CN111401052A - Semantic understanding-based multilingual text matching method and system - Google Patents
Semantic understanding-based multilingual text matching method and system Download PDFInfo
- Publication number
- CN111401052A CN111401052A CN202010330830.6A CN202010330830A CN111401052A CN 111401052 A CN111401052 A CN 111401052A CN 202010330830 A CN202010330830 A CN 202010330830A CN 111401052 A CN111401052 A CN 111401052A
- Authority
- CN
- China
- Prior art keywords
- text
- multilingual
- words
- semantic
- preprocessed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Machine Translation (AREA)
Abstract
The invention discloses a multilingual text matching method and system based on semantic understanding, which comprises text processing, semantic representation learning, substituted word meaning analysis, substituted word translation and combined output, wherein the input text is processed, punctuations, symbols and special characters in the text are removed, the text is separated by taking words as units to obtain a preprocessed text, a deep learning model is created in a translator, a semantic analysis model is implanted in the translator, a multilingual grammar detection tool is used for compiling the text words which are translated to obtain the context semantics according to the corresponding languages, the grammar of the corresponding languages of the sentences is detected to ensure that the sentences accord with the grammar of the corresponding languages, the semantic representation of source language words can be directly utilized to find words of similar target languages, the matching is directly carried out, and the huge calculation amount caused by word matching after the translation is reduced, the semantic accuracy of the multi-language text matching is improved, and errors in the multi-language text matching are reduced.
Description
Technical Field
The invention relates to the technical field of text data processing, in particular to a multilingual text matching method and system based on semantic understanding.
Background
With the rapid development of the internet and the further deepening of the globalization trend, the network data shows the explosive growth, the big data era has come, a large amount of multilingual text data are programmed in the network text, and simultaneously, with the rise of a translation system, a large amount of texts are translated into other languages, so that the generation of the multilingual text data is promoted;
the existing multilingual text matching method mainly translates Chinese into corresponding foreign words and then matches the foreign words through machine translation, but the problems exist in the way that translation is possible to be wrong, translated words can be limited on a limited number of words, matching cannot be accurately, effectively and completely carried out, and translated foreign words are easy to be substituted into sentences, and then semantic deviation occurs.
Disclosure of Invention
The technical scheme provided by the invention can effectively solve the problems that the translation provided by the background technology is possible to have errors, the translated words can only be limited on a limited number of words, the matching cannot be accurately, effectively and completely carried out, and the translated foreign words are easy to have semantic deviation after being substituted into sentences.
In order to achieve the purpose, the invention provides the following technical scheme: the multilingual text matching method based on semantic understanding comprises the following steps:
s1, text processing: processing the input text, removing punctuations, symbols and special characters in the text, and separating the text by taking words as units to obtain a preprocessed text;
s2, semantic representation learning: creating a deep learning model in a translator, and implanting a semantic analysis model in the translator;
s3, substituting word meaning analysis: inputting the preprocessed text into a translator, and performing context semantic analysis on the input text by using a semantic analysis model;
substitution of S4 into word translation: translating the preprocessed text to obtain a multilingual text set which accords with the preprocessed text;
s5, combined output: and arranging and outputting the obtained multilingual text set according to the language grammar to obtain the multilingual text conforming to the semantics.
According to the above technical features, in S1, the input text is processed to remove punctuation, symbols, and special characters in the text, and to remove invalid characters in the input text, the text is divided in units of words, and the input text is converted into a text sequence in units of words, so as to obtain a preprocessed text.
According to the technical characteristics, in S2, a deep learning model is created in the translator, and a semantic analysis model is implanted in the translator, where the deep learning model is specifically a convolutional neural network model, and the semantic analysis model is specifically BM 25.
According to the technical characteristics, in S3, the preprocessed text is input into the translator, the words in the preprocessed text are respectively substituted into the semantic analysis model for semantic analysis, and the input words are subjected to context semantic analysis to obtain multilingual words that best meet the context semantic.
