Intelligent government affair question-answering system based on big data

The invention belongs to the field of natural language processing, and particularly relates to an intelligent government affair question-answering system based on big data, which comprises a text analysis module, a feature extraction module, an information interaction module and an answer prediction module, the text analysis module calculates the correlation between the question Q of the user and the documents, and the first K documents with the highest correlation are selected as recall documents P; a feature extraction module extracts dependency relationship features between the recalled document P and the user question Q; the information interaction module performs multiple attention information interaction on the recalled document P, the user question Q and the dependency relationship characteristics to obtain a final text vector representation; and the answer prediction module predicts the starting and ending positions of the final answer by using the final text vector representation through a full conne