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

Answer acquisition method and system based on machine reading understanding for open domain questions and answers

A technology for reading comprehension and answers, applied in the field of answer acquisition based on machine reading comprehension, and can solve problems such as poor answer results

Active Publication Date: 2020-06-23
WUHAN UNIV
View PDF12 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In view of this, the present invention provides an answer acquisition method and system based on machine reading comprehension for open domain question answering, to solve or at least partially solve the problem of answer acquisition due to the shortcomings of existing methods in information extraction and fusion methods. Technical Issues That Don't Work Well

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
  • Answer acquisition method and system based on machine reading understanding for open domain questions and answers
  • Answer acquisition method and system based on machine reading understanding for open domain questions and answers
  • Answer acquisition method and system based on machine reading understanding for open domain questions and answers

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] This embodiment provides an answer acquisition method based on machine reading comprehension for open domain question answering, please refer to figure 1 , the method includes:

[0068] S1: Obtain the reading comprehension data set, divide the training set, verification set and test set, and preprocess the acquired reading comprehension data set.

[0069] Specifically, reading comprehension datasets can be obtained from existing data sources, and preprocessing includes structured processing, data cleaning, etc.

[0070] S2: Build an answer acquisition model for reading comprehension. The answer acquisition model for reading comprehension includes a semantic encoding module for documents and questions, an information interaction encoding module, and an answer acquisition module. The semantic encoding module for documents and questions is used to extract from the input data set Capture and encode to obtain the contextual semantic representation of the document and the co...

example 1

[0131] Documentation:

[0132] Yuhang Station is a station of the Shanghai-Kunming Passenger Dedicated Line, located in the east of Liansheng Community, Nanyuan Street, Yuhang District, Hangzhou City, Zhejiang Province. Yuhang Railway Station (referred to as Yuhang Station) is a new station on the Shanghai-Hangzhou Passenger Dedicated Line. Yuhang Station has one basic platform and one side platform, and two arrival and departure lines (excluding two main lines). The opening and operation of high-speed railway Yuhang Station has made Linping Sub-City of Yuhang District a unique "city of two railways", which can enter the "30-minute urban life circle" of Greater Shanghai and Greater Hangzhou at high speed. The total length of the Yuhang Station building is about 174 meters. The design of the station fully utilizes the elements of Liangzhu culture, and the facade mainly features horizontal lines to create a "Jade Cong City". The main body of the station is a two-story island-...

example 2

[0136] Documentation:

[0137] Shangzhiling, (hiragana: じょうちれい, あげちれい), refers to the land confiscation decree issued by the Edo shogunate and the Meiji government from the 1840s to the 1870s, and is sometimes recorded as "Shangji Order". In the Edo period, the territories of temples and shrines (called temples and shrines) were recognized by the government at that time, and they were exempted from paying land rent. In 1871 (Meiji 4) and 1875 (Meiji 8), the government issued two gochi orders to confiscate those lands. At the same time as the abolition of feudal vassals and the establishment of counties, the power of the lords given to the temple community has been disintegrated, and the temple community itself has lost its legal basis. Due to the implementation of the land rent correction, under the principle that all land in the country must be paid land rent, the land that previously had tax-free privileges, including monasteries, etc., will be canceled one by one. The "Li...

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 an answer acquisition method based on machine reading understanding for open domain questions and answers. A BERT-based semantic coding module and an information interaction attention network are adopted to deeply capture potential semantic representations of problems and documents, effectively extract and fuse information between the problems and the documents, and captureglobal features of the problems and the documents; and an answer acquisition module based on Pointer Networks is adopted, and the attention weight is used as a pointer, so that the start and stop positions of the predicted answer can be positioned more accurately. The invention provides a reading comprehension-based answer acquisition method for open domain questions and answers. Empirical evaluation is performed on a CMRC 2018 data set. Experimental results show that the method can reach the standard level of open domain questioning and answering tasks, and excellent performance is obtained.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to an answer acquisition method and system based on machine reading comprehension for open domain question answering. Background technique [0002] With the development of artificial intelligence technology, human-computer interaction is becoming more and more important. Among them, the intelligent question answering of human-computer interaction is the focus of current technology development. [0003] In the field of natural language processing, the question answering system is given a corpus and a question, the algorithm system will retrieve from the corpus, and organize the answers in accurate and concise natural language to return to the user. With the continuous development of retrieval technology and knowledge base technology, open-domain question answering for a large number of document knowledge bases has developed. The representative open-domai...

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): G06F16/332G06F40/30G06N3/04G06N3/08
CPCG06F16/3329G06N3/08G06N3/045Y02D10/00
Inventor 彭敏李冬郭天翼武涵胡星灿张鼎
Owner WUHAN UNIV
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com