According to the above technical features, the S4 includes:
s41, translation: substituting the preprocessed text into a translation machine to translate by taking words as units;
s42, semantic analysis learning: the semantic representation of each word in the substituted preprocessed text in context is learned using a deep learning model.
According to the above technical features, in S41, the preprocessed text is substituted into the translator and translated in units of words to obtain a multilingual word set corresponding to the words in the preprocessed text, in S42, when the preprocessed text is substituted into the translator and translated, the deep learning model is used to learn the semantic representation of each word in the substituted preprocessed text in context, and according to the semantic representation of each word, words of similar target languages are found from the multilingual words that best meet the context semantics to obtain the multilingual text set meeting the semantics.
According to the above technical features, in S5, the obtained multilingual text set is arranged and output according to the corresponding language grammar, so that the output text conforms to the grammar of the corresponding language, and a multilingual text conforming to semantics is obtained.
According to the technical characteristics, the multilingual text matching method system based on semantic understanding comprises a text processing module, a semantic analysis module, a text translation module and a multilingual arrangement module, wherein the text processing module comprises a Chinese word segmentation model, the semantic analysis module comprises a semantic analysis model and a deep learning model, the text translation module comprises a translator, the multilingual arrangement module comprises a multilingual grammar detection tool, the output end of the text processing module is connected with the input end of the semantic analysis module, the output end of the semantic analysis module is connected with the input end of the text translation module, and the output end of the text translation module is connected with the multilingual arrangement module.
According to the technical characteristics, the text processing comprises a Chinese word segmentation model, the Chinese word segmentation model is used for removing punctuation, symbols and special symbols from an input text to obtain pure Chinese text words, the semantic analysis module comprises a semantic analysis model and a deep learning model, the semantic analysis model is used for analyzing the input text according to context semantics corresponding to the words to obtain a multilingual word set which best accords with the context semantics, and the deep learning model is used for learning semantic representation of the text translated according to the words substituted into the translation machine, so that words expressing similar meanings in different languages and different contexts are similar in semantic representation.
According to the above technical features, the multilingual arrangement module includes a multilingual grammar checking tool for compiling text words translated to conform to the context semantics in the corresponding language, and checking the grammar of the corresponding language of the obtained sentence to conform to the grammar of the corresponding language.
Compared with the prior art, the invention has the beneficial effects that: the invention has scientific and reasonable structure and safe and convenient use:
by processing an input text, punctuation, symbols and special characters in the text are removed, invalid characters in the input text are removed, the text is separated by taking words as units, the input text is converted into a text sequence by taking the words as units, a preprocessed text is obtained, a deep learning model based on a convolutional neural network model is established in a translation machine, a semantic analysis model based on BM25 is implanted in the translation machine, the situation that the Chinese text word segmentation is difficult due to the fact that the Chinese text contains the punctuation, the symbols and the special characters is avoided, and the accuracy degree of the Chinese text word segmentation is increased;
inputting the preprocessed text into a translator, respectively substituting the words in the preprocessed text into a semantic analysis model for semantic analysis, performing context semantic analysis on the input words to obtain multi-language words which best meet the context semantic, and then substituting the preprocessed text into the translator for translation by taking the words as units to obtain a multi-language word set corresponding to the words in the preprocessed text;
the method comprises the steps of learning semantic representation of each word in a substituted preprocessed text in context by using a deep learning model, finding out words of a similar target language from multi-language words which best accord with context semantics according to the semantic representation of each word, obtaining a multi-language text set which accords with semantics, arranging and outputting the obtained multi-language text set according to corresponding language grammar, enabling the output text to accord with the grammar of the corresponding language, obtaining the multi-language text which accords with the semantics, finding out the words of the similar target language by using the semantic representation of source language words, directly matching, not performing single matching after translation, improving the matching degree of the multi-language text, and increasing the matching accuracy of the multi-language text;
