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

A case prediction agent training method and system capable of automatic updating

An automatic update and training method technology, applied in the field of artificial intelligence, can solve the problems of retraining update, prediction model accuracy drop, complexity, etc., and achieve the effect of expanding the sample space

Active Publication Date: 2022-01-28
JINAN UNIVERSITY
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the technical problems of the decline in the accuracy of the prediction model in the prior art, retraining and updating are relatively complicated, the present invention proposes a case prediction agent training method and system that can be automatically updated

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
  • A case prediction agent training method and system capable of automatic updating
  • A case prediction agent training method and system capable of automatic updating
  • A case prediction agent training method and system capable of automatic updating

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] For the technical features of the present invention, the objects and effects more clearly understood, reference now be described specific embodiments of the present invention the control of.

[0031] Reinforcement Learning of the most fundamental characteristics of the disadvantages that it requires the agent to explore the environment, to explore the environmental movement made to obtain the reward to update the model parameters, which happens to be able to make up the judicial system use cases to predict the depth of learning design , which is able to go up on the basis of existing data sets to explore the sights and state does not appear, and then be able to adapt to changes in judicial cases of social changes caused by, changes over time to avoid the problem of declining model accuracy, therefore, for justice predict aspects of the issue in the case, the use of reinforcement learning method to solve with high research value.

[0032] Cases can be automatically updated a...

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 discloses a case prediction intelligent body training method and system capable of automatic updating. The method includes obtaining a certain number of judicial case trial documents as samples, extracting key information such as facts from the trial documents, processing the data, and receiving New data marked by experts, construct a new data set, model for judicial trial issues, obtain the accuracy calculation model for key information, extract the parameters required for training from the model, define the reward function for model training, and then use The BCQ algorithm trains the agent model according to parameters and data sets. The invention solves the problem of inaccurate prediction of judicial cases by using the reinforcement learning method, and can improve and optimize at the same time, and maintain the accuracy rate of the model for a long time.

Description

Technical field [0001] The present invention relates to the field of artificial intelligence, primarily with respect to application of artificial intelligence in the judicial process, in particular, a case can be automatically updated anticipation training methods and systems Agent. Background technique [0002] With the legal system continues to rely on social perfected reality, the number of judicial cases has gradually increased, resulting in lower courts around the heavy workload, under immense pressure. For occupy most civil cases, contract disputes are simple and reproducible characteristics, but because the number is too large, occupying a large number of judicial cases of judicial resources, resulting in a backlog of cases across the country too frequently failed to effectively deal with the problem, often one cases need to wait several months before the trial, the efficiency of the judiciary and thus also decreased. [0003] Most common prediction model based on the dept...

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 Patents(China)
IPC IPC(8): G06F16/33G06N3/04G06N3/08G06Q50/18
CPCG06F16/3344G06N3/08G06Q50/18G06N3/045
Inventor 郭洪飞戴源志曾云辉何智慧任亚平张锐
Owner JINAN UNIVERSITY
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