Multi-document-based complex problem automatic solving method

A complex problem and multi-document technology, applied in the field of automatic solution to complex problems based on multi-document, can solve the problem that the single-document reading comprehension model does not meet the actual needs, and achieve the effect of improving recall rate and solving ambiguity

Active Publication Date: 2020-07-28
THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
View PDF2 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: The technical problem to be solved by the present invention is to provide a multi-document-based automatic solution method for comple

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-based complex problem automatic solving method
  • Multi-document-based complex problem automatic solving method
  • Multi-document-based complex problem automatic solving method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] Such as figure 1 As shown, the present invention provides a method for automatically solving complex problems based on multiple documents, which specifically includes the following steps:

[0036] Phase 1: Reading comprehension module, which mainly focuses on single-document reading comprehension technology for a single paragraph, mainly including two processes of document division into paragraphs and single-document reading comprehension:

[0037] Step (1) document segmentation, because the length of the document is too long, the neural network cannot handle the too long text, firstly, the document needs to be divided into paragraphs. First segment the document and then traverse each sentence to determine whether the length of the paragraph after adding the current sentence exceeds a given threshold. If it exceeds, a new paragraph will start; otherwise, add the sentence to the current paragraph;

[0038] Step (2) carries out data preprocessing to the divided paragraph, ...

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

According to the multi-document-based complex problem automatic solving method provided by the invention, a strategy of reading understanding and answer sorting is adopted, and a bridging entity is introduced for a complex problem with an inference process, so that the complex problem is effectively solved. Aiming at a multi-document machine reading understanding technology, firstly, documents aresplit into paragraphs, and then answers are predicted by utilizing a reading understanding model for each paragraph, so that the recall rate of the answers is effectively improved; the plurality of candidate answers is sorted by using a sorting model, and the most possible answer is selected; for a problem with an inference process, a bridging entity is firstly predicted, and then the bridging entity is spliced to an original problem, so that the ambiguity of the inference problem is effectively solved. Through the above three steps, the method can achieve the solving of a complex problem based on a plurality of documents.

Description

technical field [0001] The invention relates to a method for automatically solving complex problems based on multiple documents. Background technique [0002] Allowing machines to complete reading comprehension and question answering is a hot topic at the forefront of the current AI field, mainly involving deep learning, natural language processing and information retrieval. Machine reading comprehension has high research value in both industry and academia. It enables computers to help humans find the answers they want in a large amount of text, thereby reducing the cost of people's access to information. Perhaps, the future search engine will not only return user-related links and web pages, but will directly return the answer to the user by reading and comprehending the massive resources on the Internet. [0003] At present, the reading comprehension technology based on a single document is relatively mature, but the reading comprehension of a single document does not me...

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): G06F16/33G06F16/332G06F40/211G06F16/35G06F40/289
CPCG06F16/3329G06F16/3344G06F16/35
Inventor 徐建吴蔚李晓冬王鑫鹏徐琳阮国庆王羽
Owner THE 28TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP
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