Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

English event co-reference resolution method and system

An event, English technology, applied in the field of event co-reference resolution, can solve problems such as not being a long-term solution, affecting the judgment of whether events are co-reference, unable to extract the global information of event sentences, etc., to achieve the effect of improving the resolution performance.

Active Publication Date: 2019-02-19
SUZHOU UNIV
View PDF7 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] At present, there are two main problems in most event co-reference resolution: 1) In previous studies, probability-based models or graph-based models are usually used [4], these methods usually cannot extract the global information of event sentences, event Partial information between words in a sentence and relatively important parts in an event sentence
However, this method also has a disadvantage. There may be conflicts between event pairs, which will affect the judgment of whether the event pairs refer to each other. Although rules can be imposed to alleviate the interference of this problem, it will not last for a long time after all. However, if the corpus is changed, the rules will also change accordingly

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
  • English event co-reference resolution method and system
  • English event co-reference resolution method and system
  • English event co-reference resolution method and system

Examples

Experimental program
Comparison scheme
Effect test

example

[0062] S101. Read the event annotation information in the annotation document according to the tag to obtain an annotation corpus event set. The format of each event sentence instance in the labeled corpus event set is as follows:

[0063]

[0064] Take Example 1 as an example,

example 1

[0065] Example 1: Annotate the document:

[0066]

[0067]

[0068] The conviction of 29-year-old Ali AsadChandia is the government's 11th conviction in its investigation of what it called a ``Virginia jihad network" that used paintball games to train forglobal holy war .

[0069] conviction

[0070]

[0071]

[0072] A Maryland teacher has been convicted of lending aid to a Pakistani terrorist organization.

[0073] convicted

[0074]

[0075]

[0076]

[0077]

[0078] A federal jury convicted Chandia in June of three counts of providing material support to Lashkar-e-Taiba or conspiring to do so and acquired him on a fourth count of supporting terrorists.

[0079] providing

[0080]

[0081]

[0082]Among them, the attribute id of hopper indicates the ID of the same reference, the subtree of hopper is event mention, the attribute id of eventmention is the event sentence ID, type is the type of event sentence, subtype is the type of event ...

example 2

[0084] Example 2:

[0085] S20, for each instance in the marked corpus event set, call the lemma restoration tool to restore the trigger word (trigger), and obtain the trigger word restoration word form trig_lemma; call the data cleaning tool and the part-of-speech tagging tool to clean and mark the event sentence (text), Obtain the new event sentence sentence and part-of-speech information postag; take 3 words in the event sentence before and after each trigger word (trigger) to obtain the event short sentence context; Each word before and after the word has a position subscript to get the word position loc in the sentence; replace the event sentence text in the marked corpus event collection with the new event sentence sentence, and restore the part-of-speech information postag, trigger word to form trig_lemma, and event phrase context and the intra-sentence word position loc are added to the instance of the labeled corpus event set. The format of the annotat...

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 English event co-reference resolution method and system. The English event co-reference resolution method comprises the following steps: reading the event annotation information of the annotation document to obtain the annotation corpus event set; Document ID, identical ID, event sentence ID, event sentence type, event sentence type, event sentence morphology and the starting position of trigger in the tagged corpus event set are statistically analyzed to obtain the event sentence information set. The invention has the advantages that the information features extracted from the event sentences and the event pair matching features are utilized to judge the homonymy relationship between the event pairs by adopting a circulating neural network; The invention adoptsa circulating neural network model and utilizes depth learning to solve the problem of event homonym resolution. Compared with the existing method and system, the method and system of the invention obviously improve the English event homonym resolution performance.

Description

technical field [0001] The invention relates to the field of natural language processing, in particular to an event co-reference resolution method. Background technique [0002] Co-reference resolution is a difficult problem in natural language processing. In recent years, most studies have focused on entity co-referencing, and relatively few studies on event co-referencing, which is still a challenging task. Event co-reference resolution is to link two event sentences in the text to the same event chain on the basis of judging whether they point to the same event in the real world. According to whether the event sentences are in the same document, the task of event co-reference resolution can be divided into intra-document event co-reference resolution and cross-document event co-reference resolution. [0003] At present, compared with the research on entity co-referencing [1], there are few researches on event co-referencing. Event synonymous resolution is evolved from ...

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): G06F16/332G06F16/35
Inventor 程昊熠李培峰朱巧明周国栋朱晓旭
Owner SUZHOU UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com