📊 COVID-19 Metadata Analysis Dashboard 🦠

Welcome to the CORD-19 Metadata Analysis Dashboard! This Streamlit app lets you explore the COVID-19 Open Research Dataset metadata interactively. Dive into publication trends, top journals, and visualize key insights from the research papers. 📚🔍

---

🚀 Features

• Data Loading & Cleaning 🧹
  Efficiently loads the metadata.csv file, drops columns with excessive missing data, fills missing key fields, and extracts important features like publication year and abstract word count.

• Interactive Filters 🎛️
  Filter data dynamically by publication year range and select specific journals to narrow your investigation.

• Visual Insights 📈

  • Publications Over Time: Line chart showing yearly publication counts.
  • Top Journals: Bar chart highlighting most prolific publishers.
  • Word Cloud: Highlights trending words in paper titles.
  • Source Distribution: Bar chart showing paper counts by source.

• Data Preview 👀
  Display a scrollable table featuring a sample of the filtered research metadata.

---

🛠️ How It Works

1. Load and Clean Data
   Uses Pandas to load the CSV dataset, drops columns with over 50% missing values, fills important missing fields, converts publication dates to years, and counts abstract words.

2. Filter Data
   The sidebar lets users select the publication year range and pick a journal from the dataset to filter the results.

3. Display & Visualize
   Shows a sample of the filtered data and updates visualizations accordingly in real-time.

4. Efficient Caching ⏱️
   Data loading is cached for 1 hour to speed up performance and reduce repetitive computation.

---

📋 Requirements

• Python 3.7+
• Streamlit
• Pandas
• Matplotlib
• Seaborn
• WordCloud

Install dependencies via:

pip install streamlit pandas matplotlib seaborn wordcloud

---

🏃 Running the App

1. Place your metadata.csv file in the same directory as this app script.
2. Run the app with:

streamlit run Frameworks.py

3. Open the provided URL in your browser and explore! 🌐

---

🧐 Code Highlights

• @st.cache_data decorator to cache data loading and cleaning.
• Robust data preprocessing including dropping mostly empty columns and imputing missing values.
• Multiple visualizations built with Matplotlib and Seaborn.
• Intuitive Streamlit widgets for dynamic filtering.

---

⚠️ Notes

• The CSV file should include columns like title, publish_time, journal, abstract, and optionally source_x.
• The app gracefully handles missing columns by notifying the user.
• Performance may vary based on dataset size; caching helps improve responsiveness.

---

❤️ Contributions & Feedback

Feel free to open issues or submit pull requests. Your feedback helps make this dashboard better!

---

Thank you for exploring the COVID-19 research metadata with this dashboard! 🙌📊🧪