A New Hybrid Deep Learning Algorithm for Prediction of Wide Traffic Congestion in Smart Cities

Pseudocode 1

Pseudocode for the boosted LSTM predictor.

1. Inputs sample training sets where where  = no of input samples and where is a multiclass label associated with

2. Initialize

3. For

4.     Train the LSTM classifier using the distribution

5.      Get the hypothesis with error function with respect to

6.        Error function is calculated at each stage which is then weighted

7.          Choose -network parameter calculation

8. Update the

9.  Calculate the error function and repeat step 4

10. If error is less than

11. Then, ensemble all the outputs

12. Else

13. Go to step 4

14. End

15. End