An Automatic Identification Method of Controversial Focus Based on Hierarchical Attention Neural Network Model

A neural network model and automatic recognition technology, applied in biological neural network models, neural learning methods, unstructured text data retrieval, etc., can solve problems such as unfavorable large-scale expansion, time-consuming and laborious construction process, and achieve good scalability. , the effect of reducing manual work and reducing the burden on users

Active Publication Date: 2022-05-03
ZHEJIANG UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] However, on the one hand, in order to automatically identify the focus of disputes in documents, it is necessary to establish a focus of dispute system for each field, but the construction of the focus of dispute system relies on domain experts, and the construction process is time-consuming and laborious, which is not conducive to large-scale expansion; on the other hand, in legal documents Among them, different words and sentences contain different levels of "information" and have different effects on judging the focus of disputes. When building an automatic identification model for the focus of disputes, it is necessary to pay attention to the role played by different words and sentences

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
  • An Automatic Identification Method of Controversial Focus Based on Hierarchical Attention Neural Network Model
  • An Automatic Identification Method of Controversial Focus Based on Hierarchical Attention Neural Network Model
  • An Automatic Identification Method of Controversial Focus Based on Hierarchical Attention Neural Network Model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0082] Below in conjunction with the method of the present invention describe in detail the concrete steps that this embodiment implements, as follows:

[0083] In this embodiment, the method of the present invention is applied to court judgment documents in the field of commercial housing sales disputes, and the focus of disputes in the documents is automatically identified.

[0084] 1) Use regular expressions to process a total of about 336,000 judgment documents, and extract the court's summary and expression of the disputed focus of the case. Among them, there are more than 15,000 documents containing expressions of the focus of disputes. From these 15,000 documents, a total of 6,418 non-repetitive sentences expressing the focus of disputes can be obtained. First, the TF-IDF algorithm is used to vectorize the text. First, preprocess the text by cutting words and removing stop words, and then build a word bag space. Read all the documents into the program, and follow the...

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 method for automatically identifying the focus of disputes based on a hierarchical attention neural network model. First of all, extract the court's summary of the dispute focus sentence from the documents containing the court's summary of the dispute focus, and use the hierarchical clustering method to construct the dispute focus system. Then use the dispute focus system to label each document with multiple different categories, construct a data set, and transform the problem of identifying focus of dispute into a multi-label and multi-category problem. Afterwards, the hierarchical attention neural network model is trained to focus more attention on important words, sentences, and paragraphs that contain more information to form a dispute focus recognizer. Finally, the text that needs to identify the focus of dispute is input into the focus of dispute recognizer to obtain the focus of dispute of the input text. The method of the invention has a high prediction accuracy rate, can accurately identify and judge the dispute focus of documents, and has good scalability.

Description

technical field [0001] The invention relates to an automatic identification method for dispute focus based on a layered attention neural network model. Background technique [0002] Legal service is a traditional industry, but it is also an industry with great potential. In order to improve the efficiency of legal services, reform the traditional form of legal services, use artificial intelligence technology to assist in identifying the focus of disputes in documents, so as to help people judge and understand the focus of disputes in cases faster and better. [0003] However, on the one hand, in order to automatically identify the focus of disputes in documents, it is necessary to establish a focus of dispute system for each field, but the construction of the focus of dispute system relies on domain experts, and the construction process is time-consuming and laborious, which is not conducive to large-scale expansion; on the other hand, in legal documents In , different word...

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): G06F16/35G06F40/30G06N3/08
CPCG06F16/35G06N3/08
Inventor 鲁伟明贾程皓庄越挺
Owner ZHEJIANG 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