Shamrock

Inspiration

The rising concerns around public safety in India inspired us to analyze long-term trends in IPC (Indian Penal Code) crimes and build a tool that could forecast future patterns. By making the data visual and predictive, we aim to support better planning and resource allocation.

What it does

The project analyzes historical IPC crime data and forecasts future crime counts using Linear Regression and ARIMA models. It also features an interactive dashboard to explore trends across years and regions, making crime data more accessible and insightful.

How we built it

We used Python for data cleaning and modeling, with streamlit, pandas, matplotlib, scikit-learn, and statsmodels for analysis and forecasting. ARIMA was used for time series forecasting. The dashboard was built using plotly and ipywidgets in Jupyter Notebook.

Challenges we ran into

Tuning the ARIMA model parameters to get stable predictions was challenging. We also had to deal with inconsistencies in the dataset and make it suitable for time series modeling.

Accomplishments that we're proud of

Successfully combining statistical forecasting with interactive visualization. We’re proud of making the data approachable through visual tools and drawing meaningful insights from complex trends.

What we learned

We deepened our understanding of time series forecasting, data visualization, and the practical challenges of working with real-world datasets. We also gained experience in building user-friendly interactive dashboards.