The invention provides an oil well yield prediction method based on a deep learning algorithm. The oil well yield prediction method based on the deep learning algorithm comprises the following steps:the step 1, acquiring data and performing quality inspection; the step 2, performing data processing and division; the step 3, establishing a learning model; the step 4, carrying out training by adopting the model built in the step 3, and carrying out verification; and the step 5, predicting the oil well yield. According to the oil well yield prediction method based on the deep learning algorithm,the relations between reservoir physical properties, working systems, development stages and other factors and the oil yield and the liquid yield are established through training, the advantages of adata driving algorithm are exerted, and a multi-factor oil well yield prediction model is established.