Man-machine conversation method based on artificial intelligence, and model training method and device

Abstract

The embodiment of the invention discloses a man-machine conversation method based on artificial intelligence and a model training method and device, and belongs to the field of natural language processing. The method comprises the steps of obtaining historical dialogue information and knowledge text information; encoding the historical dialogue information and the knowledge text information through a dialogue generation model to obtain a first semantic feature and a second semantic feature; performing fusion processing on the first semantic feature and the second semantic feature to obtain a third semantic feature; and decoding the third semantic feature to obtain dialogue reply information. The historical dialogue information is considered, and the knowledge text information associated with the historical dialogue information is also considered, so that the amount of information referenced when the dialogue reply information is generated is increased, the accuracy of generating the dialogue reply information is improved, and the diversity of the dialogue reply information is improved. Moreover, the dialogue reply information generated according to the third semantic feature is smoother, and the accuracy of generating the dialogue reply information is further improved.

Description

technical field [0001] The embodiments of the present application relate to the field of natural language processing, and in particular to an artificial intelligence-based man-machine dialogue method, model training method and device. Background technique [0002] With the rapid development of natural language processing technology, through the human-machine dialogue system or intelligent robot, dialogue reply information can be obtained according to the question text provided by the user, thereby realizing the dialogue between the user and the human-machine dialogue system or intelligent robot. [0003] In related technologies, during the dialogue process, each time a question text is input, dialogue reply information corresponding to the question text can be automatically generated. And in order to improve the accuracy rate, not only the question text, but also the question text and dialogue reply information before the question text can be considered when generating the d...

AI Technical Summary

Problems solved by technology

[0004] However, the above scheme only decodes the vectors of historical dialogue information to obtain dialogue reply information. The scheme is relatively simple, and the accuracy of generating dialogue reply information is low.

Images