Text classification model training method, text classification method and related devices

Abstract

An embodiment of the invention provides a text classification model training method, a text classification method and related devices. The text classification model training method comprises the stepsof: acquiring a text category data set for classification training; acquiring each target word vector of training classification texts by using a text classification model, and acquiring the number of word vectors; acquiring a position hierarchy and a hierarchical vector based on the number of word vectors, the position of each target word vector and the number of predetermined hierarchies; acquiring a training position text matrix according to each target word vector and each hierarchical vector, and acquiring estimated text categories of the training classification texts; and acquiring an accuracy rate of the estimated classification categories of the text category data set, and adjusting parameters of the text classification model according to the accuracy rate until the accuracy ratereaches an accuracy rate threshold value, so as to obtain the trained text classification model. According to the text classification model training method, the text classification method and the related devices of the invention, a basis can be provided for improving the text classification accuracy, so that the text classification accuracy is improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of data processing, and in particular to a text classification model training method, a text classification method and related devices. Background technique [0002] In order to recommend information for users in a more targeted manner and improve the efficiency and quality of users' acquisition of required information, users need to be classified, that is, user tags are constructed to provide different types of information for different types of users. [0003] When constructing user tags, it can be done from multiple aspects based on a variety of information, and with the popularity of online Internet platforms, it is possible to use address text to classify users. [0004] However, the current text classification methods are less accurate in classifying similar texts such as addresses. Therefore, how to provide a basis for improving the accuracy of text classification, and then i...

AI Technical Summary

Problems solved by technology

[0004] However, the current text classification methods are less accurate in classifying similar texts such as addresses. Therefore, how to provide a basis for improving the accuracy of text classification, and then improve the accuracy of text classification, has become an urgent problem to be solved. technical problem

Images