A method to automatically correct parts of text - judged by Chinese parts of speech

Abstract

The present invention mainly relates to the judgment and correction of the three characters of "de", "get" and "di". After the translator completes the translation, this method will automatically check the "de" and "de" and "di" used in the translator's manuscript. According to the rules, if it is used incorrectly, it will be automatically corrected to the correct "de" or "get" or "land". According to the method provided by the present invention, at first detect all the sentences that contain "', "get" and "地" in the document, and judge whether it belongs to a content word or a function word according to the word segmentation method; when it belongs to a content word, directly skip it; otherwise, according to Relevant rules carry out the correction and judgment of "de", "get" and "land". By adopting the invention, the expression accuracy of translated documents can be improved, and the problem of low efficiency of manual verification in the prior art can be avoided.

Description

technical field [0001] The invention belongs to the technical field of document processing, and in particular relates to a method for automatically correcting part of text. Background technique [0002] In Chinese grammar, when "de", "de" and "di" are used as function words, they are important bridges connecting words and characters in the article, and are widely used in various written expressions. However, many people do not clearly know the difference between the three, and usually use them interchangeably, or even replace them all with "的". [0003] From the perspective of grammatical research, "Grammar and Rhetorical Speeches" written by Lu Shuxiang and Zhu Dexi believes that "de" has too many concurrent jobs and the burden is too heavy, and advocates a strict division of labor in "de, place, and acquisition". Many modern Chinese treatises and textbooks since the 1950s generally hold this view. [0004] Judging from the use of written language, the division of labor b...

AI Technical Summary

Problems solved by technology

[0007] Although this usage rule is well known by linguists, for translators, they need to give the translation results of Chinese sentences before using the above rules. In the case of a huge amount of document translation, it is unrealistic to manually screen them one by one; In addition, most translators are not researchers who strictly grasp grammatical rules, and they may pay more attention to the substantive content itself, rather than the importance of these three words; more importantly, "的,地,得 "The above situation will only occur when making function words. The "de, place, and get" that are part of the content words do not need to consider the above problems. Therefore, simply using the three-character search method will only increase the workload

[0008] In addition, even if the above-mentioned usage rules of "de, place, and de" are known, in the Chinese corpus, the definition of the so-called attributive / adverbial / complement cannot be accurately identified because the computer cannot strictly define its characteristics, so it is impossible to directly Realize the above-mentioned judging method by computer flow

Images