Text classification method and device based on intention confusion, equipment and storage medium

Abstract

The invention relates to the field of artificial intelligence, discloses a text classification method, device and equipment based on intention confusion and a storage medium, and is used for reducingthe process of setting keyword categories and improving the recognition capability of a model for multiple intentions. The method comprises the steps of obtaining a to-be-recognized text, wherein theto-be-recognized text is used for indicating at least one intention of a target user; calling an input layer of the improved rapid text classification model to recognize the text to be recognized to obtain a plurality of word segmentation vectors; calling a hidden layer of the improved rapid text classification model to carry out superposition averaging on the plurality of word segmentation vectors to obtain a document vector; calling an output layer of the improved rapid text classification model to carry out confusion intention identification on a plurality of potential intentions in the document vector to obtain a plurality of candidate intention tags; and marking the to-be-identified text based on the plurality of candidate intention tags to obtain a plurality of classified texts.

Description

technical field [0001] The present invention relates to the technical field of classification algorithms, in particular to a text classification method, device, equipment and storage medium based on intentional confusion. Background technique [0002] In a multi-turn dialogue system, natural language understanding (NLU) is a crucial step for the machine to obtain information from the user. Intent recognition for text is one of the most commonly used methods for natural language understanding, but common intent recognition is limited to one sentence corresponding to one intent, and the classifier classifies the user's words into one of the pre-designed multiple categories Go, that is, the multi-classification method. The actual situation is that a sentence of the user often expresses more than one intention. In this case, only identifying the intention cannot meet the business needs. Therefore, the dialogue system needs a natural language understanding module that can recog...

AI Technical Summary

Problems solved by technology

[0003] However, multi-intent recognition has always been a problem in the industry. The usual method is to use rule matching. The principle is to manually design keywords for all intents in advance. If a sentence matches multiple keywords, it is considered to have hit multiple intents. However, this method has disadvantages such as poor scalability, heavy manual workload, and incomplete coverage.

Images