Text classification model training method and device based on multi-task fusion

Abstract

The invention provides a text classification model training method and device based on multi-task fusion, and relates to the technical field of text classification. According to the method, the method comprises the steps of obtaining first representation vector and a second representation vector based on two sample texts included in a sample text pair; obtaining a first category prediction vector and a second category prediction vector based on the first representation vector and the second representation vector; processing a splicing representation vector obtained based on the first representation vector and the second representation vector to obtain a binary classification similarity probability result; obtaining a prediction loss value based on the first category prediction vector, the second category prediction vector and the dichotomy similarity probability result; performing parameter updating on the target fusion model based on the prediction loss value to obtain an updated target fusion model; and constructing a text classification model based on the network parameters included in the updated target fusion model. Based on the method, the problem of poor classification effect in the existing text classification technology can be improved.

Description

technical field [0001] The present application relates to the technical field of text classification, in particular, to a multi-task fusion-based text classification model training method and device. Background technique [0002] Text classification is a very important module in text processing, and its applications are also very extensive, such as garbage filtering, news classification, part-of-speech tagging, sentiment analysis, etc. Among them, text classification refers to classifying and marking text sets (other entities or objects) according to a certain classification system or standard. For example, according to the set of training documents that have been labeled, find the relationship model between document features and document categories, and then use this learned relationship model to judge the category of new documents. [0003] Among them, the current mainstream text classification algorithms are statistical machine learning and deep learning. Statistical ma...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the purpose of this application is to provide a text classification model training method and device based on multi-task fusion, to improve existing text classification technology problems with poor classification performance

Images