Short text clustering method based on adaptive variational encoder

The invention discloses a short text clustering method based on an adaptive variational encoder, and relates to the technical field of text clustering. The method comprises the following steps: firstly, performing text representation on a short text by using a sense-Bert method; secondly, converting the vector into a low-dimensional feature vector by using an auto-encoder, and extracting a clustering center by using a K-means method; then, the clustering center is used as an expected mean value of the variational auto-encoder to pre-train an input vector, and the input vector is converted into a feature vector which meets distribution with the clustering center as the expected mean value; and constructing a classifier by using the feature vectors according to a K-means algorithm, and performing fine adjustment on the weights of the classifier and the encoder through distribution after classification. And finally, obtaining a clustering result according to the fine-tuned encoder and classifier. According to th