punctuation, symbols and special symbols of an input text are removed through a Chinese word segmentation model to obtain pure Chinese text words, the input text is analyzed according to context semantics corresponding to the words by using a semantic analysis model to obtain a multilingual word set which best accords with the context semantics, invalid characters appearing during Chinese text word segmentation are reduced, and the purity degree of the Chinese text word segmentation is improved;
the method comprises the steps of translating input texts according to words by using a translator, learning semantic representation of texts substituted into the translator according to words by using a deep learning model, enabling words expressing similar meanings in different languages and different contexts to be similar in semantic representation, compiling text words according with context semantics by using a multi-language grammar detection tool, detecting grammars of corresponding languages of sentences to obtain grammars of corresponding languages, enabling the grammars to be in line with the grammars of the corresponding languages, directly utilizing semantic representation of source language words to find words of similar target languages, directly matching, reducing huge calculation amount brought by word matching after translation, increasing semantic accuracy of multi-language text matching, and reducing errors occurring when the multi-language texts are matched.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
In the drawings:
FIG. 1 is a schematic diagram of the process configuration of the present invention;
FIG. 2 is a schematic structural view of S41 and S42 of the present invention;
fig. 3 is a schematic diagram of the system architecture of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Example 1: as shown in fig. 1-2, the present invention provides a technical solution, a multilingual text matching method based on semantic understanding, comprising the following steps:
s1, text processing: processing the input text, removing punctuations, symbols and special characters in the text, and separating the text by taking words as units to obtain a preprocessed text;
s2, semantic representation learning: creating a deep learning model in a translator, and implanting a semantic analysis model in the translator;
s3, substituting word meaning analysis: inputting the preprocessed text into a translator, and performing context semantic analysis on the input text by using a semantic analysis model;
substitution of S4 into word translation: translating the preprocessed text to obtain a multilingual text set which accords with the preprocessed text;
s5, combined output: and arranging and outputting the obtained multilingual text set according to the language grammar to obtain the multilingual text conforming to the semantics.
According to the technical features, in S1, the input text is processed to remove punctuation, symbols and special characters in the text, and remove invalid characters in the input text, the text is divided in units of words, and the input text is converted into a text sequence in units of words to obtain a preprocessed text.
According to the technical characteristics, in S2, a deep learning model is created in the translator, and a semantic analysis model is implanted in the translator, wherein the deep learning model is specifically a convolutional neural network model, and the semantic analysis model is specifically BM 25.
According to the above technical features, in S3, the preprocessed text is input into the translator, the words in the preprocessed text are respectively substituted into the semantic analysis model for semantic analysis, and the input words are subjected to context semantic analysis to obtain multilingual words that best meet the context semantic.
According to the above technical features, S4 includes:
s41, translation: substituting the preprocessed text into a translation machine to translate by taking words as units;
s42, semantic analysis learning: the semantic representation of each word in the substituted preprocessed text in context is learned using a deep learning model.
According to the above technical features, in S41, the preprocessed text is substituted into the translator and translated in units of words to obtain a multilingual word set corresponding to the words in the preprocessed text, in S42, when the preprocessed text is substituted into the translator and translated, the deep learning model is used to learn the semantic representation of each word in the substituted preprocessed text in context, and according to the semantic representation of each word, words of a similar target language are found from the multilingual words that best fit the context semantics to obtain a multilingual text set that fits the semantics.
According to the above technical features, in S5, the obtained multilingual text set is arranged and output according to the corresponding language grammar, so that the output text conforms to the grammar of the corresponding language, and a multilingual text conforming to semantics is obtained.
Example 2: as shown in FIG. 3, the present invention provides a technical solution, a system for matching a multilingual text based on semantic understanding, comprising a text processing module, a semantic analysis module, a text translation module and a multilingual arrangement module, wherein the text processing module comprises a Chinese word segmentation module, the semantic analysis module comprises a semantic analysis module and a deep learning module, the text translation module comprises a translator, the multilingual arrangement module comprises a multilingual grammar detection tool, an output end of the text processing module is connected with an input end of the semantic analysis module, an output end of the semantic analysis module is connected with an input end of the text translation module, and an output end of the text translation module is connected with the multilingual arrangement module.
