Relation pair library table construction method

A relational pair and relational technology, applied in the field of relational pair library table construction, can solve problems such as missing and achieve the effect of ensuring accuracy

Active Publication Date: 2019-11-05
北京中科瑞通信息科技有限公司
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the restriction is that the templates in the schema set must be identical, it is likely not to be found

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
  • Relation pair library table construction method
  • Relation pair library table construction method
  • Relation pair library table construction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The user enters two relationships to form a relationship pair. After clicking Search, the system will check whether the relationship pair has been extracted. If it has been extracted, it means that there are corresponding record results in the database, and these results will be returned directly to the page. In the boxes of "candidate set" and "sentence template"; if it has not been extracted, the system will immediately extract the incoming relationship pair, and after the extraction is completed, the generated template and relationship will be written into the database, and the The results are returned to the page. The logical process is as figure 1 shown.

[0055] Relationship extraction follows this logic: input two different relationship pairs, and then extract sentences containing two relationship entities in a relationship pair in ES (ElasticSearch), and then use the two relationship entity words in these sentences with Entity placeholders arg0 and arg1 instea...

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 character processing, and provides a relation pair library table construction method. The method aims to manually specify several initial seeds. After that, a system searches for sentence pattern templates meeting the requirement for manually providing the seeds. New seeds are found through the obtained templates. The iteration is conducted continuously. Finally, the purpose of one-reverse-three is achieved. A large number of triples meeting the manually specified relation are found. According to the main scheme, the method comprises the steps ofinputting multiple relation pairs to form a relation pair set; according to the two entities in the relationship pair, searching for literature statements containing the entities in a search server;replacing two entity words in the sentence with entity placeholders arg0 and arg1 to form a sentence pattern template; and comparing whether the relationship pairs corresponding to the sentence pattern templates conform to the two input relationship pairs at the same time, if so, leaving the sentence pattern templates, searching sentences containing the left sentence pattern templates in the search server, and finding out the corresponding parts of arg0 and arg1 to obtain the generated relationship pairs.

Description

technical field [0001] The invention relates to the technical field of word processing, and provides a method for constructing relational database tables. Background technique [0002] In the part of seed bank table construction, we use the BootStrapping algorithm to implement seed expansion. [0003] Bootstrapping algorithm, also known as self-expanding technology, is a machine learning technology widely used in knowledge acquisition. It is a step-by-step semi-supervised learning method that only requires a small number of seeds. Based on this, through training again and again, the seeds are effectively expanded, and finally the required data information scale is reached. [0004] The main steps of the Bootstrapping algorithm [0005] (1) Establish an initial seed set; [0006] (2) According to the seed set, a context pattern with a certain window size is extracted, and a candidate pattern set is established; [0007] (3) Use pattern matching to identify samples and for...

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): G06F16/31G06F16/36G06F16/33
CPCG06F16/31G06F16/367G06F16/3331
Inventor 付立军常志军刘雨江武靖恺李凯张永宏
Owner 北京中科瑞通信息科技有限公司
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