CN113807105B - French place name machine translation method - Google Patents
French place name machine translation method Download PDFInfo
- Publication number
- CN113807105B CN113807105B CN202111122788.XA CN202111122788A CN113807105B CN 113807105 B CN113807105 B CN 113807105B CN 202111122788 A CN202111122788 A CN 202111122788A CN 113807105 B CN113807105 B CN 113807105B
- Authority
- CN
- China
- Prior art keywords
- name
- place name
- french
- place
- transliteration
- 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.)
- Active
Links
- 238000013519 translation Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000000354 decomposition reaction Methods 0.000 claims description 22
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000014616 translation Effects 0.000 description 28
- 230000006870 function Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
- G06F40/44—Statistical methods, e.g. probability models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
- G06F40/47—Machine-assisted translation, e.g. using translation memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/58—Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/53—Processing of non-Latin text
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a French place name machine translation method, which comprises the following steps: firstly, distinguishing the preprocessed French place name group to obtain a place name full name part and a place name special name part; secondly, translating the place name transliteration part according to the letter combination in the legal Chinese transliteration table to obtain a transliteration result; then translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result; and finally, integrating the special name transliteration result and the full name transliteration result to obtain a place name translation result. The technical scheme disclosed by the invention realizes machine translation, reduces the manpower consumption during the generation of the place names, and improves the efficiency of translating French place names.
Description
Technical Field
The invention relates to the technical field of place name translation, in particular to a French place name machine translation method.
Background
Place name translation refers to the conversion of the expression of a geographic entity in one language into the expression in another language. In general, place names are divided into a place name common name part and a place name special name part, wherein the place name common name part is a general word summarizing a certain place feature commonality (category) and plays a qualitative role; the place name part refers to a special word which is used for distinguishing the similar places and is used for positioning a certain geographic entity. Automatic place name translation or machine place name translation is part of the named entity translation in machine translation. However, this section is the most difficult one of all named entity translations. First, the place name composition is complex, and place name compositions under different scales have great differences. Secondly, the translation of the place name needs to consider two parts of the full name and the special name, namely the transliteration and the intention translation, so that the translation problem of the place name cannot be independently solved by more general machine translations, and the conventional French place name translation is mainly performed manually, so that the problem of low efficiency exists.
Disclosure of Invention
The invention aims to provide a French place name machine translation method so as to improve the efficiency of French place name translation.
To achieve the above object, the present invention provides a machine translation method for french names, the method comprising:
acquiring a legal place name group to be translated;
preprocessing the French place name group;
distinguishing based on the preprocessed French place name group to obtain a place name through part and a place name special part;
translating the place name special name part according to the letter combination in the legal Chinese transliteration table to obtain a special name transliteration result;
translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result;
and integrating the special name transliteration result and the full name transliteration result to obtain a place name translation result.
Optionally, the distinguishing based on the preprocessed french place name group to obtain a place name common name part and a place name special name part specifically includes:
determining a place name through name template based on the preprocessed Laplace name phrase stored in the place name corpus;
determining all the common name structure decomposition schemes according to each place name common name template;
calculating the corresponding logarithmic frequency of each place name through name template;
summing the logarithmic frequencies in the same common name structure decomposition scheme to obtain a logarithmic frequency sum corresponding to each common name structure decomposition scheme;
taking the common name structure decomposition scheme with the maximum log frequency sum as the place name structure tree;
taking cotyledon nodes of the place name structure tree as the place name special name part; and taking the non-cotyledon node of the place name structure tree as the place name through name part.
