Research Article
Multiobjective Personalized Recommendation Algorithm Using Extreme Point Guided Evolutionary Computation
Input the training data, the test data, and a binary matrix R | Set g=0, L=10, N=100, T=3000, K=0.5 | Cluster all the users U in the training data into four clusters with size , | and | Set with | for j=1 to 4 | Compute the similarities of different users in based on Eq. (7) | Set | for k=1 to | Compute the available items of user , as recorded by the set | for =1 to Ns | Compute the predict rating for user to item by Eq. (8) | EPOP = Initialization(L, N, R, pr) | while g T | Epa = Objective_Func(EPOP) | Epa = Fast_Non_Dominated_Sort(EPa) | EPOP = Selection(EPOP, Epa) | EPOP = Uniform_Crossover(EPOP) | NPOP = Mutation(EPOP) | POP = | EPOP = Update(POP, Epa, N) | g = g+1 | = EPOP | Return P |
|