Text matching method and device, equipment and medium

Abstract

Embodiments of the invention relate to a text matching method and and device, equipment and a medium. The method comprises the steps of obtaining a target short text and a target long text; determining a candidate long text matched with the target short text; matching the target long text with the candidate long text by adopting a first matching model, and determining a first matching result; and obtaining a first matching result of the target short text and the target long text based on the first matching result. By the adoption of the technical scheme, matching between the short text and the long text is converted into matching between the two long texts, the problem that matching is inaccurate due to the fact that the information amount of the short text is too small or the information of the short text and the information of the long text are not equal can be solved, and the matching accuracy between the short text and the long text is improved.

Description

technical field

[0001] The present disclosure relates to the technical field of semantic processing, in particular to a text matching method, device, equipment and medium. Background technique

[0002] Text matching is a core issue in the field of natural language processing. It plays a central supporting role in many natural language processing applications by calculating the correlation between texts, such as information retrieval, question answering systems, dialogue systems, recommendation systems, etc.

[0003] The existing text matching technology usually extracts the features of the text and uses some similarity measurement methods to calculate the similarity between the texts or judge whether the texts are related. In addition, deep learning technology is also used to obtain the semantic vector representation of text based on neural network training, and then text matching is performed; or deep learning methods are used directly for end-to-end text matching. However...

Images