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

Text classification method based on CNN and Bi-GRU

A text classification and text technology, applied in neural learning methods, semantic analysis, instruments, etc., can solve problems such as limiting the accuracy of text classification, gradient explosion, gradient disappearance, etc., to achieve good classification effect, solid theoretical foundation, and wide application Effect

Inactive Publication Date: 2020-05-12
NO 30 INST OF CHINA ELECTRONIC TECH GRP CORP
View PDF1 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) are widely used in natural language processing, but due to the contextual dependencies in the structure of natural language, only relying on Convolutional Neural Networks to achieve text classification will ignore the context of words Meaning, and the traditional cyclic neural network has the problem of gradient disappearance or gradient explosion, which limits the accuracy of text classification

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
  • Text classification method based on CNN and Bi-GRU
  • Text classification method based on CNN and Bi-GRU
  • Text classification method based on CNN and Bi-GRU

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] A text classification method based on CNN and Bi-GRU, the framework is as follows figure 1 shown. This method mainly obtains two abstract high-level feature expressions of text through two neural network structures of convolutional neural network and bidirectional GRU cyclic neural network, and uses a classifier to classify text through feature fusion.

[0020] Include the following steps:

[0021] 1) Model the text with a multi-angle convolutional neural network, including different filter types and pooling types, remove the last layer of softmax layer, and obtain the feature expression of local hidden information. Specific steps are as follows:

[0022] 1.1) Establish two different types of filters, one is an overall filter, which is a filter that matches the entire word vector, and the other is a single-dimensional filter, which is to match on each dimension of a word vector; assuming a sentence Input ∈ R length×Dim is a sequence of length words, each word is rep...

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 text classification method based on CNN and Bi-GRU, and the method comprises the following steps: 1, carrying out the convolutional neural network modeling of text data, andobtaining a first text feature expression containing local implicit information; 2, performing Bi-GRU neural network modeling on the text data to obtain a second text feature expression containing sequence information of the whole sentence in two directions; and 3, performing feature fusion on the two text feature expressions obtained in the step 1 and the step 2, and performing classification byusing an LSSVM classifier. According to the method, not only are sentence local features and context semantic information captured, but also more diversified and richer feature expressions of the textare obtained by fusing two different text feature expressions, and the classification accuracy is further improved.

Description

technical field [0001] The invention relates to a text classification method based on CNN and Bi-GRU. Background technique [0002] Text classification technology is an important basis for information retrieval and text mining, and its main task is to determine its category according to its content under a given set of category tags. Text classification has a wide range of applications in natural language processing and understanding, information organization and management, content information filtering and other fields. There are many commonly used methods for text classification, such as unsupervised methods based on dictionaries and rules, and supervised methods based on machine learning. The dictionary-based method uses authoritative dictionaries and artificially constructs features according to experience. The accuracy of the model is high, but the recall rate of the model is low due to the low coverage of the dictionary. Supervised methods based on machine learning,...

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): G06F40/211G06F40/30G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/2411
Inventor 姬少培颜亮董贵山刘栋
Owner NO 30 INST OF CHINA ELECTRONIC TECH GRP CORP
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