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

Method and system for automatically constructing affair graph

An automatic construction and map technology, applied in the field of event map, can solve the problems of error cascade, low accuracy rate, backward technology, etc., and achieve the effect of avoiding error cascade, improving extraction accuracy, and improving accuracy

Active Publication Date: 2022-08-09
云孚科技(北京)有限公司
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] 1. The technology is backward, and the latest achievements of deep learning have not been applied;
[0008] 2. Event extraction and event causality extraction are carried out in stages, and there are common problems such as error cascade, low accuracy, and slow speed.

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
  • Method and system for automatically constructing affair graph
  • Method and system for automatically constructing affair graph
  • Method and system for automatically constructing affair graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0117] The invention discloses an automatic construction method of an event map. figure 1 It is a flowchart of a method for automatically constructing an event map according to an embodiment of the present invention, such as figure 1 and figure 2 As shown, the method includes:

[0118] Step S1, using the text encoding model of the trigger word enhancement based on the Transformer class pre-training model transformation to carry out text encoding on the sentence to obtain the semantic vector of the sentence;

[0119] Step S2, using a unified character-to-multiple labeling method to label the event labeling sequence, the event causality header labeling sequence and the event causality tail labeling sequence, and then according to the labelled event labeling sequence, event causality header labeling sequence and event causality Tail label sequence, get event set and causal event pair set;

[0120] Step S3, constructing an event pair set through the event set, and then applyin...

Embodiment 2

[0195] Example 2: Next, take a text set composed of two sentences as an example to construct an example of an event graph, such as Figure 7 shown. Given a set of texts: {"Over-issue of currency will cause inflation", "The reason for rising house prices is attributed to the release of currency"}.

[0196] Extract the event {"money over-issuance", "inflation"} from the text "money over-issuance will cause inflation", and extract the causal event pair {("money over-issuance", "inflation")}; from From the text "The reason for the rise in house prices is attributed to the release of money", the event {"the rise in house prices", "the rise in money"} is extracted, and the causal event pair {("the rise in money", "the rise in house prices")} is extracted.

[0197] Through the event aggregation module, the nodes with highly similar events "currency over-issuance" and "currency release" are merged.

[0198] Finally, an event map is constructed, showing that the event of "money over-...

Embodiment 3

[0200] The invention discloses an automatic construction system of an event map. Figure 8 is a structural diagram of a system for automatically constructing an event map according to an embodiment of the present invention; such as Figure 8 As shown, the system 100 includes:

[0201] The first processing module 101 is configured to use the text encoding model of trigger word enhancement based on the Transformer class pre-training model to carry out text encoding on the sentence to obtain the semantic vector of the sentence;

[0202] The second processing module 102 is configured to use a unified character-to-multiple labeling method to label the event labeling sequence, the event causality header labeling sequence and the event causality tail labeling sequence, and then according to the labelled event labeling sequence and event causality The head annotation sequence and the event causality tail annotation sequence are used to obtain an event set and a causal event pair set;...

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 provides a method and a system for automatically constructing a affair graph. The method comprises the following steps: carrying out text coding on sentences by adopting a trigger word enhanced text coding model transformed based on a deep learning network to obtain semantic vectors of the sentences; labeling an event labeling sequence, an event causal relationship head labeling sequence and an event causal relationship tail labeling sequence by adopting a uniform character pair multi-head labeling mode, and obtaining an event set and a causal relationship event pair set; applying an event vector learning method based on comparative learning to the event pair set to obtain semantic vectors of event pairs, and obtaining the similarity degree of every two events; and according to the similarity between every two events, aggregating the events in the initial version of the affair graph to obtain the affair graph of the final version. According to the scheme provided by the invention, the extraction accuracy of the event and the causal relationship thereof can be greatly improved; and the accuracy of event similarity calculation can be greatly improved.

Description

technical field [0001] The invention belongs to the field of event maps, and in particular relates to a method and system for automatically constructing event maps. Background technique [0002] Events are a very important concept in human society, and many activities in human society are often driven by events. The evolution law between events is a kind of very valuable knowledge, and it is of great significance to mine this kind of logical knowledge for us to understand the law of the development and change of human society. Event Logic Graph (ELG for short) is an event logic knowledge base that describes the evolutionary laws and patterns between events, including the relationships between events, such as inheritance, causality, conditions, and upper and lower positions, and is used to describe and record The objective evolution law of human behavior activities and events. Structurally, an event graph is a directed cyclic graph, in which nodes represent events and direc...

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): G06F16/36G06F40/126G06F40/211G06F40/284G06F40/30G06N3/04G06N3/08
CPCG06F16/367G06F40/126G06F40/211G06F40/30G06F40/284G06N3/08G06N3/045
Inventor 张文斌曾俊瑀贾显伏程尧周建行辛洁
Owner 云孚科技(北京)有限公司
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