Automatic sentence filling method based on BERT

The invention discloses an automatic sentence filling method based on BERT, which comprises the steps of obtaining a complete original sentence used for generating training data, and performing second-order processing on the sentence by using a word segmentation tool and a syntactic analysis tree; judging words in the processed sentences in combination with priori, and replacing the words which donot conform to the priori with placeholders to obtain screened keyword groups; combining the screened phrases with the original sentences to construct training data; constructing a main network by using the pre-trained BERT as an encoder and using Transformer as a decoder, and training the network by using training data; and testing the new keyword group by utilizing the trained main network to obtain a sentence which is finally filled according to the keyword group, namely a final result. The sentence filling method has the advantages that the ability of filling sentences according to the keyword groups is good, the