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

A positioning method for near-character errors in text

A positioning method and a near-word technology, which can be used in text database query, unstructured text data retrieval, electronic digital data processing, etc., and can solve problems such as speed problem, word segmentation result accuracy, text-like near-word error, etc.

Active Publication Date: 2021-08-20
KUNMING UNIV OF SCI & TECH
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a positioning method for text shape and near-word errors, which is used to solve the speed problem and the accuracy of word segmentation results caused by word segmentation during text error detection, eliminating the need for word segmentation and calculation The time consumed by probability can quickly locate the error position in the text, paving the way for the next proofreading work

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
  • A positioning method for near-character errors in text
  • A positioning method for near-character errors in text

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0022] Embodiment 1: as Figure 1-2 As shown in Fig. 1, a method for locating errors in the form of text, firstly divide the long sentence into n short sentences with a length of m, and then use the Chinese character form library to find the corresponding form of each word in each short sentence Words, and form a candidate word vector with the original character, use the commonly used character library to remove the uncommon words in the vector, and form a candidate matrix with the candidate word vectors of all words, so as to obtain the candidate word matrix of each short sentence; The adjacent vectors in a matrix are bundled into words, and the correct words combined are added to the word set w, and the vectors that cannot be combined into words are extracted and the stop words are added to the stop word set d; Extract the words in the connecting part of a short sentence and combine them. If there are words, add them to the set w; finally, compare the words in the set w and ...

Embodiment 2

[0040] Embodiment 2: a kind of positioning method for text shape near word error, the concrete steps of described method are as follows:

[0041] Step1. Establish a database, which includes font library X, corpus Y, commonly used font library Q, and disabled thesaurus T.

[0042] Step2, select the sample sentence A to be processed, such as 'I can't believe my eyes. ’ The wrong character is sunny (eye).

[0043] Step3. Sentence A is preprocessed, and the punctuation marks in the sentence are removed to obtain a new character string. B='I can't believe my eyes' n=11 is the length of character string B.

[0044] Step4. Divide B='I can't believe my eyes' with length m=5, g={n / m}, (n / m) means the smallest integer not less than this number, then g=3 , then L=[L 1 L 2 L 3 ]=['I can't believe it','Believe in my own eyes','Sunny'],L 1 , L 2 length 5, L 3 has a length of 1.

[0045] Step5, find out L respectively 1 , L 2 , L 3 The candidate word vector matrix, such as L 1 T...

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 method for locating near-character errors in text, and belongs to the technical field of natural language processing. First, the long sentence is divided into multiple short sentences, and then the Chinese character near-character library is used to find out the corresponding character in each short sentence, and the candidate word vector is formed with the original character, and the common character library is used to convert the character in the vector Uncommon words are eliminated, and the candidate word vectors of all words are combined into a candidate matrix to obtain a candidate word matrix for each short sentence; secondly, the adjacent vectors in each candidate matrix are bound into words, and the correct words formed are added to In the word set, the vectors that cannot be combined into words are extracted and added to the stop word set; then the words in the connecting part of two adjacent short sentences are extracted and combined, if there is Words are added to the word set; finally, the words in the word set and the stop word set are compared with the original text, and these words are removed, and the rest are the positions where the wrong words exist.

Description

technical field [0001] The invention relates to a method for locating near-character errors in text, and belongs to the technical field of natural language processing. Background technique [0002] At present, due to the application of OCR text recognition technology, when translating paper texts into computer texts, some texts are often misrecognized and recognized as other characters, and most of these characters are close characters of the original characters. In a large amount of text proofreading, it is a prerequisite for text proofreading to quickly find out the position of typos in the text. [0003] Using N-gram to locate the error position in the text through the connection strength of the context is a common method for text error detection and proofreading. Word segmentation is a prerequisite for using N-gram, but for word segmentation, the accuracy of word segmentation has a great impact on the error detection of text. It plays a decisive role. Word segmentation ...

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 Patents(China)
IPC IPC(8): G06F40/247G06F40/211G06F16/33
CPCG06F40/211G06F40/247
Inventor 邵玉斌王林坪龙华杜庆治
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