According to the technical characteristics, the text processing comprises a Chinese word segmentation model, the Chinese word segmentation model is used for removing punctuation, symbols and special symbols from an input text to obtain pure Chinese text words, the semantic analysis module comprises a semantic analysis model and a deep learning model, the semantic analysis model is used for analyzing the input text according to context semantics corresponding to the words to obtain a multilingual word set which best accords with the context semantics, and the deep learning model is used for learning semantic representation of the text translated according to the words substituted into the translation machine, so that words expressing similar meanings in different languages and different contexts are similar in semantic representation.
According to the above technical features, the multilingual arrangement module includes a multilingual grammar checking tool for compiling text words translated to conform to the context semantics in the corresponding language, and checking the grammar of the corresponding language of the obtained sentence to conform to the grammar of the corresponding language.
The working principle and the using process of the invention are as follows: when the multilingual text matching method based on semantic understanding is used, firstly, an input text is processed, punctuations, symbols and special characters in the text are removed, invalid characters in the input text are removed, the text is separated by taking words as units, the input text is converted into a text sequence by taking the words as units, a preprocessed text is obtained, meanwhile, a deep learning model based on a convolutional neural network model is established in a translation machine, and a semantic analysis model based on BM25 is implanted in the translation machine;
inputting the preprocessed text into a translator, respectively substituting the words in the preprocessed text into a semantic analysis model for semantic analysis, performing context semantic analysis on the input words to obtain multi-language words which best meet the context semantic, and then substituting the preprocessed text into the translator for translation by taking the words as units to obtain a multi-language word set corresponding to the words in the preprocessed text;
learning semantic representation of each word in the substituted preprocessed text in the context by using a deep learning model, finding out words of a similar target language from multi-language words which most accord with the context semantics according to the obtained semantic representation of each word to obtain a multi-language text set which accords with the semantics, and finally arranging and outputting the obtained multi-language text set according to corresponding language grammar so that the output text accords with the grammar of the corresponding language to obtain the multi-language text which accords with the semantics;
when the system is used, firstly, punctuation, symbols and special symbols of an input text are removed through a Chinese word segmentation model to obtain pure Chinese text words, the input text is analyzed according to context semantics corresponding to the words by using a semantic analysis model to obtain a multilingual word set which best accords with the context semantics, then, the input text is translated according to the words by using a translator, meanwhile, a deep learning model is used for learning the semantic representation of the text which is substituted into the translator according to the words translation, so that different languages and words which express similar meanings in different contexts are similar in semantic representation, finally, a multilingual grammar detection tool is used for compiling the text words which accord with the context semantics according to the corresponding languages, and the grammar of the language corresponding to the sentence is detected, so that it conforms to the grammar of the corresponding language.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The multilingual text matching method based on semantic understanding is characterized by comprising the following steps of:
s1, text processing: processing the input text, removing punctuations, symbols and special characters in the text, and separating the text by taking words as units to obtain a preprocessed text;
s2, semantic representation learning: creating a deep learning model in a translator, and implanting a semantic analysis model in the translator;
s3, substituting word meaning analysis: inputting the preprocessed text into a translator, and performing context semantic analysis on the input text by using a semantic analysis model;
substitution of S4 into word translation: translating the preprocessed text to obtain a multilingual text set which accords with the preprocessed text;
s5, combined output: and arranging and outputting the obtained multilingual text set according to the language grammar to obtain the multilingual text conforming to the semantics.
2. The semantic understanding-based multilingual text matching method according to claim 1, wherein in S1, the input text is processed to remove punctuation, symbols and special characters from the text and to remove invalid characters from the input text, the text is separated in units of words, and the input text is converted into a text sequence in units of words to obtain a preprocessed text.
3. The semantic understanding-based multilingual text matching method of claim 1, wherein in S2, a deep learning model, specifically a convolutional neural network model, is created in the translator, and a semantic analysis model, specifically BM25, is embedded in the translator.
