Unlock instant, AI-driven research and patent intelligence for your innovation.

Semantic recognition method and device, electronic equipment and storage medium

A semantic recognition and semantic technology, applied in semantic analysis, digital data processing, natural language data processing, etc., can solve the problems of cumbersome semantic recognition process, low recognition accuracy, low performance and efficiency, etc. Simplify training process and improve efficiency

Pending Publication Date: 2021-05-21
ZTE CORP
View PDF0 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The semantic recognition method, device, electronic equipment, and storage medium provided by the embodiments of the present invention at least solve the problems in related technologies that the model training process and semantic recognition process are cumbersome, the performance and efficiency are low, and the recognition accuracy is also low.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Semantic recognition method and device, electronic equipment and storage medium
  • Semantic recognition method and device, electronic equipment and storage medium
  • Semantic recognition method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0024] In view of the problems that the model training process and semantic recognition process are cumbersome, the performance and efficiency are low, and the recognition accuracy rate is also low, this embodiment performs joint training on the intent recognition model and the slot value labeling model during the model training process. Separate training for each model can simplify the model training process and improve efficiency.

[0025] At the same time, in semantic recognition, after the query text to be recognized is obtained, the query file to be recognized can be recognized through the semantic joint recognition model, and the result of intent recognition and slot value recognition is output. The semantic joint recognition model includes: according to the first model loss of the intent recognition model and the second model loss of the slot value tagging model, the intent recognition model and the slot value tagging model are jointly trained to obtain a relevant intent...

Embodiment 2

[0084] The method provided by the foregoing embodiments can be applied to various electronic devices such as servers, common PCs or even embedded mobile devices, and the devices can have hardware devices such as a central processing unit CPU, memory or even a graphics processing unit GPU (optional), and Power supply is required; the method provided by the foregoing embodiments is also applicable to operating system software (Linux, Windows, etc.), and can be combined with deep learning platforms (such as TensorFlow, Pytorch, etc.), machine learning software libraries (such as sk-learn, etc.) or Relevant algorithm code implementation realized by the integrated development environment of the software based on the relevant computer language.

[0085] This embodiment provides a semantic recognition model training device, which can be set in electronic equipment, please refer to Figure 6 shown, including:

[0086] The information extraction module 601 is used to extract the langu...

Embodiment 3

[0108] This embodiment also provides an electronic device, which can be a server and various terminals as above, see Figure 10 As shown, it includes a processor 1001, a memory 1002 and a communication bus 1003;

[0109] The communication bus 1003 is used to realize the communication connection between the processor 1001 and the memory 1002;

[0110] In an example, the processor 1001 may be configured to execute a computer program stored in the memory 1002, so as to implement the steps of the semantic recognition method in the above embodiments.

[0111] The present embodiment also provides a computer-readable storage medium, which includes information implemented in any method or technology for storing information, such as computer-readable instructions, data structures, computer program modules, or other data. volatile or nonvolatile, removable or non-removable media. Computer-readable storage media include but are not limited to RAM (Random Access Memory, random access me...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The embodiment of the invention provides a semantic recognition method and device, electronic equipment and a storage medium. Joint training is performed on an intention recognition model and a slot value labeling model to obtain a semantic joint recognition model, a query text is predicted, and associated prediction output of an intention prediction result and a slot value prediction result of the query text can be directly realized; the intention recognition model and the slot value labeling model are jointly trained in the model training process, so that each model does not need to be independently trained, the model training process can be simplified, and the efficiency can be improved; and the association prediction output of a query text intention prediction result and a slot value prediction result can be directly realized by utilizing the intention recognition model and the slot value labeling model which are obtained by training and have association, so that the recognition efficiency can be improved, and the recognition accuracy can be improved by utilizing the association.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a semantic recognition method, device, electronic equipment and storage medium. Background technique [0002] Human-machine dialogue systems supported by voice technology and NLP technology are widely used in scenarios such as smartphones, smart homes, and vehicle-mounted equipment. It usually includes three parts: semantic recognition (also called semantic understanding), dialogue management and reply generation. Semantic recognition is an important part of it and the basis of each subsequent step. It refers to recognizing the intent of the query text entered by the user and the entity information contained in it. Specifically, it mainly includes two main tasks: intent recognition and slot value labeling. In related technologies, semantic recognition is only based on the global language using the training language model to recognize the query text, and usually requires se...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/30G06F40/289G06F16/33G06F16/332G06F16/35
CPCG06F16/3329G06F16/3344G06F16/35
Inventor 李向阳胡韧奋谢志华
Owner ZTE CORP