Intent classification method and device based on business corpus and intelligent question answering method

A classification method and intent technology, applied in the field of artificial intelligence, can solve problems such as influence, lack of adaptation to business corpus, and inconsistency between intent classification and actual business needs, and achieve the effect of accurate reply and improved reply friendliness.

Inactive Publication Date: 2018-11-23
北京玄一科技有限公司
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, the intent classification method in the prior art only generally classifies user intent according to the business domain, and cannot closely relate to the actual business needs generated by the user in the specific business domain, nor does it adapt to each demand. Business corpus, it is easy to cause errors that the intent classification does not match the actual business needs, and further affect the accuracy of the reply content generated by the question answering system under specific business needs

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
  • Intent classification method and device based on business corpus and intelligent question answering method
  • Intent classification method and device based on business corpus and intelligent question answering method
  • Intent classification method and device based on business corpus and intelligent question answering method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] The embodiment of this application provides a business corpus-based intent classification method, see figure 1 , is a flow chart of a business corpus-based intent classification method provided in the embodiment of this application, such as figure 1 As shown, the method includes the following steps:

[0033] Step S110, setting the type of intent according to business requirements;

[0034] In the embodiment of this application, the intent is composed of category, user question method and reply source, wherein the category is set according to business requirements. In an optional implementation, the category can include FAQ intent, slot intent, greeting intent and structure Intent, where:

[0035] FAQ is intended to be used in a question-and-answer situation where one question corresponds to one reply. For example, there is a welcome robot in the lobby of a certain bank. When a user asks the welcome robot: "Which window is the large-value transfer?", the user's busine...

Embodiment 2

[0094]The embodiment of the present application provides an intention classification device based on business corpus, see image 3 , which is a block diagram of a business corpus-based intent classification device provided in the embodiment of the present application, such as image 3 As shown, the device includes:

[0095] A category setting unit 210, configured to set the category of the intention according to business requirements;

[0096] A user question setting unit 220, configured to set a user question matching the business requirement according to the category of the intention;

[0097] The reply source setting unit 230 is configured to set the reply source of the intention, and the reply source is associated with a corresponding knowledge base, knowledge category or knowledge entry in the business corpus.

[0098] Figure 4 A block diagram of another device for classifying intentions based on business corpus provided by the embodiment of the present application. ...

Embodiment 3

[0105] The embodiment of this application provides an intelligent question answering method, see Figure 5 , is a flow chart of an intelligent question answering method provided in the embodiment of the present application, such as Figure 5 As shown, the method includes the following steps:

[0106] Step S310, according to the user question method in the classified intentions, identify the intention contained in the user question.

[0107] In step S310, the content of the user's question is matched with the user's usage in each classified intention. If the content of the user's question is successfully matched with the user's question in a certain intention, the user's question contains the matched intention. .

[0108] Step S320, obtaining the reply content from the knowledge base, knowledge category or knowledge entry associated with the intended reply source.

[0109] In this application, the intended reply source is associated with the knowledge base, knowledge categor...

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 embodiments of the present application provide an intent classification method and device based on a business corpus and an intelligent question answering method. An intent consists of a category,a user question and a reply source. In the intent classification, the method comprises the steps of setting an intent category according to the business requirement; setting a user question matchingthe business requirement according to the intent category; and setting a reply source of the intent, the reply source being associated with a corresponding knowledge base, knowledge category or knowledge item in the business corpus, so that the intention and the business corpus are logically independent of each other, and form a corresponding relationship in the form of the user question and the reply source in the intent. After the corresponding relationship is established, the process of training an intelligent question answering system by using the business corpus becomes a process of intention training for the intelligent question answering system, so that the intelligent question answering system has the ability to accurately identify the user's intent from the user's problem after training, the reply to the question is more accurate and the reply friendliness of the intelligent question answering system is improved.

Description

technical field [0001] The present application relates to the technical field of artificial intelligence, in particular to a business corpus-based intent classification method, device and intelligent question answering method. Background technique [0002] Artificial intelligence (English: Artificial Intelligence, AI), also known as machine intelligence, refers to the intelligence shown by machines made by humans. Generally, artificial intelligence refers to human-like intelligence technology realized through computer programs. [0003] Intelligent question answering system is a research field of artificial intelligence. The intelligent question answering system can understand user questions described in natural language, obtain answers matching user questions by searching the question answer knowledge base, etc., and generate reply content containing answers. Compared with search engines, intelligent question answering systems can better understand the real intention of u...

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): G06F17/30G06N5/02
Inventor 杨凯程张正蒋宏飞
Owner 北京玄一科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap