Research Article
Hybrid Gradient Descent Grey Wolf Optimizer for Optimal Feature Selection
1. Begin | 2. Randomly initialize all wolves within the function’s limits | 3. Evaluate fitness values of all wolves and sort in ascending order | a. Set the alpha wolf as the highest fitness value | b. Set the beta wolf as the second highest fitness value | c. Set the delta wolf as the third highest fitness value | 4. While the maximum number of iterations is not exceeded | a. For each wolf | i. Evaluate and using equations (3) and (4) | ii. Evaluate all 3 values of using equations ((6a), (6b), (6c)) | iii. Evaluate , , and using equations ((7a), (7b), (7c)) | iv. Evaluate the new positions using equation (8) | b. End | d. Evaluate the partial derivative of the alpha, beta, and delta wolves from equation (10) | e. Update the bottom 3 fitness value wolves using equation (9) with the update locations as the alpha, beta, and delta wolves. | c. Evaluate the fitness values of the wolves | d. Update the alpha, beta, and delta wolves | 5. End | 6. |
|