Unsupervised classification method for long texts

Abstract

The invention relates to an unsupervised classification method for long texts, which comprises the following steps of: filtering a to-be-classified long text, and extracting three parts, namely a title text, a body text and a text issuing department text, in the to-be-classified long text; extracting weight coefficients of the title text, the body text and the issuing department text; according to the extracted weight coefficient, fusing the title text, the body text and the text issuing department text into a new long text T; performing Chinese word segmentation on the new long text T, and extracting word segmentation information; inputting the word segmentation information into a word vector model to obtain word vector information; calculating a feature vector of the long text T according to the word vector information; and clustering the feature vectors of the long text T to obtain text classification. According to the invention, the long text classification method is improved, the time complexity of long text classification is reduced, the accuracy of long text classification is improved, and a user can read and classify long texts more conveniently.

Description

technical field [0001] The invention relates to the technical field of network information, in particular to an unsupervised classification method for long texts. Background technique [0002] In order to support the better development of enterprises, the country and the government have issued various texts needed by developing enterprises. Under the guidance of relevant texts, enterprises can more directly and accurately understand the government's guidance, and also understand the market to a large extent, so as to produce products that are more in line with market demand. The text of government preferential support involves various aspects such as departmental investment promotion, tax reduction and exemption, financing support, optimization of the environment, employment of talents, etc., directly or indirectly promotes the healthy development of enterprises. Various texts of the state and the government have become an important basis for the formulation of enterprise d...

AI Technical Summary

Problems solved by technology

Since the long text has the characteristics of long text, text structure and word habits are different from ordinary text, the time complexity of classification is high, and the classification accuracy is low

At present, there are relatively few studies on long text classification.

Images