The embodiment of the invention provides a text classification method and a text classification device and is applied to the technical field of computers. The method comprise the following steps: mapping a to-be-classified text to a first text vector in a vector space model VSM through text representation; mapping the first text vector to a first mapping vector in a high-dimensional space according to a pre-established semantic smooth matrix, wherein the semantic smooth matrix is obtained by calculating a statistic similarity matrix and a semantic similarity matrix; and classifying the first mapping vector to obtain a text classification result of the to-be-classified text. According to the embodiment of the invention, by applying the semantic similarity based on world knowledge and the semantic similarity based on statistics to text classification, the accuracy of text classification is improved.