Text processing method and device and computer storage medium

Abstract

Embodiments of the invention provide a text processing method and device, and a computer storage medium. The method comprises the steps of obtaining a to-be-processed text; calling a first classification network to process the secondary category coding representation, the category matching parameters and the semantic coding representation of the to-be-processed text to obtain secondary category classification information of the to-be-processed text; determining a classification result of the to-be-processed text based on the classification information of the second-level category and the classification information of the first-level category of the to-be-processed text, with the classification information of the first-level category of the to-be-processed text being obtained by processing the to-be-processed text through a second classification network, the accuracy of hierarchical classification of texts can be improved.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to a text processing method, device and computer storage medium. Background technique [0002] Hierarchical Multi-Label Classification (HMC) is an important multi-classification task in the field of natural language processing / computer vision. The granularity becomes finer at the lower levels. For example, the upper-level category is "Game", and its lower-level categories are "Mini Games", "Mobile Games", and "Client Games". [0003] At present, hierarchical classification is usually used as a basic multi-classification task, that is, the lower-level category is directly predicted, and the upper-level category is obtained by inversely deriving the lower-level category. If the predicted lower-level category is "small game", then you can get The upper-level category is "game", but directly predicting the lower-level categories often faces the problem of a small training s...

AI Technical Summary

Problems solved by technology

[0003] At present, hierarchical classification is usually used as a basic multi-classification task, that is, the lower-level category is directly predicted, and the upper-level category is obtained by inversely deriving the lower-level category. If the predicted lower-level category is "small game", then you can get The upper-level category is "game", but directly predicting the lower-level categories often faces the problem of fewer training sample sets, and due to the large number of lower-level categories, it is often impossible to classify accurately

Images