Check patentability & draft patents in minutes with Patsnap Eureka AI!

Multi-fragment reading method and device based on deep learning

A deep learning and reading device technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve the problems of reduced model parameters and training time, weak multi-segment text processing performance, etc., to achieve small improvement space, The effect of performance improvement and strong fitting ability

Pending Publication Date: 2022-02-15
SICHUAN INFORMATION TECH COLLEGE
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

MTQA is good at processing multi-type text corpora, but the performance of multi-segment text processing is weak
While the accuracy of the above three models needs to be improved when processing multi-segment corpus, the amount of parameters and training time of the models need to be further reduced

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-fragment reading method and device based on deep learning
  • Multi-fragment reading method and device based on deep learning
  • Multi-fragment reading method and device based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0068] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the application, not all of them. Based on the embodiments of the application, those of ordinary skill in the art All other embodiments obtained under the premise of no creative work belong to the scope of protection of this application.

[0069] In this example, if Figure 1-4 Execute as shown in the following steps:

[0070] In this embodiment, firstly, the corpus MuitSpan_DROP is made by using the multi-segment corpus in the disclosed DROP corpus and the corpus MuitSpan_NMU is made by collecting corpus according to the present invention. In the production process of the corpus, the following requirements should be followed.

[0071] (1) All answer fragments must appear in the text corresponding to the question and the number of answer fragments for each question is ...

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 relates to the technical field of data processing, and in particular relates to a multi-fragment reading method and device based on deep learning. According to the method, the improved ALBERT and the improved answer boundary prediction method SBoundary are fused. The method comprises the steps that firstly, ALBERT is improved at the upper half part of a model so that the ALBERT conforms to a reading understanding model; and secondly, an answer prediction mechanism in the lower half part of the model is improved, so that the accuracy is improved to a certain extent compared with other multi-fragment answer prediction mechanisms on the premise of conforming to multi-fragment answer extraction. Compared with other models, the accuracy of the ALBERT_SBoundary model provided by the invention is obviously improved, and meanwhile, the ALBERT_SBoundary model has smaller parameter quantity and relatively shorter training time.

Description

technical field [0001] The present invention relates to the technical field of natural language processing, in particular to a multi-segment reading method and device based on deep learning. Background technique [0002] With the rapid development of computer hardware and software and the advent of the era of artificial intelligence, various applications based on deep learning are in the ascendant. As the jewel in the crown of artificial intelligence, natural language processing is known as the last bridge to open up human-computer interaction, and it plays an important role in deep learning. As one of the important applications of text extraction in natural language processing, reading comprehension plays an important role in practical applications such as search engines and man-machine reading competitions. In recent years, research based on reading comprehension tasks has received a lot of attention at home and abroad, for example: Stanford University, Google Research In...

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): G06F16/35G06F40/289G06N3/08G06F16/332
CPCG06F16/3329G06N3/08G06F18/214
Inventor 陈德光乔治锡何雪锋廖海张倩莉俞天均
Owner SICHUAN INFORMATION TECH COLLEGE
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com