Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Multi-document machine reading understanding method based on hash learning

A reading comprehension and multi-document technology, which is applied in the field of multi-document machine reading comprehension based on hash learning, can solve the problems of increasing storage consumption and achieve the effect of reducing storage consumption and improving accuracy

Pending Publication Date: 2020-07-28
NANJING UNIV
View PDF5 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Considering other candidate documents when predicting the answer of each document can improve the accuracy of reading comprehension, and further increase storage consumption, causing many limitations to practical applications

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
  • Multi-document machine reading understanding method based on hash learning
  • Multi-document machine reading understanding method based on hash learning
  • Multi-document machine reading understanding method based on hash learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] Below in conjunction with specific embodiment, further illustrate the present invention, should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention, after having read the present invention, those skilled in the art will understand various equivalent forms of the present invention All modifications fall within the scope defined by the appended claims of the present application.

[0026] The multi-document machine reading comprehension method based on hash learning needs to train a multi-document reading comprehension model based on hash learning to realize the prediction of multi-document machine reading comprehension, such as figure 1 As shown, the network structure used to train the model includes an embedding layer, an encoding layer, a hashing layer, and a dynamic pointer decoding layer. The embedding layer converts the input text sequence into a vector representation, ...

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 multi-document machine reading understanding method based on hash learning. The method can achieve the effects of high prediction accuracy and low memory overhead in an online multi-document reading understanding scene. According to the method, a hash learning-based multi-document reading understanding model is trained; during machine online prediction, firstly, a pre-trained self-attention model is used for extracting text feature information of a problem and a document, then binary matrix representations corresponding to the documents are calculated, answers and probabilities of the documents are predicated by using a dynamic pointer decoder, probabilities that the documents contain correct answers are also predicated, all the answers are sequenced by combiningthe two probabilities, and the foremost answer is selected and output. The multi-document dynamic pointer decoder considers semantic information of other documents when predicting answers to each document, so that the model accuracy is improved. During model prediction, binary matrix representations of all documents are stored in a memory, so that the storage cost is reduced.

Description

technical field [0001] The invention relates to a multi-document machine reading comprehension method based on hash learning, relates to natural language processing technology, and utilizes low memory overhead to realize efficient machine reading comprehension. Background technique [0002] Multi-document reading comprehension is to give the correct answer to a given question by reading and understanding multiple related documents. Multi-document reading comprehension can be applied in an open-domain question answering system to read relevant documents queried for a given question and directly return the predicted answer to the user. Multi-document reading comprehension is an important research direction in the field of natural language processing. It also has high application value in the field of engineering and can bring great convenience to people's lives. [0003] Most of the existing multi-document reading comprehension models use the pre-trained self-attention model ...

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): G06F16/383G06F16/33G06F40/126G06N3/04G06N3/08
CPCG06F16/383G06F16/3346G06N3/049G06N3/082G06N3/084G06N3/045Y02D10/00
Inventor 李武军江悦
Owner NANJING UNIV
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
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