Unlock instant, AI-driven research and patent intelligence for your innovation.

Chinese text error correction method based on seq2seq + attention

A text error correction, Chinese technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as gradient disappearance and gradient explosion, achieve strong fitting ability, good effect, and reduce manual workload.

Inactive Publication Date: 2019-04-12
WUHAN UNIV
View PDF4 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

RNN will encounter great difficulties when dealing with long-term dependencies (nodes far away in time series), because the calculation of the connection between nodes far away will involve multiple multiplications of the Jacobian matrix, which will bring The problem of gradient disappearance or gradient explosion

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
  • Chinese text error correction method based on seq2seq + attention
  • Chinese text error correction method based on seq2seq + attention
  • Chinese text error correction method based on seq2seq + attention

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049] During specific implementation, the technical solution provided by the present invention can be realized by those skilled in the art by using computer software technology to realize the automatic operation process. The technical solution of the present invention will be described in detail below in conjunction with the drawings and embodiments.

[0050] Step 1: Text Preprocessing

[0051] Use relevant tools in python to read the maintenance records in the database, extract all the content in the document file, and then use regular expressions to perform Chinese sentence segmentation operations, and store the results in the text file, each line corresponds to a sentence, and manually The correct text of the annotation is stored in another text file, which is in one-to-one correspondence with the original file. The proprietary symbols in the electric power communication field are recorded, and the common Chinese character list together constitutes the character lis...

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 Chinese text error correction method based on seq2seq + attention, belonging to the research field of data quality and relating to the technical fields of RNN, bidirectionalRNN, LSTM, seq2seq, attention mechanism and the like. Aiming at communication equipment maintenance records, a seq2seq + attention neural network model is constructed, an Adam optimization method isadopted for model training, and a trained model is used for carrying out error correction tasks. The neural network model used in the method can be widely applied to text error correction in other fields, and redesign of the model is avoided to a certain extent.

Description

technical field [0001] The invention belongs to the technical field of Chinese text error correction, in particular to the field of error correction of communication equipment maintenance records generated in a power communication management system. Background technique [0002] The main research objects, key technologies and practical application values ​​involved in this field mainly include: [0003] Power communication management system: It is a dedicated power communication network system that is an important support for smart grids. It is a "two-level deployment" of headquarters and provincial companies, and a communication management system for "four-level applications" of headquarters, branches, provincial companies, and city and county companies. SG-TMS". Through standardized project construction and vigorous promotion of the practical application of the system, "SG-TMS" has been deeply integrated into the daily work of tens of thousands of power communication prof...

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): G06F17/27G06F17/22G06N3/04G06N3/08
CPCG06N3/08G06F40/126G06F40/232G06N3/044G06N3/045
Inventor 李石君邓永康杨济海余伟余放李宇轩
Owner WUHAN UNIV