A self-learning method and device for question answering system

A self-learning method and technology of question answering system, applied in special data processing applications, instruments, electrical and digital data processing, etc., can solve the problem that the knowledge base cannot be updated automatically, and achieve the effect of smooth human-computer dialogue process.

Active Publication Date: 2018-08-21
BEIJING GUANGNIAN WUXIAN SCI & TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] One of the purposes of the present invention is to solve the technical defect that the knowledge base cannot be updated automatically in the existing question answering system

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
  • A self-learning method and device for question answering system
  • A self-learning method and device for question answering system
  • A self-learning method and device for question answering system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] This embodiment provides a self-learning method oriented to the question answering system, specifically extracting the question and answer records during the man-machine dialogue, and after verifying the relevance of the questions and answers, the real knowledge records are obtained and stored in the knowledge base to achieve The automatic expansion of the knowledge base enables the question answering system to learn the user's dialogue habits.

[0069] For the sake of clarity, the "first input information" and "second input information" appearing in the step flow will be described first. Here, "first input information" means user input information received earlier in time series, and "second input information" means user input information received later in time series. It should be noted that the first input information and the second input information may come from the same user or different users.

[0070] Wherein, the input information may be in various forms, incl...

Embodiment 2

[0089] This embodiment also provides another self-learning method. After automatically improving the learning knowledge base, the system can score each newly entered knowledge record in time, and update the score value of each record, so that it will not meet the user's dialogue habits. Knowledge records decrease the score. In the process of man-machine dialogue, the evaluation information usually follows the previous question-and-answer process. The self-learning method provided by this embodiment will score this newly entered knowledge record at any time after entering the knowledge record into the knowledge base. It can guarantee the accurate update of the score value.

[0090] The following combination figure 2 The steps and flow of the self-learning method in this embodiment will be described. and figure 1 The same steps will not be repeated. The difference from Embodiment 1 is that after the knowledge record is obtained in step S106, step S108 is executed to receive...

Embodiment 3

[0102] This embodiment provides a self-learning device for question answering systems, which can extract question and answer records during man-machine dialogue and store them in a knowledge base, realize automatic expansion of the knowledge base, and enable the question answering system to learn the user's dialogue habits. image 3 It is a schematic diagram of the structure of the self-learning device. The self-learning device mainly includes a first receiving unit 301 , a self-learning starting unit 302 , an extracting unit 303 and a storage unit 304 .

[0103] Wherein, the first receiving unit 301 is configured to receive first input information. The self-learning initiating unit 302 is configured to determine that there is a previous question-and-answer process within a preset time before receiving the first input information, and start a self-learning process.

[0104] The extracting unit 303 is configured to extract the output information in the previous question answer...

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 self-learning method and device for a question-answering system. The method comprises the steps of receiving first input information; starting the self-learning process after it is determined that a previous question-answer process exists within the preset period of time before the first input information is received; extracting output information in the previous question-answer process, storing the output information and the first input information into a knowledge base to be learnt as potential storage knowledge; calculating the relevancy between the output information and the first input information, grading the relevancy, and storing the potential storage knowledge with relevancy meeting the preset requirement in the knowledge base to obtain knowledge record. By the adoption of the method and device, feedback of user input can be achieved by means of abundant and accurate corpora stored in the knowledge base, role conversion with users can be achieved according to dialogue corpora, and then interaction with users can be achieved in the anthropomorphic mode and the man-machine dialogue procedure is smoother.

Description

technical field [0001] The present invention relates to the technical field of intelligent services, in particular to a question answering system-oriented self-learning method and device. Background technique [0002] Intelligent Q&A robot belongs to the intersection of artificial intelligence and natural language processing. It can communicate with users and answer questions through natural language, and answer questions raised by users in natural language, so as to ensure that people can quickly and accurately obtain information and satisfy people's companionship, entertainment, etc. need. [0003] Existing question-answering robots mostly select answer information matching the user's question from the knowledge base for feedback after receiving the question raised by the user. However, with the continuous emergence of new vocabulary and new sentences in real life, the corpus contained in the question and answer content is constantly enriched. In order to improve the answ...

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 Patents(China)
IPC IPC(8): G06F17/30
CPCG06F16/90332
Inventor 孙永超
Owner BEIJING GUANGNIAN WUXIAN SCI & TECH
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