Algorithm 2 Multi-objective optimization of clustering by genetic algorithm

Input: Population size m, iteration times T

Output: Optimal clustering results (the optimal chromosomes)

Step 1: Generation of m chromosomes and formation of initial paternal population P based on improved K-medoids algorithm;

Step 2: Generation of m chromosomes by two-point crossover selection to form a new offspring population Q;

Step 3: Combining the parent population P with the offspring population Q to form a new population A;

Step 4: Non-dominant sorting;

Step 5: Crowding-distance operator computing;

Step 6: Based on the non-dominant sorting value and crowding-distance information, the optimal m chromosomes were selected from the new population A, and the parent population P was updated.;

Step 7: If the optimal termination condition is satisfied (the optimal result is obtained or the maximum number of iterations is reached), the optimal chromosome in the optimal population is selected, and the final result of clustering for sub-network is output, otherwise go to Step 2.