This project aims to create a system that predicts traffic congestion at future intervals for particular junctions using Gated Recurrent Unit (GRU) model. The system then assigns scores to each junction based on the congestion levels. The scores provide us an indication of the traffic density, with higher scores representing higher congestion. The system utilizes the scores with the Dijkstra algorithm to identify the optimal routes while avoiding highly congested junctions.

The following are the steps taken to predict the traffic:

• Data collection
• Data preprocessing
• Data exploration
• Data transformation
• Splitting the dataset
• Creating the model
• Fitting the model

Following steps taken to find the optimal route:

• Assign congestion scores to each junction.
• Using Dijkstra's algorithm to find the optimal route in each junction.