Optionally, the determining the place name through name template based on the preprocessed french place name group stored in the place name corpus specifically includes:
by means ofCalculating mutual information of any ordered word pairs in the place name corpus; wherein P is a Representing the frequency of occurrence of the pretreated French place name group a in place name corpus, P b Representing the frequency of occurrence of the pretreated French place name group b in place name corpus, P ab Representing co-occurrence frequency, MI, between pre-processed French place name groups a and b ab Mutual information representing ordered word pairs (a, b); the pretreated French place name group a and the pretreated French place name group b form ordered word pairs (a, b);
storing ordered word pairs with co-occurrence frequency greater than a first set value and mutual information greater than a second set value in the place name corpus into an ordered word pair library;
traversing all sentences, and taking the word groups of the French place names which form pretreatment in each sentence as points on the directed acyclic graph; when two preprocessed French place name groups in sentences belong to ordered word pairs in the ordered word pair library, connecting two points into a line on a directed acyclic graph, and drawing the line into directed edges;
finding out all paths on the directed acyclic graph, and generating a candidate place name template for each path according to the preprocessed French place name group corresponding to the access node; the path is composed of a plurality of directed edges;
and counting the frequency of each candidate place name template, and taking the candidate place name templates with the frequency larger than the set frequency threshold as place name universal name templates.
Optionally, the translating the place name special name part according to the letter combination in the law-chinese transliteration table to obtain a special name transliteration result specifically includes:
performing unsupervised learning on word letters by taking the minimum entropy as a principle to obtain letter combination distribution;
dividing and combining letters in the special name part of the place name by adopting a shortest path method word dividing method to obtain a plurality of letter combinations;
calculating average entropy values of different letter combinations according to the letter combination distribution;
selecting the letter combination with the minimum entropy value as the optimal letter combination;
and translating and writing the optimal letter combination into Chinese characters according to a French transliteration table, so as to obtain a special name transliteration result.
Optionally, the translating the public place name part according to the category of the geographical entity pointed by the French public place name to obtain a public place name transliteration result specifically includes:
translating and analyzing the place name through part by using a Fabry-Perot dictionary and the place name through template to obtain a place name hierarchical grammar structure;
and according to the principle of bottom up, converting the Law place name elements in the place name hierarchical grammar structure into Chinese place name elements layer by layer to obtain a full name transliteration result.
The invention discloses a French place name machine translation method, which comprises the following steps: firstly, distinguishing the preprocessed French place name group to obtain a place name full name part and a place name special name part; secondly, translating the place name transliteration part according to the letter combination in the legal Chinese transliteration table to obtain a transliteration result; then translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result; and finally, integrating the special name transliteration result and the full name transliteration result to obtain a place name translation result. The technical scheme disclosed by the invention realizes machine translation, reduces the manpower consumption during the generation of the place names, and improves the efficiency of translating French place names.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are needed in the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a French place name machine translation method of the present invention;
FIG. 2 is a flow chart showing the distinguishing of a place name part and a place name part according to the present invention;
FIG. 3 is a flowchart of a specific method for determining a place name through name template based on a preprocessed place name phrase stored in a place name corpus;
FIG. 4 is a flowchart of a transliteration of the invention;
fig. 5 is a flowchart of transliteration of the generic name of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a French place name machine translation method so as to improve the efficiency of French place name translation.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
As shown in fig. 1, the invention discloses a french place name machine translation method, which is characterized in that the method comprises the following steps:
step S1: acquiring a legal place name group to be translated; the phrase constituting the french place name is referred to as the french place name phrase.
Step S2: and preprocessing the French place name group.
Step S3: and distinguishing based on the preprocessed French place name group to obtain a place name full name part and a place name special name part.
Step S4: and translating the place name special name part according to the letter combination in the French-Chinese transliteration table to obtain a special name transliteration result.
Step S5: and translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result.
Step S6: and integrating the special name transliteration result and the full name transliteration result to obtain a place name translation result.
The steps are discussed in detail below:
step S2: preprocessing the French place name group; the method has the advantages that the special signs such as the articles, the illegal letters and the punctuation marks which are unfavorable for the machine translation of the place names are removed before the place names and the special names are distinguished, so that the accuracy of distinguishing the place names and the special names in the follow-up process is improved.
