Integrated classification method for mass multi-word short texts

Abstract

The invention discloses an integrated classification method for mass multi-word short texts, which comprises the following steps of: 1, acquiring a multi-word short text set, and performing word segmentation preprocessing on the multi-word short text; 2, obtaining a word vector representation model on the word segmentation result by utilizing a CBOW continuous word bag model in a Word2vec word vector representation method; 3, based on a word vector representation model, using a Sentence2vec sentence vector representation method to represent PV-in the method; The DM model is used for constructing sentence vector representation; And 4, on the basis of the sentence vector representation model, utilizing a kNN classifier to predict a category label of the labeled data. According to the method,the problem of'dimensionality disaster 'of a traditional representation learning method can be solved, so that the short text representation learning effect is improved, the text classification precision is improved, and the method has relatively high robustness and practicability.

Description

technical field [0001] The invention relates to the field of text representation and representation learning methods, in particular to an integrated classification method for massive multi-word short texts. Background technique [0002] With the continuous prosperity of the commodity economy, various new commodities and services are constantly emerging. According to national regulations, enterprises and individuals need to issue value-added tax invoices in accordance with the requirements in their operations. When issuing invoices, the commodities on the invoices should be associated with the tax codes approved by the State Administration of Taxation. However, there are more than 4,200 tax codes approved by the State Administration of Taxation, and there are many types. The traditional method of manually selecting tax classification codes not only requires taxpayers to have certain professional knowledge, but is also prone to filling errors, which increases the cost of busin...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the shortcomings of the above-mentioned prior art, the present invention provides an integrated classification method for massive multi-word short texts, in order to solve the problem of "dimension disaster" of traditional representation learning methods, thereby improving the effect of short text representation learning , improve the accuracy of text classification, and have high robustness and practicability

Images