An interactive data analytics dashboard built on NCRB crime data (2021–2023), covering state rankings, crime type trends, women crime analysis, per capita rates, and growth-based trend extrapolation.

---

📸 Screenshots

Crime Trend Analysis	State Ranking (2023)

Crime Type Growth	Per Capita Crime Rate

---

📊 What It Covers

8 interactive views, all filterable by state and crime type:

Tab	What It Shows
📈 Trends	Year-wise crime count per state with YoY growth %
🏆 State Ranking	States ranked by total crimes in 2023
📊 YOY Growth	% change in crime volume per state (2021–2023)
📉 Crime Type Trends	Growth rates across specific IPC crime categories
👩 Women Crime	Women crime percentage by state
🗺️ Map	Geographic distribution of crime across India
🔭 Trend Extrapolation	Forward projection based on 2021–2023 growth rates
👤 Per Capita	Crime per 100K population — removes population size bias

---

🗃️ Dataset

• Source: National Crime Records Bureau (NCRB), Government of India
• Coverage: All Indian states and UTs, 2001–2023
• Crime types: IPC categories including murder, theft, kidnapping, dowry deaths, counterfeiting, women crimes, and more
• Normalization: Per capita analysis uses state-level population data to remove population size bias

---

🛠️ Tech Stack

Layer	Technology
Language	Python 3
Dashboard	Streamlit
Data processing	Pandas, NumPy
Visualizations	Plotly Express
Deployment	Streamlit Cloud

---

🚀 Local Setup

# Clone
git clone https://github.com/arka562/india-crime-dashboard.git
cd india-crime-dashboard

# Install dependencies
pip install -r requirements.txt

# Run
streamlit run app.py

---

📁 Project Structure

india-crime-dashboard/
├── app.py                  # Main Streamlit app
├── crime_model.py          # Trend extrapolation logic
├── crime_data.csv          # NCRB dataset (2001–2023)
├── screenshots/
└── requirements.txt

---

⚠️ Limitations

• Trend extrapolation is not ML — it projects forward using the 2021–2023 growth rate. This assumes that short-term trend continues, which may not hold. Treat projections as directional, not predictive.
• 2-year base window (2021–2023) is narrow. Projections are sensitive to anomalies in those specific years (e.g., COVID-period distortions in 2021).
• NCRB data reflects reported crimes only — actual crime rates may differ due to underreporting, especially for women crimes.

---

👤 Author

Arkaprava Ghosh
B.Tech IoT & Intelligent Systems, Manipal University Jaipur

---

_{Data source: NCRB Annual Reports — publicly available at ncrb.gov.in}