As shown in fig. 2-3, step S3: distinguishing based on the preprocessed French place name group to obtain a place name full name part and a place name special name part, wherein the distinguishing comprises the following steps of:
step S31: the method for determining the place name through name template based on the preprocessed Laplace name phrase stored in the place name corpus specifically comprises the following steps:
step S311: calculating mutual information of any ordered word pairs in the place name corpus, wherein the specific formula is as follows:wherein P is a Representing the frequency of occurrence of the pretreated French place name group a in place name corpus, P b Representing the frequency of occurrence of the pretreated French place name group b in place name corpus, P ab Representing co-occurrence frequency, MI, between pre-processed French place name groups a and b ab Mutual information representing ordered word pairs (a, b); pretreated French phrase group a and pretreatedThe french word group b constitutes an ordered word pair (a, b).
Step S312: co-occurrence frequency P in the place name corpus ab Is greater than a first set value T 1 And mutual information MI ab Is greater than the second set value T 2 Is stored in the ordered word pair library G.
Step S313: traversing all sentences, and taking the word groups of the French place names which form pretreatment in each sentence as points on the directed acyclic graph; when two preprocessed French place name groups in a sentence belong to ordered word pairs in the ordered word pair library, connecting two points into a line on a directed acyclic graph, and drawing a directed edge 'a- > b'.
Step S314: finding out all paths on the directed acyclic graph, and generating a candidate place name template for each path according to the preprocessed French place name group corresponding to the access node; the path is made up of a plurality of directed edges. In this embodiment, if adjacent nodes are crossed, a placeholder is inserted.
Step S315: and counting the frequency of each candidate place name template, and taking the candidate place name templates with the frequency larger than the set frequency threshold as place name universal name templates.
Step S32: and determining all the through name structure decomposition schemes according to the through name templates of the place names.
Step S33: and calculating the logarithmic frequency corresponding to each place name through name template.
Step S34: and summing the logarithmic frequencies in the same common name structure decomposition scheme to obtain the logarithmic frequency sum corresponding to each common name structure decomposition scheme.
Step S35: and taking the common name structure decomposition scheme with the maximum log frequency sum as the place name structure tree.
Step S36: taking cotyledon nodes of the place name structure tree as the place name special name part; and taking the non-cotyledon node of the place name structure tree as the place name through name part.
Specifically, given a pre-processed french place name group is input, the place name through name template obtained in the step S31 is utilized to realize the structural decomposition of the pre-processed french place name group so as to generate a place name structural tree. To achieve structural decomposition of the pre-processed French place words, the "projective (projection) hypothesis" of syntactic analysis is consulted. The assumption of the structural decomposition proposed by the present invention is as follows:
1. the preprocessed French place name group is composed of a plurality of place name through name templates, and the templates are not intersected with each other;
2. the placeholder part of the place name template is also a place name through name template;
3. each individual pre-processed french place name group can be viewed as a special place name generic template.
Based on the assumption, the invention provides a place name group structure decomposition method: a series of place name through name templates are found to cover each word in the place name groups without duplication, omission and intersection, so that the sum of the probability logarithms of the place name through name templates is maximum. Specifically, the method can be divided into two steps:
1. scanning out all possible place name template conditions in place name entries;
2. traversing all the combined structural decomposition schemes according to all the possible place name generic name templates, and calculating the logarithmic frequency (i.e. taking the logarithm of the frequency) of each structural decomposition scheme: and calculating the sum of all log probability of the place name through name template in the decomposition scheme to obtain a log frequency value corresponding to each through name structure decomposition scheme, and selecting the structure decomposition scheme with the largest log frequency as a place name structure tree.
In the place name structure tree, based on the assumption of structural decomposition, non-cotyledon nodes are regarded as common place nouns, and cotyledon nodes are regarded as candidate place nouns, so that place name common place special name recognition is achieved in a classification mode.
