False news detection method for theme embedded multi-mask prompt template

The invention relates to a false news detection method of a theme embedded multi-mask prompt template, and belongs to the field of natural language processing and machine learning. The method comprises the following steps: firstly, making a template for a false news detection task, and respectively designing answer words according to the false of a news text and the possibility of news occurrence; secondly, extracting a news subject term embedding template by using an LDA subject model, and inputting the template and a news text into a pre-training language model to obtain a word vector; and finally, enabling the two word vectors at the mask positions to pass through a multi-layer perceptron, outputting probability distribution of answer words, inputting the probability distribution of the answer words into a softmax layer to obtain probability distribution of news false and news occurrence possibility, and then deciding and outputting a detection result. According to the method for embedding the multi-mask p