Graph convolution-based relationship extraction method

A technology of relational extraction and convolution, applied in the field of relational extraction based on graph convolution, which can solve problems such as inaccurate features, lack of natural language processing tools, and method limitations

Inactive Publication Date: 2021-09-28
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Statistical learning methods based on feature engineering and kernel functions have considerable shortcomings in model scalability. At the same time, the extraction of these artificially designed features relies on natural language processing tools, and the process of feature extraction is also a pipeline process. , the result of the previous step of natural language processing is used as the input of the next step, so these natural language processing tools are prone to error accumulation and transmission, making the extracted features inaccurate
At the same time, when facing small languages, the lack of relevant natural language processing tools makes the above methods subject to greater limitations

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
  • Graph convolution-based relationship extraction method
  • Graph convolution-based relationship extraction method
  • Graph convolution-based relationship extraction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0065] Input data set: Enter the Linguistic Data Consortium official website to download the ACE 2005 data set. The data under the ACE2005 corpus folder includes Arabic, English and Chinese, and there are multiple data sources in each language.

[0066] With the help of natural language analysis tools, the original sentence in the data set is segmented to obtain the word segmentation result of the original sentence, so that the sentence after the word segmentation is obtained X =[ X 1 ,… X n ], using entity recognition tools to perform entity recognition on the original sentence, and the obtained entities are called subject entities and object entities according to the order in which they appear in the original sentence; with the help of natural language analysis tools, the original sentences in the data set are subjected to dependency syntax analysis , each word is represented as a node, and the semantic dependencies between words are used as the edges between the correspo...

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 graph convolution-based relationship extraction method, which comprises the following steps of: language analysis preprocessing: performing word segmentation and dependency syntactic analysis on an original sentence in a data set by means of a natural language analysis tool to obtain a word segmentation result of the original sentence, and constructing a dependency syntactic tree for representing a semantic dependency relationship between words in the original sentence, generating an adjacent matrix according to a topological relation between nodes in the dependency syntax tree; querying word vectors: converting each word of the original sentence into a corresponding word vector by querying a word vector table to obtain a vectorized representation of the original sentence; feature extraction through the graph convolutional neural network: inputting the adjacent matrix and the vectorized representation of each word into the graph convolutional neural network, and performing learning to obtain feature representation; and relation classification: splicing the feature representations and then sending the spliced feature representations into a learning neural network to obtain a final representation, then obtaining probability distribution of the entity pairs on each relation according to the feature representations, and predicting the relation with the maximum probability, namely the relation type of the subject entity and the object entity in the sentence predicted by the model.

Description

technical field [0001] The invention relates to the field of text data relation extraction, in particular to a relation extraction method based on graph convolution. Background technique [0002] In the era of information explosion, a large amount of text data emerges on the Internet every day, such as news reports, blogs, research documents, and social media comments, etc. How to quickly and effectively mine valuable information from these massive text data has become a Challenges that need to be addressed. Relation extraction is to identify the semantic relationship between named entities for a given text sentence and the marked named entities. [0003] Existing relationship extraction techniques generally feature sentences and words near entities as the input features of the model. After a series of processing, an overall representation is obtained. Finally, the relationship classification probability is obtained after a trained classifier. [0004] Disadvantages of exi...

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): G06F16/33G06F16/35G06F40/279G06N3/04G06N3/08
CPCG06F16/3344G06F16/35G06F40/279G06N3/08G06N3/044G06N3/045
Inventor 陶建华张华张大伟杨国花刘通
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
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