Traffic violation trends : machine learning driven predictive model & data visualization

Abstract

Our project presents an interactive dashboard designed for analyzing and visualizing traffic violations across Jharkhand. Built using Streamlit, the dashboard incorporates geospatial data and interactive filters to offer insights into violation trends. The system supports filtering by violation type, district, subdistrict, time, and date range, along with advanced features such as address-based radius filtering and historical data comparison. Leveraging Folium for geospatial mapping and Plotly for dynamic visualizations, the dashboard facilitates easy identification of high-violation zones, trends, and hotspots. It ensures efficient handling of large datasets with optimized visual rendering, offering actionable insights for traffic management and policy-making.