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

Topic model-based public opinion news event tracking method

A public opinion and news technology, applied in the field of knowledge discovery, can solve the problems that the news topic detection and tracking technology does not consider the characteristics of public opinion news, and cannot effectively track public opinion news events, etc.

Pending Publication Date: 2020-12-22
YUNNAN UNIV
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] In order to solve the problems that the known news topic detection and tracking technology does not consider the characteristics of public opinion news and cannot effectively track public opinion news events, the present invention provides a public opinion topic based on the characteristics of public opinion news, such as sparseness, timing, and easy evolution. The construction method of the model MBTM, through which the hidden theme of public opinion news data is inferred

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
  • Topic model-based public opinion news event tracking method
  • Topic model-based public opinion news event tracking method
  • Topic model-based public opinion news event tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0078] Example: Analysis of the tracking process of the public opinion news event of "Express Robot Entering the Campus".

[0079] Online acquisition and preprocessing of public opinion news data

[0080] 1.1: Data online acquisition

[0081] First, continue to crawl the news data related to the "Express Robot Entering the Campus" event through the news webpage and social media, and the total number of news data obtained M 50,000,000, and the time span of news releases is from January 9, 2019 to October 23, 2019.

[0082] Table 1 Some data examples corresponding to the theme of "Express Robot Entering Campus"

[0083]

[0084] 1.2: Preprocessing

[0085] 1.2.1: Obtain the Chinese stop word set Stop_words suitable for news classification from https: / / github.com / goto456 / stopwords.

[0086] Obtain the Chinese word segmentation component jieba from https: / / github.com / fxsjy / jieba, and perform word segmentation processing on the text content of public opinion news data. B...

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 topic model-based public opinion news event tracking method. Aiming at the characteristics of sparsity, time sequence, evolutionary easiness and the like of public opinion news, the invention provides a construction method of a minority biterm topic model (MBTM). Then, the method comprises the following steps: using the MBTM for modeling public opinion news data, deducingimplicit topics, further setting time nodes, sampling topic word distribution on each time node, acquiring lexical items in the corresponding topics. mining a topic evolution track by utilizing two indexes of JS divergence and topic strength, and completing online tracking of a specific public opinion news event. In the MBTM construction process, the approximate deduction method based on the random updating thought is used for updating the parameters, so that public opinion news event tracking can be accurately and efficiently achieved, and reference is provided for the news event tracking technology in the specific field.

Description

technical field [0001] The invention discloses a method for tracking public opinion news events based on a theme model, which involves inferring implicit themes of public opinion news events based on a theme model, mining public opinion news topic evolution tracks by using JS divergence and theme strength, and realizing public opinion news events The method of tracking belongs to the field of knowledge discovery. Background technique [0002] News topic detection and tracking technology is a research hotspot in the field of information processing. Its task is to take the information flow of news media as the research object in the early stage of event development. Organize news related to a certain topic and present it to users in the form of news threads. With the rapid development of social media, using topic models to track topic evolution in social media text content has become a research hotspot in recent years. Applications have an important guiding role. There are ...

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): G06F16/951G06F16/9535G06F40/289G06F40/242G06F40/216G06Q50/00
CPCG06F16/951G06F16/9535G06Q50/01
Inventor 马子娟岳昆段亮吴鑫然李维华赵天资
Owner YUNNAN UNIV
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