Event extraction method based on multi-task learning

An event extraction and event technology, applied in computer parts, text database query, unstructured text data retrieval, etc., can solve the problem of ignoring correlation, avoid error transmission and reduce the amount of memory occupied

Active Publication Date: 2022-07-12
NAT UNIV OF DEFENSE TECH
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] Existing event extraction techniques mostly use the method of segmentation extraction, while ignoring the correlation between the subtasks of event extraction

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
  • Event extraction method based on multi-task learning
  • Event extraction method based on multi-task learning
  • Event extraction method based on multi-task learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0062] figure 2 is the overall flow chart of the present invention; such as figure 2 As shown, the present invention comprises the following steps:

[0063] Step 1: Build an event extraction system. The event extraction system consists of client, negative sample database, event screening module, feature extraction module, event classification module, event trigger word and argument extraction module and event information integrator.

[0064] The client is connected with the event screening module and the event information integrator, and sends the event text X input by the user into the event screening module and the event information integrator.

[0065] The negative sample database stores the text collection obtained from the Internet and other channels, including P texts, where P is a positive integer and P>3000, and is connected to the event screening module. The negative sample database is read by the event screening module. The text set in the negative sample databas...

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 an event extraction method based on multi-task learning, and aims to realize quick and efficient event extraction when the number of texts is huge. According to the technical scheme, firstly, an event extraction system composed of a client, a negative sample database, an event screening module, a feature extraction module, an event classification module, an event trigger word and argument extraction module and an event information integrator is constructed. Then selecting a training set, and training the event extraction system by using the training set to obtain network weight parameters; carrying out event screening on the events by adopting the trained event extraction system, judging whether the events are events of predefined categories, and if so, carrying out feature extraction, event classification, event trigger word and argument extraction and event information integration to obtain an event extraction result; and if not, discarding the current event. By the adoption of the method, information implied in data can be fully mined, classification attributes of multiple fields of texts can be rapidly obtained, and the event extraction accuracy and efficiency are improved.

Description

technical field [0001] The invention relates to the field of natural language processing event extraction, in particular to a method for structured processing of event information contained in unstructured text based on multi-task learning, which belongs to an event extraction method. Background technique [0002] Natural language is a language that is deliberately created by humans for some specific purposes, and is the essential feature that distinguishes humans from other animals. Natural language is the crystallization of human wisdom. Most of human knowledge is stored and circulated in language and words. As a language recording tool, words can save and record rich semantic information and characteristic content, helping people record and express what they want. information conveyed. How to make machines recognize and understand the rich semantic features recorded by natural language is the focus of scientists in today's information age, and natural language processing...

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): G06F16/33G06F16/35G06K9/62G06N3/04
CPCG06F16/334G06F16/353G06N3/044G06N3/045G06F18/214Y02D10/00
Inventor 黄震陈一凡刘攀王博阳陈易欣周文博李东升
Owner NAT UNIV OF DEFENSE TECH
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