Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Method and device for optimizing intention recognition confidence threshold and storage medium

A confidence threshold and confidence technology, which can be used in digital data information retrieval, unstructured text data retrieval, special data processing applications, etc., and can solve the problems of intention recognition accuracy and low recognition rate.

Active Publication Date: 2021-05-14
南京山猫齐动信息技术有限公司
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present application provides a method, device and storage medium for optimizing the confidence threshold of intent recognition to solve the problem of low accuracy and low recognition rate of intent recognition caused by fixed thresholds

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
  • Method and device for optimizing intention recognition confidence threshold and storage medium
  • Method and device for optimizing intention recognition confidence threshold and storage medium
  • Method and device for optimizing intention recognition confidence threshold and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] figure 1 A method for optimizing the confidence threshold of intent identification provided by the embodiment of this application, such as figure 1 As shown, the method includes:

[0036] Set n thresholds Ft, 1<=t<=n, each threshold interval is sp, and calculate the recall rate, accuracy rate and overall benefit value of intent recognition corresponding to each threshold Ft;

[0037] The intent recognition threshold Ft corresponding to the maximum overall benefit value is taken as the optimal threshold Ftmax, and the intent recognition confidence threshold is set as the optimal threshold Ftmax.

[0038] Step (1): Run the intent recognition model, record the man-machine conversation, the intent recognition results during the conversation, and the corresponding intent recognition confidence value Ctx;

[0039] Step (2): Determine whether the user's actual intention is consistent with the preset intention Ix in the intention recognition model, and analyze the results of ...

Embodiment 2

[0075] The present application provides a device for optimizing the confidence threshold of intention recognition, the device includes a memory, a processor, and a computer program stored in the memory and operable on the processor, the computer program is processed by the When the device is executed, the method described in Embodiment 1 is realized.

[0076] It should be understood that in this embodiment, the memory may include a read-only memory and a random access memory, and provide instructions and data to the processor, and a part of the memory may also include a non-volatile random access memory. For example, the memory may also store device type information.

[0077] The processor can be a central processing unit CPU, and the processor can also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, off-the-shelf programmable gate array FPGA or other programmable logic devices, discrete gate or transistor logi...

Embodiment 3

[0083] The present application provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described in the first embodiment is implemented.

[0084] Storage media include: U disk, mobile hard disk, read-only memory (English full name: Read-OnlyMemory, English abbreviation: ROM), random access memory (English full name: Random Access Memory, English abbreviation: RAM), magnetic disk or optical disc, etc. Various media that can store program code.

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 invention discloses a method and device for optimizing an intention recognition confidence coefficient threshold and a storage medium, and belongs to the field of man-machine interaction robots. The method for optimizing the intention recognition confidence threshold comprises the following steps: setting n thresholds Ft, wherein the t is greater than 1 and lower than or equal to n ; calculating an intention recognition recall rate, an accuracy rate and an overall benefit value corresponding to each threshold Ft; and taking the intention recognition threshold value Ft corresponding to the maximum overall benefit value as the optimal threshold value Ftmax, and setting the intention recognition confidence coefficient threshold value as the optimal threshold value Ftmax. According to the method, personal perceptual experience relied on by manual setting of an intention recognition confidence coefficient threshold in the prior art is solved, the natural language understanding ability of the conversation robot is improved, the threshold can be automatically and regularly updated, data are more reasonable, and the intention recognition accuracy and recognition rate of the conversation robot are high.

Description

technical field [0001] The present invention relates to the field of human-computer interaction robots, in particular to a method, device and storage medium for optimizing the confidence threshold of intention recognition. Background technique [0002] During the use of the human-machine dialogue robot, the intention recognition of the user's language will be involved. According to the training model, the confidence level of different intentions will be given to the sentence. When it is higher than a certain threshold, the system can clarify the intention expressed by the sentence, and when it is lower than the threshold, it will enter the state of not recognizing the intention. If the threshold is set too low, the intent recognition error will be too large, and if the threshold is set too high, the intent recognition rate will be too low. Moreover, changes over time and application scenarios will also cause the initially preset intent recognition thresholds that perform we...

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
IPC IPC(8): G10L17/22G06F16/33G06F16/332
CPCG10L17/22G06F16/3329G06F16/3344
Inventor 陈帆吴文睿其他发明人请求不公开姓名
Owner 南京山猫齐动信息技术有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products