Abstract:
Deep learning has improved short-term traffic forecasting. However, traffic flow stochasticity and distribution imbalance cause uncertainty and network overfitting during deep learning. This paper proposes M-B-LSTM, an end-to-end hybrid deep learning network model for short-term traffic flow forecasting, to address the issues. As a data mapping layer in the M-B-LSTM model, an online self-learning network learns and equalizes the traffic flow statistic distribution to reduce distribution imbalance and overfitting. In the stochasticity reducing layer, the deep bidirectional long short-term memory network (DBLSTM) reduces uncertainty by forward and reverse contexts approximation, and in the forecasting layer, the LSTM predicts traffic flow. The proposed model solves uncertainty and overfitting better than state-of-the-art methods, according to sufficient comparative experiments.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
