Considering the problems of the existing Graph Convolutional Network(GCN) recommendation models, such as low model convergence efficiency, over-smoothing and deteriorative recommendations for long-tail items due to the effect of high-degree nodes on presentation learning. This paper presents a Contrastive Learning -based

Simplified Graph Convolutional Network for recommendation algorithm (SGCN - CL). The algorithm uses the self-supervised learning method to generate multiple views for the user and item nodes for contrastive learning, in order to improve the accuracy of model recommendation and effectively improve the recommendation of long-tail items; Each view carries out the same feature extraction task for different inputs, an improved message propagation model SGCN is proposed to carry out feature extraction and enhance model efficiency. The algorithm was evaluated on Amazon-Book, Yelp2018 and Gowalla datasets. The results showed that the recall rates of three data sets increased by 15.4%、4.3%、1.4%, and NDCG increased by 17.8%、4.1%、1.6%, respectively. Additionally, the efficiency of model has increased more than 55%. After the introduction of Contrastive Learning method, the recommendation effect of non-popular long-tail items is also improved.