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

Science and technology news automatic writing system based on deep learning

A technology of deep learning and journalism, applied in neural learning methods, natural language data processing, biological neural network models, etc., can solve the problems of high cost of news reports, improve the generalization learning ability of deep learning, improve collection speed, and high recognition The effect of accuracy

Pending Publication Date: 2020-10-30
郑州市混沌信息技术有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

With the development of the Internet, the occurrence of technological figures and technological events, there are more and more technology-related reports every day, and relatively speaking, there are more and more news reports on technology, which leads to an increase in the cost of news reports. high

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
  • Science and technology news automatic writing system based on deep learning
  • Science and technology news automatic writing system based on deep learning
  • Science and technology news automatic writing system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] Such as Figure 1 to Figure 4 As shown, a technology news automatic writing system based on deep learning is characterized in that it includes the following modules:

[0043] Web crawler module: This module collects technology channels, technology news, and relevant content of each technology website from various websites, extracts the text of the collected data, and stores them in the database;

[0044] Science and technology news preprocessing module: for the collected news, word segmentation, named entity recognition, entity relationship extraction, syntax analysis, semantic analysis;

[0045] Sci-tech news classification and clustering module: Mainly aiming at the content of sci-tech news, it is dedicated to further refinement, using intelligent classification and clustering technology to carry out detailed classification for sci-tech news. The generative memory model based on deep learning is used to train and learn news content, and finally realize a News generat...

Embodiment 2

[0057] Such as Figure 1 to Figure 4 As shown, in order to make the inner side occupy less, this embodiment has made further improvements on the basis of embodiment 1, specifically: the crf++ is implemented in c++ language, and a large number of stl data structures are applied. After reading the source code in depth Based on the stl-related part of the source code, the code is rewritten in c language, which can be expressed as: in the tagger.cpp source code file, use the vector structure:

[0058]

[0059] And there is no release of memory after the feature is encoded, the present invention uses char* instead of std::vector>TaggerImpl::x_, and the forced release of memory immediately obtains 10% by experiment and comparison In addition, it is modified for CRF++ and L-BFGS. The L-BFGS algorithm is an improvement of the quasi-Newton algorithm. Its name already tells us that it is based on the improvement of the quasi-Newton method BFGS algorithm. The basic idea of ​​the L-B...

Embodiment 3

[0061] Such as Figure 1 to Figure 4 As shown, the news named entity recognition module uses the crf++ model to train the corpus and identify the names of people, places, organizations, products, professional terms and time of occurrence of news. The automatic news writing system is characterized in that: the automatic news generation module includes a user interaction module and a news generation module.

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 science and technology news automatic writing system based on deep learning, and relates to the technical field of news writing. The science and technology news automatic writing system includes a web crawler module; a science and technology news preprocessing module; a science and technology news classification clustering module; a science and technology news deep learning generation training module; an automatic news generation module; and a generated news display module. Rapid generation of science and technology news is achieved, and news forms of different stylescan be generated according to different website styles and the like.

Description

technical field [0001] The invention relates to the technical field of news writing, and is used for information processing and press release writing of scientific and technological news, and more specifically relates to an automatic writing system of scientific and technological news based on deep learning. Background technique [0002] There are many categories of news works, such as people's livelihood, current affairs, military affairs, etc. These can be seen in various columns or pages of newspapers, and Internet news is gradually increasing, and various news websites emerge in endlessly. [0003] The science and technology news is the report of the latest and unique scientific and technological facts. Since most of the science and technology news is conference news, the materials are mostly conference drafts and related reports, and there are few special interviews, so the materials are very important. For science and technology news, the requirements for reporters 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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F40/186G06F40/216G06F40/247G06F40/295G06F40/30G06F16/906G06F16/951G06N3/08
CPCG06F40/186G06F40/216G06F40/247G06F16/951G06F16/906G06F40/295G06F40/30G06N3/08
Inventor 刘超刘霖雯
Owner 郑州市混沌信息技术有限公司
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