📊 CORD-19 Data Exploration and Visualization

📖 Overview

This project explores the CORD-19 metadata dataset, focusing on metadata of COVID-19 related research papers.
The workflow follows the data science pipeline: loading, cleaning, analyzing, visualizing, and deploying with Streamlit.

---

🗂️ Project Structure

• Part 1: Data Loading and Basic Exploration

  • Download and load metadata.csv
  • Explore the first rows, data types, dimensions, and missing values
  • Generate summary statistics

• Part 2: Data Cleaning and Preparation

  • Handle missing values (drop/fill where needed)
  • Convert publish_time to datetime
  • Extract year from publication date
  • Create additional columns (e.g., abstract word count)

• Part 3: Data Analysis and Visualization

  • Count publications by year
  • Identify top publishing journals
  • Find frequent words in titles
  • Visualizations:
    • Publications over time
    • Bar chart of top journals
    • Word cloud of titles
    • Distribution by source

• Part 4: Streamlit Application

  • Interactive dashboard for exploration
  • Widgets (sliders, dropdowns) for filtering
  • Displays charts and sample data

• Part 5: Documentation and Reflection

  • Well-commented code
  • Short report summarizing findings
  • Reflections on challenges and lessons learned

---

🚀 Getting Started

1️⃣ Prerequisites

Install the required Python packages:

pip install pandas matplotlib streamlit wordcloud