Natural language automatic prediction inference method with interpretation generation

A predictive inference, natural language technology, applied in natural language data processing, semantic analysis, instrumentation, etc., can solve problems such as lack of premise and hypothetical interaction information, difficulty in capturing and generating explanations containing correct logical semantics, etc.

Active Publication Date: 2020-07-24
NANJING UNIV
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since logical information exists implicitly in premise-hypothesis text pairs, it is difficult for existing discriminative explanation generators to effectively capture and generate explanations containing correct logical semantics
[0008] In addition, although Camburu et al. [Camburu et ...

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
  • Natural language automatic prediction inference method with interpretation generation
  • Natural language automatic prediction inference method with interpretation generation
  • Natural language automatic prediction inference method with interpretation generation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0056] The invention provides a method for interpreting and generating natural language, including the following content:

[0057] First, a variational interpretation generation model based on Transformer [Vaswani et al., 2017] is given. The flow chart is as follows figure 1 shown. The following notation is used to mark the input and output of the generative model based on variational interpretation: the input is the premise text hypothetical text and standard interpretation text where m and n are the maximum lengths of the premise text and the hypothesis text respectively, is the mth word in the premise text, is the nth word in the premise text, is the uth word of the standard interpretation text; the output is the generated interpretation text and the logical relation l, where is the uth word in the generated interpretation text. Such as figure 1 As shown, the model is divided into the following steps:

[0058] Step 1-1, splicing the premise text and hypo...

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 natural language automatic prediction inference method with interpretation generation. The natural language automatic prediction inference method comprises the following steps: 1, training a variational interpretation generation model; 2, training an interpretation-based logic relationship prediction model; and 3, explaining and generating natural language inference through the variational explaining production model, and carrying out logic relationship prediction through the generated explaining.

Description

technical field [0001] The invention belongs to the field of natural language processing, in particular to an automatic prediction and inference method of natural language with explanation generation. Background technique [0002] With the continuous development of artificial intelligence technology, all areas of social production and life are undergoing tremendous changes. As one of the basic technologies for machines to understand human language, natural language inference has an increasing impact on all aspects of people's lives. Natural language inference refers to judging the logical relationship between the hypothetical text and it - implication, irrelevance or contradiction - given the premise text information. This technology supports many applications in the upper-level text field, and can help search engines locate content containing user search information in document collections, filter and screen information that is irrelevant or contradictory to search keyword...

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): G06F16/332G06F16/33G06F40/289G06F40/30G06F40/211
CPCG06F16/3329G06F16/3344
Inventor 戴新宇成臻何亮张建兵黄书剑陈家骏
Owner NANJING UNIV
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