The recommender system is an intelligent system that provides recommendations to the user based on his interests. Currently, many different companies that have large sites use recommender systems to advance their work. With the increase of data in the net space, the use of traditional recommender algorithms faces problems. For this reason, it is necessary to reduce the complexity of algorithms as much as possible. The best choice to reduce the complexity of recommender algorithms is to use information preprocessing methods and optimization algorithms. Among optimization algorithms, basic methods such as particle swarm optimization algorithm (PSO), genetic optimization algorithm (GA) and... are more widely used. But these algorithms also have their own problems. To solve these problems, in this research, a new recommender system is presented using the combination of PSO-GA algorithms in data clustering. The use of GA has caused the problem of lack of diversity in the initial population of the PSO algorithm to be removed and hence the problem of getting stuck in the local optimum for PSO. The simulation results on Movielens dataset show that the error rate of the proposed method is equal to 0.67.