4. The semantic understanding-based multilingual text matching method of claim 1, wherein in S3, the preprocessed text is inputted into a translation engine, the words in the preprocessed text are respectively substituted into a semantic analysis model for semantic analysis, and the inputted words are subjected to context semantic analysis to obtain multilingual words that best meet the context semantic.
5. The semantic understanding-based multilingual text-matching method of claim 1, wherein the S4 comprises:
s41, translation: substituting the preprocessed text into a translation machine to translate by taking words as units;
s42, semantic analysis learning: the semantic representation of each word in the substituted preprocessed text in context is learned using a deep learning model.
6. The semantic understanding-based multilingual text matching method of claim 5, wherein in S41, the preprocessed text is translated into the translator in units of words to obtain a multilingual word set corresponding to the words in the preprocessed text, and in S42, when the preprocessed text is translated into the translator, a deep learning model is used to learn the semantic representation of each word in the preprocessed text in context, and words of a similar target language are found from the multilingual words that best satisfy the context semantic meaning according to the semantic representation of each word obtained to obtain the multilingual text set satisfying the semantic meaning.
7. The semantic understanding-based multilingual text matching method according to claim 1, wherein in S5, the obtained multilingual text sets are output in a corresponding language grammar arrangement such that the output text conforms to the grammar of the corresponding language, resulting in semantically conforming multilingual text.
8. The semantic understanding-based multilingual text matching method system of any of claims 1-7, comprising a text processing module, a semantic analysis module, a text translation module, and a multilingual arrangement module, wherein the text processing module comprises a chinese word segmentation model, the semantic analysis module comprises a semantic analysis model and a deep learning model, the text translation module comprises a translator, the multilingual arrangement module comprises a multilingual grammar detection tool, an output of the text processing module is connected to an input of the semantic analysis module, an output of the semantic analysis module is connected to an input of the text translation module, and an output of the text translation module is connected to the multilingual arrangement module.
9. The method and system for matching multilingual texts based on semantic understanding according to claim 8, wherein the text processing includes a chinese word segmentation model for de-punctuation, sign and special sign processing of the input text to obtain pure chinese text words, the semantic analysis module includes a semantic analysis model for analyzing the input text according to context semantics corresponding to the words to obtain a set of multilingual words that best meet the context semantics, and a deep learning model for learning semantic representations of the text translated according to the words substituted into the translator, so that words in different languages and in different contexts expressing similar meanings are similar in semantic representation.
10. The semantic understanding-based multilingual text matching method and system according to claim 8, wherein the multilingual ranking module comprises a multilingual grammar checking tool for compiling text words translated to conform to the context semantics in the corresponding language, and checking the grammar of the corresponding language of the resulting sentence to conform to the grammar of the corresponding language.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010330830.6A CN111401052A (en) | 2020-04-24 | 2020-04-24 | Semantic understanding-based multilingual text matching method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010330830.6A CN111401052A (en) | 2020-04-24 | 2020-04-24 | Semantic understanding-based multilingual text matching method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111401052A true CN111401052A (en) | 2020-07-10 |
Family
ID=71429783
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010330830.6A Pending CN111401052A (en) | 2020-04-24 | 2020-04-24 | Semantic understanding-based multilingual text matching method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111401052A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066589A (en) * | 2017-04-17 | 2017-08-18 | 河南工业大学 | A kind of sort method and device of Entity Semantics and word frequency based on comprehensive knowledge |
