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

Judicial public opinion text abstracting method fusing theme information

A technology of subject information and text, applied in the field of natural language processing, can solve problems such as unsatisfactory summary effect

Active Publication Date: 2020-03-24
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention provides a method for summarizing judicial public opinion texts that integrates subject information, so as to solve the problem that the traditional sequence-to-sequence model has unsatisfactory summary effects on judicial public opinion news without domain knowledge supervision

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
  • Judicial public opinion text abstracting method fusing theme information
  • Judicial public opinion text abstracting method fusing theme information
  • Judicial public opinion text abstracting method fusing theme information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0088] Embodiment 1: as Figure 1-4 As shown, a method for summarizing judicial public opinion texts that fuse subject information, the specific steps of the method for summarizing judicial public opinion texts that fuse subject information are as follows:

[0089] Step1, collect data from the Internet, process the data, and construct a judicial public opinion text abstract data set; each text in the judicial public opinion text abstract data set corresponds to a standard abstract Y; news texts and corresponding abstract samples are shown in Table 1 :

[0090] Table 1

[0091]

[0092] The statistics of the processed training set, development set and test set are shown in Table 2. The invention trains the model on the training set, uses the development set to adjust the hyperparameters of the model, and finally verifies the validity of the model on the test table.

[0093] Table 2 Dataset Statistics

[0094]

[0095] Step2. Based on the judicial public opinion text 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 relates to a judicial public opinion text abstracting method fusing theme information. The method comprises the steps of firstly constructing a judicial public opinion text abstract dataset; utilizing a topic model to generate a judicial public opinion topic knowledge base containing topic words and document topic probability distribution; and on the basis of an attention-based encoding and decoding model, providing a judicial public opinion text abstract model based on topic information constraint. On one hand, theme words are encoded to obtain theme vectors related to news texts, and the theme vectors serve as external knowledge to be fused into the decoding process to guide abstract generation; on the other hand, theme distribution of the generated abstract is obtained through calculation, and a pre-trained news text theme distribution optimization model is utilized. According to the method, the topic information is used as external knowledge, the decoding process isguided more effectively, the text abstract with higher quality is generated, a referential method is provided for developing judicial public opinion text abstract research integrated with domain knowledge, and meanwhile strong support is provided for timely, effective and rapid judicial public opinion emergency disposal.

Description

technical field [0001] The invention relates to a method for summarizing judicial public opinion texts fused with subject information, and belongs to the technical field of natural language processing. Background technique [0002] Judicial public opinion news focuses on social hot issues such as social justice and corruption. Some sensitive events can easily trigger a social hot topic in a short period of time, leading to a viral outbreak of public opinion. Therefore, how to guide and deal with judicial public opinion in a timely and effective manner and eliminate negative effects is crucial to social stability. The text summarization task can generate short, fluent, and readable text summaries for long judicial public opinion news documents, which is conducive to understanding judicial public opinion news and provides technical support for fast and effective public opinion disposal. Contents of the invention [0003] The present invention provides a method for summarizi...

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): G06F16/34G06F16/31
CPCG06F16/345G06F16/313
Inventor 余正涛黄于欣相艳郭军军线岩团于志强
Owner KUNMING UNIV OF SCI & TECH
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