As shown in fig. 4, step S4: translating the place name special name part according to letter combinations in a legal Chinese transliteration table to obtain a special name transliteration result, wherein the method specifically comprises the following steps of:
step S41: and performing unsupervised learning on word letters by taking the minimum entropy as a principle to obtain letter combination distribution.
Step S42: dividing and combining letters in the special name part of the place name by adopting a shortest path method word dividing method to obtain a plurality of letter combinations; the letter combination division of the invention needs to divide the letter combination as little as possible so as to meet the requirement of transliteration as short as possible.
Step S43: and calculating the average entropy values of different letter combinations according to the letter combination distribution.
Entropy is a measure of the amount of information contained in a variable, and for a random variable x, the probability density function (distribution) is P, then the entropy of the random variable is e= - ≡p (x) log P (x), which in the discrete case may be expressed as e= - Σp (x) log P (x). The minimum entropy principle is to modify the probability density function (distribution) to minimize the entropy E, so as to minimize redundant information contained in the variable.
The invention first defines an average letter entropy e= - Σ x∈Γ P (x) logP (x), where P (x) represents the probability distribution of the occurrence of the letter combinations x in the corpus and Γ represents the set of letter combinations. Calculating the average entropy value for different letter combinations can therefore be expressed as:
wherein P (x) represents the probability distribution of the occurrence of the letter combination x in the corpus, l x Representing the length of the letter combination (consisting of several letters), Θ represents the set of letter combinations, and E represents the average entropy of the different letter combinations.
Step S44: and selecting the letter combination with the minimum entropy value as the optimal letter combination.
Step S45: and translating and writing the optimal letter combination into Chinese characters according to a French transliteration table, so as to obtain a special name transliteration result. At present, the rule mainly uses a French transliteration table formulated by national standard (foreign language, famous and Chinese characters, translation and writing guidance (GB/T17693.3-1999)).
As shown in fig. 5, step S5: translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result, which comprises the following steps:
step S51: and translating and analyzing the place name through part by using the Fabry-Perot dictionary and the place name through template to obtain a place name hierarchical grammar structure.
Step S52: and according to the principle of bottom up, converting the Law place name elements in the place name hierarchical grammar structure into Chinese place name elements layer by layer to obtain a full name transliteration result.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (1)
1. A method of machine translation of french names, the method comprising:
acquiring a legal place name group to be translated;
preprocessing the French place name group;
distinguishing based on the preprocessed French place name group to obtain a place name through part and a place name special part;
translating the place name special name part according to the letter combination in the legal Chinese transliteration table to obtain a special name transliteration result;
translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result;
integrating the special name transliteration result and the full name transliteration result to obtain a place name translation result;
the distinguishing is performed on the basis of the preprocessed French place name group, and a place name full name part and a place name special name part are obtained, which specifically comprises the following steps:
determining a place name through name template based on the preprocessed Laplace name phrase stored in the place name corpus;
determining all the common name structure decomposition schemes according to each place name common name template;
calculating the corresponding logarithmic frequency of each place name through name template;
summing the logarithmic frequencies in the same common name structure decomposition scheme to obtain a logarithmic frequency sum corresponding to each common name structure decomposition scheme;
taking the common name structure decomposition scheme with the maximum log frequency sum as a place name structure tree;
