Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Self-adaptive architecture semantic distribution text understanding method and self-adaptive architecture semantic distribution text understanding system

An adaptive classification and self-adaptive technology, applied in semantic analysis, natural language data processing, unstructured text data retrieval, etc., can solve problems such as poor performance of classification models

Active Publication Date: 2020-10-23
前海企保科技(深圳)有限公司
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this case, classification models such as pre-training tend to perform poorly

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
  • Self-adaptive architecture semantic distribution text understanding method and self-adaptive architecture semantic distribution text understanding system
  • Self-adaptive architecture semantic distribution text understanding method and self-adaptive architecture semantic distribution text understanding system
  • Self-adaptive architecture semantic distribution text understanding method and self-adaptive architecture semantic distribution text understanding system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] In order to facilitate those of ordinary skill in the art to understand and implement the present invention, the embodiments of the present invention will be further described below in conjunction with the accompanying drawings.

[0043] refer to Figure 1 to Figure 3 , the present invention provides an adaptive architecture semantic distribution text understanding method, comprising steps:

[0044] S1. Encode the input text through the text basic encoder module to obtain text feature representation;

[0045] S2. The text feature representation is processed by a multi-head attention mechanism through the semantic association key distribution representation module to form a self-attention representation of the text feature representation;

[0046] S3. Using the architecture-adaptive classification network module to automatically select a classification network for the self-attention representation of the text feature representation, to achieve the final classification. ...

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 relates to an adaptive architecture semantic distribution text understanding method and an adaptive architecture semantic distribution text understanding system. The adaptive architecture semantic distribution text understanding system comprises the following main modules: a text basic encoder, a semantic association key distribution representation module and an architecture adaptiveclassification network module. According to the method, context multi-head attention semantic distribution representation in the semantic association key distribution representation module is utilized, so that context information can be better utilized, and the ambiguity problem of context representation is eliminated; meanwhile, key information such as keywords and phrases can be extracted through multi-head attention semantic distribution representation of the key information, and the problem that the key information in the subdivision field is sensitive to classification tasks is solved. And the architecture self-adaptive classification network module can select a classification network layer in a self-adaptive manner according to the similarity of classification categories, so that the similar classifications share weight parameters, the classifications with large classification category differences do not share weights, and the problem that the classification categories cannot becompletely independent can be better solved.

Description

technical field [0001] The invention relates to the field of classification and processing of natural language texts, in particular to a text understanding method and system for adaptive framework semantic distribution. Background technique [0002] In the Internet era of information explosion, any website needs to process a large amount of text data, and it is essential to classify and structure the information, and manual processing has completely failed to meet the efficiency requirements of the Internet age. As a natural language processing The most basic task of text classification is important and indispensable. The automatic classification of text can summarize and concentrate a large amount of information, further structure the text information, save manpower and time costs, increase the attention to target information, and thus improve the speed of information processing and decision-making efficiency. [0003] Common applications of text classification include new...

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): G06F40/289G06F40/30G06F16/33G06F16/35G06K9/62G06N3/04G06N3/08
CPCG06F40/289G06F40/30G06F16/3344G06F16/35G06N3/08G06N3/047G06N3/045G06F18/2415G06F18/241
Inventor 庞文君杨猛黄志青
Owner 前海企保科技(深圳)有限公司
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
Eureka Blog
Learn More
PatSnap group products