Aiming at the characteristics of P2P platform loan portfolio diversity, risk and return preference, the traditional recommendation algorithm can not balance income and risk, and the recommendation accuracy is low, a portfolio loan recommendation model based on multi-objective evolutionary algorithm is proposed. Firstly, the prediction rating and consistency objective function of the recommended portfolio are established to balance the accuracy and return of the recommended loan portfolio; Secondly, the objective function is solved based on the improved multi-objective evolutionary recommendation algorithm. Then, in order to solve the problem of too high dimension of decision variable space, an improved initialization strategy for reducing the dimension of decision space is proposed, so as to speed up the convergence speed of the population and improve the search efficiency of the algorithm. In the experimental part, the proposed model is compared with CF, NCF, Probs, PSO, DPA, MOEA-Probs and other models. The results show that the performance of the proposed model is improved, the average accuracy is 0.1047, the average profit coefficient is 0.1542, and the evaluation risk coefficient is 0.0023.