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

Cause and effect reason map construction method based on integration of multiple neural networks

A neural network and construction method technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as lack of logic mining

Active Publication Date: 2020-10-13
QINGDAO UNIV
View PDF4 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, its main research is the relationship between entities and entities, and there is still a lack of logic mining between events and events.

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
  • Cause and effect reason map construction method based on integration of multiple neural networks
  • Cause and effect reason map construction method based on integration of multiple neural networks
  • Cause and effect reason map construction method based on integration of multiple neural networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0050] In this embodiment, the process of constructing a causal event map for hot topics includes the following steps:

[0051] Step 1: Use the open source Scrapy framework to crawl the data released by the Internet platform. The crawled content is the title, content and time, and the hot topic text data set is stored in descending order of time;

[0052] Step 2: According to step 1, use the unsupervised learning kmeans algorithm to perform text clustering analysis on the hot topic text data set obtained;

[0053] Step 3: Define the constituent elements of the event, and then use the data source obtained in step 2 to label the text data with the BIO sequence labeling system;

[0054] Step 4: According to the data marked in step 3, use the BERT model space vectorization data source, and then combine the Bi-LSTM+Attention+CRF model to extract event elements;

[0055] Step 5: Construct candidate event pairs according to the event extraction results of step 4;

[00...

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 belongs to the technical field of network information. The invention relates to a cause and effect reason map construction method based on integration of multiple neural networks. The method comprises the following steps of: marking data of an obtained corpus by adopting a BIO sequence marking system and segmenting the data; performing event extraction by using a BERT + Bi-LSTM + Attention + CRF model, combining events based on event extraction elements (O, V) and defining event pairs and rule features during event relationship extraction, and extracting causal relationships in combination with the rule features between the events and the Bi-GRU model; carrying out similarity calculation based on event pairs; selecting two events with the highest scores to be combined into a(event i, similarity, event j) triple; extracting triples of combinations (reason events, cause and effect, and result events) on the basis of event relationships; then, adopting a Neo4j graph database for persistent storage, building a reason logic knowledge base is built to construct a cause and effect reason graph for hot topics, the constructed cause and effect reason graph can deeply extractsemantic information, and real-time control over hot events by related supervision departments and personal users is facilitated.

Description

Technical field: [0001] The invention belongs to the field of network information technology, and relates to a method for constructing a causal event map based on the integration of various neural networks. Background technique: [0002] Since Hinton published an article on deep learning in Natures, a new wave of artificial intelligence has followed. In 2012, CNN won the ImageNet championship. In 2014, the DeppID algorithm model of the Hong Kong Chinese Lab surpassed the human face recognition rate for the first time. In 2016, AlphaGo defeated human high-level Go players. In 2018, Google released Cloud AutoML. But then the development of artificial intelligence entered the next stage, allowing machines to learn and master human knowledge. For example, understand the common sense knowledge of human beings that they will be "not hungry" after they are "full". The research on implicit consumption intention based on deep learning can let the machine know that the "marriage" ev...

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): G06F16/36G06F16/35G06F40/211G06F40/289G06F40/30G06K9/62G06N3/04G06N3/08
CPCG06F16/367G06F16/35G06N3/08G06N3/045G06F18/23213Y02A90/10
Inventor 云红艳胡欢云洋李正民
Owner QINGDAO 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