taking cotyledon nodes of the place name structure tree as the place name special name part; taking a non-cotyledon node of the place name structure tree as the place name through name part;
the method for determining the place name through name template based on the preprocessed French place name group stored in the place name corpus specifically comprises the following steps:
by means ofCalculating mutual information of any ordered word pairs in the place name corpus; wherein P is a Representing the frequency of occurrence of the pretreated French place name group a in place name corpus, P b Representing the frequency of occurrence of the pretreated French place name group b in place name corpus, P ab Representing co-occurrence frequency, MI, between pre-processed French place name groups a and b ab Mutual information representing ordered word pairs (a, b); the pretreated French place name group a and the pretreated French place name group b form ordered word pairs (a, b);
storing ordered word pairs with co-occurrence frequency greater than a first set value and mutual information greater than a second set value in the place name corpus into an ordered word pair library;
traversing all sentences, and taking the word groups of the French place names which form pretreatment in each sentence as points on the directed acyclic graph; when two preprocessed French place name groups in sentences belong to ordered word pairs in the ordered word pair library, connecting two points into a line on a directed acyclic graph, and drawing the line into directed edges;
finding out all paths on the directed acyclic graph, and generating a candidate place name template for each path according to the preprocessed French place name group corresponding to the access node; the path is composed of a plurality of directed edges;
counting the frequency of each candidate place name template, and taking the candidate place name templates with the frequency larger than a set frequency threshold as place name universal name templates;
the method comprises the steps of translating the place name transliteration part according to letter combinations in a legal Chinese transliteration table to obtain a transliteration result, and specifically comprises the following steps:
performing unsupervised learning on word letters by taking the minimum entropy as a principle to obtain letter combination distribution;
dividing and combining letters in the special name part of the place name by adopting a shortest path method word dividing method to obtain a plurality of letter combinations;
calculating average entropy values of different letter combinations according to the letter combination distribution;
selecting the letter combination with the minimum entropy value as the optimal letter combination;
the optimal letter combination is translated and written into Chinese characters according to a law Chinese transliteration table, so that a special name transliteration result is obtained;
the step of translating the place name through part according to the geographical entity category pointed by the French through name to obtain a through name transliteration result, which comprises the following steps:
translating and analyzing the place name through part by using a Fabry-Perot dictionary and the place name through template to obtain a place name hierarchical grammar structure;
and according to the principle of bottom up, converting the Law place name elements in the place name hierarchical grammar structure into Chinese place name elements layer by layer to obtain a full name transliteration result.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| AU2021104429A AU2021104429A4 (en) | 2021-07-22 | 2021-07-22 | Machine Translation Method for French Geographical Names |
| AU2021104429 | 2021-07-22 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113807105A CN113807105A (en) | 2021-12-17 |
| CN113807105B true CN113807105B (en) | 2023-07-21 |
Family
ID=77666647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111122788.XA Active CN113807105B (en) | 2021-07-22 | 2021-09-24 | French place name machine translation method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113807105B (en) |
| AU (1) | AU2021104429A4 (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116050420B (en) * | 2022-11-12 | 2023-09-22 | 武汉大学 | Chinese and French voice semantic recognition method and device based on preposition sentence |
| CN117592462B (en) * | 2024-01-18 | 2024-04-16 | 航天宏图信息技术股份有限公司 | Correlation processing method and device for open source place name data based on place group |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006252290A (en) * | 2005-03-11 | 2006-09-21 | Advanced Telecommunication Research Institute International | Machine translation device and computer program |
