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

Legal instrument reading model and construction method

A construction method and document technology, applied in the field of document reading, can solve problems such as inability to give answers to preset questions, inability to deal with multiple complex types, and unanswerable questions

Inactive Publication Date: 2021-02-05
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] However, for a judgment document, researchers may ask some questions that can be directly answered in the document, such as the sentence, crime location, etc., or may ask questions that require inference to give answers, such as whether there is a gang crime, etc.
At the same time, some instruments may not be able to give answers to preset questions, that is, unanswerable questions
The traditional machine reading comprehension model of fragment extraction cannot deal with many complex types of problems; therefore, a legal document reading model is needed to solve the above problems

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
  • Legal instrument reading model and construction method
  • Legal instrument reading model and construction method
  • Legal instrument reading model and construction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] Such as figure 1 As shown, the present embodiment provides a method for constructing a legal document reading model, which includes the following steps:

[0060] 1. The Bert layer encodes the input chapters and questions; this layer uses the Bert-Chinese implementation proposed by Google;

[0061] 2. Some previous studies have shown that adding some prior features related to words can improve the performance of the model to a certain extent, which is also applicable to the scene of legal judgment documents. Adding named entity recognition vectors will help the model recognize the names of criminals, crime locations, and criminal gang names; adding part-of-speech vectors will help the model recognize some entity words, quantifiers, etc.; therefore, after obtaining the semantic encoding vector, the feature fusion layer Fusion of part-of-speech tagging and named entity tagging vectors;

[0062] 3. The modeling layer models the fragment extraction prediction and type judg...

Embodiment 2

[0105]In this embodiment, two benchmarks are set: BIDAF and Bert, which are tested together with the model LegalSelfReader proposed in this embodiment.

[0106] lab environment

[0107] Experiment on a machine with 64-bit Windows system. The external storage space of the machine is 930GB, the memory space is 48GB, the CPU type is single-core Intel i7-8700K, the GPU type is NVIDA GeForceGTX 1080Ti, and the GPU size is 11GB. All experimental programs in this embodiment are written in python language, and the deep learning framework used is Pytorch, and the version number is 1.13.0.

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 the technical field of document reading, in particular to a legal document reading model and a construction method, and the method comprises the following steps that 1, a Bertlayer codes input chapters and questions; 2, the feature fusion layer fuses part-of-speech tagging and named entity tagging vectors; 3, a modeling layer carries out fragment extraction prediction andnon-type judgment modeling; and 4, an output layer outputs fragment prediction and right and wrong prediction. According to the method, three types of problems of fragment extraction, non-judgment and rejection can be better handled.

Description

technical field [0001] The present invention relates to the technical field of document reading, in particular to a legal document reading model and a construction method. Background technique [0002] The application of artificial intelligence technology to the legal field can speed up and improve the legal research process, reduce the time cost and funds of legal research, which makes legal intelligence research a very promising field. Katz pointed out in his 2012 research that with the rapid development of artificial intelligence, traditional legal tasks such as generating legal documents and predicting case outcomes will usher in changes. This change can also be glimpsed from three other aspects. First, speech recognition technology was used to record court proceedings. Second, use machine learning methods to assist lawyers in reviewing legal documents. Furthermore, some machine learning methods have also been applied to build intelligent referee systems [5,6]. [00...

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): G06F40/216G06F40/295G06F40/30G06F40/126G06F16/35G06N3/04G06Q50/18
CPCG06F40/216G06F40/295G06F40/30G06F40/126G06F16/35G06N3/049G06Q50/18G06N3/047G06N3/045
Inventor 张引胡刚杜锦阳刘铨张可
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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