Intelligent man-machine conversation system in closed domain

Abstract

The invention discloses an intelligent man-machine conversation system in a closed domain. The system comprises a first modeling module, a second modeling module and a third modeling module, wherein the first modeling module is used for constructing a multi-feature fusion deep intention recognition model on the basis of a bidirectional long / short-term memory network and a convolutional neural network; the second modeling module is used for constructing an MC-BLST-MSCNN-based conversation state tracking model by adoption of a man-machine conversation system current state input and context statement combined modeling manner; and the third modeling module is used for constructing a Bi-LSTM matching model on the basis of an extraterritorial recover mechanism of a displacement attention mechanism, and inputting recognized user intentions and user slot values into a displacement network to carry out weight distribution the attention mechanism, so as to realize conversation state encoding andconversation control matching. The system has higher intention recognition accuracy, conversation state tracking correctness and conversation control stability, so as to enhance the cognition intelligence capability of man-machine systems.

Description

technical field [0001] The invention relates to the technical field of man-machine dialogue, in particular to an intelligent man-machine dialogue system in a closed domain. Background technique [0002] At present, most of the human-computer dialogue systems in the industry use the method of manually sorting out business logic and writing them into rules for dialogue matching, which do not have higher cognitive intelligence. However, the academic circles mostly focus on a certain implementation part of the human-computer dialogue system, such as dialogue status. Tracking, natural language understanding, etc., do not have a complete solution that fully considers business scenarios, and existing technologies cannot easily solve the low fluency and single-domain problems of human-machine dialogue systems. However, there are three difficulties in the dialogue system of human-computer interaction as follows. [0003] (1) Short text and lack of grammar challenge the accuracy of i...

AI Technical Summary

Problems solved by technology

[0005] (2) The complexity of repeated rounds of human-computer dialogue challenges the intelligence and migration of the dialogue model

[0007] (3) Challenges brought about by all AI-based aspects of the man-machine dialogue system

[0010] However, the intention recognition in the existing dialogue system ignores the short text and lack of grammar in the human-computer dialogue scene, resulting in low recognition accuracy; the dialogue state tracking has not been well applied, resulting in the neglect of the semantic information of the context; dialogue control Ignore the control and recovery of the dialog jumping to a non-domain, and do not implement different feedback for the context, and cannot respond to the dynamic changes of the dialog

Images