Intent acquiring method, electronic device and computer-readable storage medium

Abstract

The invention discloses a method for acquiring intent, which includes the steps of: decomposing a text sentence into multiple words; mapping the decomposed multiple words to preset keywords; determining a The intent node that is closest to the semantics of the mapped word; according to the preset slot filling method, perform semantic filling for the closest intention node to obtain the intention node after filling the slot; if the intention node after filling the slot is The exit node, according to the intention information corresponding to the intention node after filling the slot, obtains the detailed intention information associated with the intention node after filling the slot within the specified range in the detailed intention layer of the knowledge graph, and according to the acquired The detailed intention information of the user determines the user's intention; if it is a non-exit node, the user's intention is determined according to the preset inquiry mode. The invention improves migration efficiency and human-computer interaction response speed.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to an intention acquisition method, an electronic device and a computer-readable storage medium. Background technique [0002] At present, when business colleagues maintain the knowledge base, they are not very professional and do not consider the identification of knowledge classification. As a result, knowledge with poor identification may affect the overall situation after being added to the knowledge base. In addition, most of the current customer service robots do not support multi-round dialogue interaction and dialogue management, or even if they support it, manual intervention and configuration are required. Moreover, different channels such as online or telephone require customized management of multiple rounds of dialogue, resulting in poor system migration. Contents of the invention [0003] The main purpose of the present invention is to provid...

AI Technical Summary

Problems solved by technology

[0002] At present, when business colleagues maintain the knowledge base, they are not very professional and do not consider the identification of knowledge classification. As a result, knowledge with poor identification may affect the overall situation after being added to the knowledge base.

In addition, most of the current customer service robots do not support multi-round dialogue interaction and dialogue management, or even if they support it, manual intervention and configuration are required

Moreover, different channels such as online or telephone require customized management of multiple rounds of dialogue, resulting in poor system migration

Images