Research on Classification Model of Government Big Data Based on Deep Learning

摘要

Based on analyzing the textual features of Government Big Data, this paper proposes the Attention-BiLSTM-CNN hybrid classification model. Firstly, the model uses the BiLSTM model to extract the sentence vector of the text, and then introduces an attention mechanism to automatically learn the weight distribution of keywords, and finally uses the CNN model to calculate the probability of the category. And design experiments to verify the effect of the new model. The experimental results show that the classification model proposed in this paper has a good classification effect in the government big data corpus, which can improve the efficiency of government big data classification.