from Stack Overflow and a developer survey. It includes topic modeling, statistical comparisons, survey result processing, and detailed visualizations across different abstraction levels.

---

📖 Overview

This project investigates developer challenges with the Ruby programming language by mining Stack Overflow and conducting a complementary developer survey. It applies topic modeling, statistical analysis, and survey alignment to uncover real-world issues and perceptions around Ruby.

---

🧪 Key Components

• Data Preprocessing: Scripts to collect, clean, and filter Stack Overflow data.
• Analysis: Statistical methods to measure topic difficulty, popularity, and developer experience.
• Visualization: Publication-quality plots and charts for the paper.
• Topic Modeling: BERTopic-based modeling with hyperparameter tuning.
• Survey Analysis: Comparison between heuristic data and developer-reported perceptions.

---

🧪 Core Scripts & Instructions

1. preprocess.py

• Adds CombinedText = Title + Body
• Removes HTML, code blocks, punctuation, and links
• Saves clean text for topic modeling

2. find_best_model.py

• Runs BERTopic with SentenceTransformers and HDBSCAN
• Outputs topic_info1.csv and coherence scores
• Saves model in results/model

3. survey_vs_so_modular.py

• Loads and analyzes Likert-scale survey data
• Plots dot charts comparing survey difficulty vs SO difficulty
• Computes correlation and paired t-tests

4. heatmap_generation.py

• Step 1: Heatmap for fine-grained topic_name
• Step 2: Heatmap for middle_category
• Step 3: Heatmap for category
• Output saved as transparent PDFs

5. line_plot_analysis.py

• Step 1/1-1: Line plots per topic (year/month)
• Step 2: Trends for middle_category
• Step 3: Trends for category + Mann-Kendall significance test

---

📁 Repository Structure

.
├── README.md
├── data_preprocessing/                   # Cleans and merges textual data into CombinedText
├── analysis/                             # Analysis for results
├── notebooks/                            # Jupyter Notebooks for BERTopicing
├── utils/                                # Utils
├── visualization/                        # Generating charts and plots
├── results/
│   ├── data/                             # StackOverflow data
│   ├── model/                            # Yearly heatmaps for topics
│   └── survey/                           # Raw survey data

---

🚀 Getting Started

1. Clone the Repository

git clone https://github.com/niktaakbarpour/unveiling-ruby.git
cd unveiling-ruby

2. Create Environment and Install Dependencies

python -m venv venv
source venv/bin/activate    # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt

3. Run Analyses

python analysis/calculate_difficulty.py
python visualization/heat_map.py

---

🛠️ Dependencies

Install all required packages with:

pip install -r requirements.txt

Main dependencies:

• pandas, matplotlib, seaborn, tqdm
• scikit-learn, gensim, bertopic, hdbscan, umap-learn
• sentence-transformers, pymannkendall, bs4

---

📈 Sample Outputs

Outputs include:

• Dot charts (survey vs SO difficulty)
• Line plots of yearly/monthly post trends
• Mann-Kendall statistical test results
• Transparent PDF heatmaps for report inclusion

---

🗂️ Download Full Data & Figures

To access full-size datasets, survey data, and all visualizations:

📁 Google Drive Folder: Unveiling Ruby Resources

Contents:

• Cleaned survey & Stack Overflow datasets
• All dot charts, line plots, and heatmaps
• Supplementary figures in PNG and PDF format

Note: These files are large and not hosted directly in this repository due to GitHub's file size limits.

---

📄 Citation

If you use this work, please cite our paper:

@inproceedings{akbarpour2025unveiling,
  author    = {Nikta Akbarpour and Ahmad Saleem Mirza and Erfan Raoofian and Fatemeh Fard and Gema Rodr{\'i}guez-P\'{e}rez},
  title     = {Unveiling Ruby: Insights from Stack Overflow and Developer Survey},
  booktitle = {Proceedings of the 29th International Conference on Evaluation and Assessment in Software Engineering (EASE)},
  year      = {2025},
  location  = {Istanbul, Türkiye}
}

---

For questions, please contact: niktakbr@student.ubc.ca