| CN107066455A (en) * | 2017-03-30 | 2017-08-18 | 唐亮 | A kind of multilingual intelligence pretreatment real-time statistics machine translation system |
| CN107423290A (en) * | 2017-04-19 | 2017-12-01 | 厦门大学 | A kind of neural network machine translation model based on hierarchical structure |
| CN108345585A (en) * | 2018-01-11 | 2018-07-31 | 浙江大学 | A kind of automatic question-answering method based on deep learning |
| CN108806671A (en) * | 2018-05-29 | 2018-11-13 | 杭州认识科技有限公司 | Semantic analysis, device and electronic equipment |
| CN109492232A (en) * | 2018-10-22 | 2019-03-19 | 内蒙古工业大学 | A kind of illiteracy Chinese machine translation method of the enhancing semantic feature information based on Transformer |
| CN110826345A (en) * | 2019-11-14 | 2020-02-21 | 北京香侬慧语科技有限责任公司 | Machine translation method and device |
-
2020
- 2020-04-24 CN CN202010330830.6A patent/CN111401052A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107066455A (en) * | 2017-03-30 | 2017-08-18 | 唐亮 | A kind of multilingual intelligence pretreatment real-time statistics machine translation system |
| CN107066589A (en) * | 2017-04-17 | 2017-08-18 | 河南工业大学 | A kind of sort method and device of Entity Semantics and word frequency based on comprehensive knowledge |
| CN107423290A (en) * | 2017-04-19 | 2017-12-01 | 厦门大学 | A kind of neural network machine translation model based on hierarchical structure |
| CN108345585A (en) * | 2018-01-11 | 2018-07-31 | 浙江大学 | A kind of automatic question-answering method based on deep learning |
| CN108806671A (en) * | 2018-05-29 | 2018-11-13 | 杭州认识科技有限公司 | Semantic analysis, device and electronic equipment |
| CN109492232A (en) * | 2018-10-22 | 2019-03-19 | 内蒙古工业大学 | A kind of illiteracy Chinese machine translation method of the enhancing semantic feature information based on Transformer |
| CN110826345A (en) * | 2019-11-14 | 2020-02-21 | 北京香侬慧语科技有限责任公司 | Machine translation method and device |
Non-Patent Citations (2)
| Title |
|---|
| 郑晓东;胡汉辉;赵林度;吕永涛;: "一种基于词对齐的中文深层语义解析模型", 计算机工程与应用, no. 20, pages 8 - 13 * |
| 黄河燕, 李渝生: "上下文相关汉语自动分词及词法预处理算法", 应用科学学报, no. 02, pages 25 - 32 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4961755B2 (en) | Word alignment device, word alignment method, word alignment program | |
| CN107066455B (en) | Multi-language intelligent preprocessing real-time statistics machine translation system | |
| US8239188B2 (en) | Example based translation apparatus, translation method, and translation program | |
| WO2016010245A1 (en) | Method and system for robust tagging of named entities in the presence of source or translation errors | |
| WO2010046782A2 (en) | Hybrid machine translation | |
| KR20120021933A (en) | Statistical machine translation method using dependency forest | |
| Xu et al. | Do we need Chinese word segmentation for statistical machine translation? | |
| CN106372053B (en) | Syntactic analysis method and device | |
| KR20090061158A (en) | Method and apparatus for correcting of translation error by using error-correction pattern in a translation system | |
| Simard | Building and using parallel text for translation | |
| Liyanapathirana et al. | Sinspell: A comprehensive spelling checker for sinhala | |
| CN112257460A (en) | Pivot-based Hanyue combined training neural machine translation method | |
| Schottmüller et al. | Issues in translating verb-particle constructions from german to english | |
| Sinhal et al. | Machine translation approaches and design aspects | |
| CN114996387A (en) | Index data-based natural language processing method and system | |
| CN111401052A (en) | Semantic understanding-based multilingual text matching method and system | |
| CN114861628A (en) | System, method, electronic device and storage medium for training machine translation model | |
| CN107870905B (en) | Method for identifying specific vocabulary | |
| Tongtep et al. | Multi-stage automatic NE and pos annotation using pattern-based and statistical-based techniques for thai corpus construction | |
| Doermann et al. | Translation lexicon acquisition from bilingual dictionaries | |
| Amini et al. | Central Kurdish machine translation: First large scale parallel corpus and experiments | |
| CN111597827A (en) | Method and device for improving machine translation accuracy | |
| Sánchez-Martínez et al. | Using alignment templates to infer shallow-transfer machine translation rules | |
| Ghaffar et al. | English to arabic statistical machine translation system improvements using preprocessing and arabic morphology analysis | |
| CN116522966B (en) | Text translation method and system based on multilingual vocabulary entry |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |