Intention recognition method and device based on multi-round K-means algorithm, and electronic equipment

Abstract

The invention provides an intention recognition method and device based on a multi-round K-means algorithm, and electronic equipment. The method comprises steps that a sample data set is established,the sample data set comprises a plurality of semantic vectors obtained through conversion of a dialogue text, and the dialogue text is converted from voice input when a user performs dialogue with anintelligent voice robot; multiple rounds of clustering processing are conducted on the sample data set by using a K-means algorithm, and an initial clustering result is outputted; fusion denoising isconducted on all the initial clustering results to form a final clustering result; and based on the final clustering result, intention recognition is conducted on the voice input when the current userperforms conversation with the intelligent voice robot. According to the method, an improved K-means algorithm is adopted to perform multiple rounds of clustering processing, and fusion denoising isperformed on clustering results of the multiple rounds of clustering so that more accurate intention classification and identification can be realized, the intention clustering quality can be enhanced, and the method can further be optimized.

Description

technical field [0001] The invention relates to the field of computer information processing, in particular to an intention recognition method, device and electronic equipment based on a multi-round K-means algorithm. Background technique [0002] With the development of Internet technology, dialogue systems have been widely used in e-commerce, smart devices, etc., and have attracted more and more attention. Intent recognition is the primary and important task in the dialogue system, and it is a multi-classification problem. Correspondingly, a multi-classification model needs to be used for corresponding processing. In fact, the classification problem is the basic problem of machine learning, and there are regression, clustering and so on. [0003] Clustering is to classify existing data objects, so that the similarity between each category is as large as possible, and the similarity between each category is as small as possible. In general, in the process of dividing data...

AI Technical Summary

Problems solved by technology

However, there is still much room for improvement in terms of applied clustering quality and method optimization

Images