Enhanced Congestion Prediction Traffic Flow Using Hybrid Attention Based Deep Learning Model
Project Description
Traffic congestion has become a critical issue worldwide, significantly impacting the quality of life and causing economic losses in urban areas. To address this, the concept of smart cities has emerged, utilizing technologies like Artificial Intelligence (AI), Machine Learning (ML), and the Internet of Things (IoT) to optimize transportation systems. One of the main challenges in smart city traffic management is accurately predicting traffic congestion. This work proposes a novel data-driven approach to predict traffic congestion in smart cities using a hybrid model that combines Bi-directional Long Short-Term Memory (Bi-LSTM), Convolutional Neural Networks (CNN), and an Attention Network. This model leverages factors such as traffic conditions, road types, and weather data to improve prediction accuracy. The model's performance is evaluated using key metrics including RMSE (Root Mean Squared Error), MSE (Mean Squared Error), and MAE (Mean Absolute Error), across four junctions (J1, J2, J3, J4). The proposed model consistently outperforms existing methods, including GRU, LSTM, CNN, and MLP. In Junction 1, the model achieved the best results with RMSE = 0.252, MSE = 0.063, and MAE = 0.178. In Junction 2, the model again led with RMSE = 0.561, outperforming GRU and LSTM models. These results confirm that the hybrid model is more effective at predicting traffic congestion in dynamic urban environments, offering a promising solution for improving traffic flow and optimizing smart city infrastructure.
This project is built using the following libraries and frameworks:
matplotlib for plotting
seaborn for data visualization
pandas for data handling
tensorflow and keras for building deep learning models
statsmodels for statistical analysis
sklearn for preprocessing and evaluation metrics
Download the dataset from Kaggle https://www.kaggle.com/datasets/fedesoriano/traffic-prediction-dataset, and place the CSV file (traffic.csv) in the project directory.
Usage Instructions
Loading the Data:
Training the Model:
Acknowledgments
This project uses the following dataset:

https://www.kaggle.com/datasets/fedesoriano/traffic-prediction-dataset