| CN102253930A (en) * | 2010-05-18 | 2011-11-23 | 腾讯科技（深圳）有限公司 | Method and device for translating text |
| CN109829173A (en) * | 2019-01-21 | 2019-05-31 | 中国测绘科学研究院 | A kind of English place name interpretation method and device |
| CN111709238A (en) * | 2020-06-04 | 2020-09-25 | 中国地质大学（北京） | Web page geoscience correlation calculation method based on geoscience expert knowledge |
| CN112084796A (en) * | 2020-09-15 | 2020-12-15 | 南京文图景信息科技有限公司 | Multi-language place name root Chinese translation method based on Transformer deep learning model |
| CN112905595A (en) * | 2021-03-05 | 2021-06-04 | 腾讯科技（深圳）有限公司 | Data query method and device and computer readable storage medium |
| CN112966068A (en) * | 2020-11-09 | 2021-06-15 | 袭明科技(广东)有限公司 | Resume identification method and device based on webpage information |
-
2021
- 2021-07-22 AU AU2021104429A patent/AU2021104429A4/en not_active Ceased
- 2021-09-24 CN CN202111122788.XA patent/CN113807105B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006252290A (en) * | 2005-03-11 | 2006-09-21 | Advanced Telecommunication Research Institute International | Machine translation device and computer program |
| CN102253930A (en) * | 2010-05-18 | 2011-11-23 | 腾讯科技（深圳）有限公司 | Method and device for translating text |
| CN109829173A (en) * | 2019-01-21 | 2019-05-31 | 中国测绘科学研究院 | A kind of English place name interpretation method and device |
| CN111709238A (en) * | 2020-06-04 | 2020-09-25 | 中国地质大学（北京） | Web page geoscience correlation calculation method based on geoscience expert knowledge |
| CN112084796A (en) * | 2020-09-15 | 2020-12-15 | 南京文图景信息科技有限公司 | Multi-language place name root Chinese translation method based on Transformer deep learning model |
| CN112966068A (en) * | 2020-11-09 | 2021-06-15 | 袭明科技(广东)有限公司 | Resume identification method and device based on webpage information |
| CN112905595A (en) * | 2021-03-05 | 2021-06-04 | 腾讯科技（深圳）有限公司 | Data query method and device and computer readable storage medium |
Non-Patent Citations (2)
| Title |
|---|
| 外语地名机器翻译中通专名区分技术研究;颜 闻 等;《测绘地理信息》;第第 ４６ 卷卷(第第３ 期期);全文 * |
| 机 器学习 的地名专名音译技术研究;颜 闻 等;《测绘科学》;第第４４ 卷卷(第第１０ 期期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113807105A (en) | 2021-12-17 |
| AU2021104429A4 (en) | 2021-09-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10303761B2 (en) | Method, non-transitory computer-readable recording medium storing a program, apparatus, and system for creating similar sentence from original sentences to be translated | |
| US9588960B2 (en) | Automatic extraction of named entities from texts | |
| Hovy et al. | Capturing regional variation with distributed place representations and geographic retrofitting | |
| Evans | Applying machine learning toward an automatic classification of it | |
| Dellert et al. | NorthEuraLex: A wide-coverage lexical database of Northern Eurasia | |
| CN113807105B (en) | French place name machine translation method | |
| RU2636098C1 (en) | Use of depth semantic analysis of texts on natural language for creation of training samples in methods of machine training | |
| Reynaert | Character confusion versus focus word-based correction of spelling and OCR variants in corpora | |
| CN109829173B (en) | English place name translation method and device | |
| CN111488466B (en) | Chinese language marking error corpus generating method, computing device and storage medium | |
| US11379656B2 (en) | System and method of automatic template generation | |
| Attia et al. | Arabic spelling error detection and correction | |
| Way et al. | wEBMT: developing and validating an example-based machine translation system using the world wide web | |
| US20100094615A1 (en) | Document translation apparatus and method | |
| JPWO2014002774A1 (en) | Synonym extraction system, method and recording medium | |
| CN109815503B (en) | Man-machine interaction translation method | |
| Aswani et al. | A hybrid approach to align sentences and words in English-Hindi parallel corpora | |
| US20190065453A1 (en) | Reconstructing textual annotations associated with information objects | |
| Chaudhuri | Reversed word dictionary and phonetically similar word grouping based spell-checker to Bangla text | |
| Ma et al. | Improving Chinese spell checking with bidirectional LSTMs and confusionset-based decision network | |
| CN115455986A (en) | Spanish language place name translation method, device, equipment and medium | |
| CN107590132B (en) | Method for automatically correcting part of characters-judging by English part of speech | |
| Andrés et al. | Search for hyphenated words in probabilistic indices: a machine learning approach | |
| Borin et al. | Bring vs. MTRoget: Evaluating automatic thesaurus translation. | |
| CN113705223A (en) | Personalized English text simplification method taking reader as center |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |