CN102385587B - Method and system for identifying name - Google Patents
Method and system for identifying name Download PDFInfo
- Publication number
- CN102385587B CN102385587B CN201010270770.XA CN201010270770A CN102385587B CN 102385587 B CN102385587 B CN 102385587B CN 201010270770 A CN201010270770 A CN 201010270770A CN 102385587 B CN102385587 B CN 102385587B
- Authority
- CN
- China
- Prior art keywords
- name
- candidate
- initiation sequence
- entry
- frequency meter
- 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
Landscapes
- Character Discrimination (AREA)
Abstract
The invention provides method and system for identifying a name, being applicable to the field of internet and searching, and the method comprises the following steps: storing the name identified in initial sequence and the times that the name appears in a name frequency list, determining the candidate name according to lexical items in the initial sequence, and taking the candidate name as the identified name for marking if the candidate name appears in the name frequency list and the appearing times exceed preset time threshold. The technical scheme of the invention has the advantage of improving the name identifying accuracy.
Description
Technical field
The invention belongs to internet and search field, relate in particular to a kind of recognition methods and system of name.
Background technology
Along with the development of internet, user more and more searches for Chinese name by search software in internet.The recognition methods of existing name is specially: Automatic Extraction Role Information from corpus (being stored data base), take Viterbi algorithm to carry out character labeling to cutting word result, on the basis of role's sequence, carry out the maximum coupling of pattern, finally realize the identification of Chinese personal name.
, in discovery prior art, there is following technical matters in the technical scheme providing according to prior art:
The method of the technical scheme that prior art provides is carried out character labeling to cutting word result, so when cutting word result appearance mistake, easily to name identification error, identification error rate is high.
Summary of the invention
The embodiment of the present invention provides a kind of recognition methods of name, and the recognition methods that is intended to solve prior art occurs when wrong cutting word result, easily to name identification error, and the problem that identification error rate is high.
The embodiment of the present invention is achieved in that a kind of recognition methods of name, and described method comprises the steps:
The number of times that the name identifying in initiation sequence and this name are occurred in described initiation sequence is stored in name frequency meter; According to the entry in this initiation sequence, determine candidate's name;
First Chinese character of a rear entry of the name entry of two words in this initiation sequence and this entry or the first two Chinese character are formed to candidate's name, as as described in form candidate's name appear at as described in name frequency meter, and when occurrence number is greater than frequency threshold value, using described composition candidate name as the name mark identifying, and upgrade described name frequency meter;
Maybe the first two word of triliteral name entry in this initiation sequence is formed to candidate's name, as as described in form candidate's name appear at as described in name frequency meter, and when occurrence number is greater than frequency threshold value, using described composition candidate name as the name mark identifying, and upgrade described name frequency.
The present invention also provides a kind of recognition system of name, and described system comprises:
Storage unit, the number of times occurring in described initiation sequence for name that initiation sequence is identified and this name is stored in name frequency meter;
Composite module, for first Chinese character of a rear entry of the name entry of two words of this initiation sequence and this entry or the first two Chinese character are formed to candidate's name,
Or for the first two word of the triliteral name entry of this initiation sequence is formed to candidate's name;
Recognition unit, for appearing at this name frequency meter at this candidate's name, and occurrence number is while surpassing preset times threshold value, using this candidate's name as the name identifying, and upgrades described name frequency.
The embodiment of the present invention compared with prior art, beneficial effect is: technical scheme of the present invention is set up name frequency meter to the name of initiation sequence and this name occurrence number, then according to the entry of this initiation sequence, determine candidate's name, and the name in this candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, because the method be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it has advantages of the name of raising recognition accuracy.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the recognition methods of a kind of name provided by the invention;
Fig. 2 is the process flow diagram that the embodiment of the present invention one provides a kind of recognition methods of name;
Fig. 3 provides a kind of name correction process flow diagram for the embodiment of the present invention one;
Fig. 4 is the process flow diagram that the embodiment of the present invention two provides a kind of recognition methods of name;
Fig. 5 is the process flow diagram that the embodiment of the present invention three provides a kind of recognition methods of name;
Fig. 6 is the structural drawing that the invention provides a kind of recognition system of name.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.
The invention provides a kind of recognition methods of name, the method as shown in Figure 1, specifically comprises the steps:
S10, the number of times that the name identifying in initiation sequence and this name are occurred are stored in name frequency meter;
It should be noted that, above-mentioned initiation sequence can be: the sequence to name after preliminary identifying processing.The method of above-mentioned identifying processing can be the method for prior art, and for example Viterbi algorithm, can certainly be other recognition methods, as long as the method can tentatively identify name, the present invention does not limit to the concrete manifestation form of this recognition methods.
S11, according to the entry in this initiation sequence, determine candidate's name;
S12, appear in this name frequency meter as this candidate's name, and occurrence number is while surpassing preset times threshold value, using this candidate's name as the name identifying.
Optionally, the method can also comprise: mark the name that this identifies, and the number of times occurring in initiation sequence according to this candidate's name upgrades name frequency meter.
Above-mentioned preset times threshold users can be set in advance voluntarily, and for example 1,2,3 etc., the present invention does not limit to the concrete value of this frequency threshold value.
Optionally, realizing any that the concrete grammar of S11 can be in subordinate's mode, can certainly be the combination in any in subordinate's mode.
Mode A, two or more entries continuous in initiation sequence are combined into candidate's name;
Mode B, first Chinese character of a rear entry of the name entry of two words in initiation sequence and this entry is formed to candidate's name;
Mode C, the first two word of triliteral name entry in initiation sequence is formed to candidate's name.
It should be noted that, people's name recognition method provided by the invention is mainly used in the identification of Chinese personal name, if the name of other words has the feature of Chinese personal name, also can be applied to other word, such as other minority name family word etc. of the language of the Manchus or some.
The method that the present embodiment provides is set up name frequency meter to the name of initiation sequence and this name occurrence number, then according to the entry of this initiation sequence, determine candidate's name, and the name in this candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, by this candidate's name sign, and upgrade this name frequency meter, because the method be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it can solve emphatically the traditional difficult problem in name identification: without the identification of surname name and name identification ambiguity, thereby can improve name recognition accuracy.
embodiment mono-:
The present embodiment provides a kind of recognition methods of name, the technology scene that the present embodiment is realized is: the method that the present embodiment provides is completed by identification equipment, this identification equipment is specifically as follows, computing machine, mobile terminal, the digital electric equipment such as PDA, the present embodiment be take Chinese as example, the present embodiment be take hypomere document and the recognition methods of the present embodiment is described as example, it should be noted that, hypomere word can be the sequence after the identifying processing method processing through prior art, for convenience of description, the present embodiment is called initiation sequence by the sequence unification after recognition methods identifying processing.Shown in this initiation sequence is specific as follows:
Because of detonieren, be promoted to " topic player " Zeng Yike/nr enrage Bao little Bai/nr before and again became person who attract people's attention that night.Group " is supported " in interior ground judging panel still very good be once lost can/original music and the pure and fresh typhoon of nr.Be called as " sheep angel " be once lost can/the original works < < Leo > > that remains oneself that nr brings.Zeng Yike/nr remains " dispute can ".Zeng Yike/nr and second takes turns " little swallow " Li Li/nr that score is minimum and carries out ultimate PK and fight to the finish.Once be lost this moment can/the ballot score of nr and Li Li/nr is 0: 2.Then Li Li/the nr on her side has drawn time her to say: keep your wool on." wanting to swear at people " three words of Zeng Yike/nr are very clear.Reporter find to be once lost can/nr write blog all through the night.To in match because comfort Li Li/nr said improper language has carried out sincere apology.Zeng Yike/nr represents to abandon because of extraneous comment the music dream of oneself.Li Li/nr also in the blog of oneself for be once lost can/nr clarification.But being lost of PK can be still unable to bear tear successively.Li Lixi/nr obtains the 13rd, my elegant whole nation of my type of 2007 Sprites.Li Lifang/nr is won.Once be lost/nr can a ticket difference be defeated by Liu Xijun/nr and transfer to undetermined.
Wherein, the entry of " nr " in above-mentioned initiation sequence is the name identifying.Above-mentioned entry can be more predefined words in dictionary, for example " can ", " whole nation " etc., can certainly be artificial some words that arrange, such as " Li Lianjie ", " Cheng Long ", " Jordon " etc.; It should be noted that, the entry in initiation sequence separates by space character, for example, in " but successively ", entry " but " and entry " successively " by space character, separate.The method that the present embodiment provides as shown in Figure 2, specifically comprises the steps:
S20, the number of times that the name identifying in this initiation sequence and this name are occurred are stored in name frequency meter;
The name frequency meter of above-mentioned initiation sequence specifically can be as shown in table 1:
Table 1:
Name | Number of times |
Zeng Yike | 10 |
Bao little Bai | 1 |
Li Li | 7 |
Li Lixi | 1 |
Li Lifang | 1 |
Once be lost | 1 |
S21, two or more continuous entries are combined into candidate's name;
It should be noted that, above-mentioned entry can be the entry of single character, for example " "; Certainly, in actual conditions, can be also the entry of a plurality of words, for example " language ".
It should be noted that, during if any continuous a plurality of individual character entry, candidate's name of its composition can be also a plurality of candidate's names, here with two continuous individual character entries, be combined into candidate's example by name, " drawn time she say " can form 4 candidate's names, be respectively: " leaving behind ", " time ", " lower she ", " she says ".It should be noted that, combinations thereof becomes the number of name list of the candidates words bar to be generally 2,3,4; Certainly the definition of this number is just stipulated by the custom of China name number of words, do not get rid of when custom changes, name number of words becomes the numbers of words such as 8,9,10, and for example the number of the foreigner's Chinese name is the number that surpasses 4, above-mentioned a plurality of can setting according to actual conditions.
S22, appear in above-mentioned name frequency meter as this candidate's name, and occurrence number is while being greater than frequency threshold value, using this candidate's name as the name mark identifying, and upgrades this name frequency meter.
The concrete mode of this name frequency meter of above-mentioned renewal can be: the number of times to this candidate's name occurring in this name frequency meter upgrades, and for example this candidate's name occurred 2 times, will in this name frequency meter, the number of times of this candidate's name be increased by 2 times.
The process flow diagram of the correction in the present embodiment method as shown in Figure 3, wherein, the name of S20 can be stored in the name frequency meter of Fig. 3, and error correction can complete the operation of S21 and S22.
The method that the present embodiment provides is set up name frequency meter to the name of initiation sequence and this name occurrence number, then the name two or more continuous entries being formed in candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, by this candidate's name sign, and upgrade this name frequency meter, because the method be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it can solve emphatically the traditional difficult problem in name identification: without the identification of surname name and name identification ambiguity, thereby can improve the advantage of name recognition accuracy.
embodiment bis-:
The present embodiment provides a kind of recognition methods of name, and the technology scene that the technology scene that the present embodiment provides provides with embodiment mono-is identical, and the method as shown in Figure 4, comprises the steps:
S40, the number of times that the name identifying in this initiation sequence and this name are occurred are stored in name frequency meter;
This name frequency meter specifically can be as shown in table 1.
S41, first Chinese character of a rear entry of the name entry of two words and this entry is formed to candidate's name;
The implementation method of S41 is described with an actual example below, with the name entry of above-mentioned two words, " was once lost/nr " here, a rear entry of this entry be " can ", the candidate " Zeng Yike " by name who forms.Certainly, in actual conditions, also the first two Chinese character of a rear entry of the name entry of two words and this entry can be formed to candidate's name.
S42, appear in above-mentioned name frequency meter as this candidate's name, and occurrence number is while being greater than frequency threshold value, using this candidate's name as the name mark identifying, and upgrades this name frequency meter.
Here suppose that frequency threshold value is 3 times, certainly in actual conditions, can be arranged to other numeral, for example 2,4 or 1 etc., the number of times occurring in name frequency meter due to candidate's name " Zeng Yike " is 10 times, is greater than frequency threshold value, so " Zeng Yike " carried out to name mark, and upgrade name frequency meter, the name frequency meter after renewal is as shown in table 2:
Table 2
Name | Number of times |
Zeng Yike | 11 |
Bao little Bai | 1 |
Li Li | 7 |
Li Lixi | 1 |
Li Lifang | 1 |
Once be lost | 1 |
Sequence after the name identifying is marked is:
" beautiful side/nr is won.Zeng Yike/nr is defeated by Liu Xijun with the difference of a ticket ".It should be noted that, because this mark is only changed the row second from the bottom of above-mentioned initiation sequence, so only write a line of change here.
The method that the present embodiment provides is set up name frequency meter to the name of initiation sequence and this name occurrence number, then the name first Chinese character of a rear entry of the name entry of two words and this entry being formed in candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, by this candidate's name sign, and upgrade this name frequency meter, because the method be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it can solve emphatically the traditional difficult problem in name identification: without the identification of surname name and name identification ambiguity, ambiguity name or become the identification problem of composer of ci poetry's name with context for example, thereby can improve the advantage of name recognition accuracy.
embodiment tri-:
The present embodiment provides a kind of recognition methods of name, and the technology scene that the technology scene that the present embodiment provides provides with embodiment mono-is identical, and the method as shown in Figure 4, comprises the steps:
S50, the number of times that the name identifying in this initiation sequence and this name are occurred are stored in name frequency meter;
This name frequency meter specifically can be as shown in table 1.
S51, the first two word of triliteral name entry is formed to candidate's name;
The implementation method of S51 is described with an actual example below, and here with above-mentioned triliteral name entry " Zeng Yike/nr ", the candidate of composition is by name: " being once lost "; The candidate that " Li Lixi/nr " and " Li Lifang/nr " forms is by name: " Li Li ".
S52, appear in above-mentioned name frequency meter as this candidate's name, and occurrence number is while being greater than frequency threshold value, using this candidate's name as the name mark identifying, and upgrades this name frequency meter.
Here suppose that frequency threshold value is 3 times, because the occurrence number of " being once lost " is for once, so it is not greater than frequency threshold value; And the number of times that " Li Li " occurs is 7 times, be greater than frequency threshold value, so " Li Lixi/nr " is modified as to " Li Li/nr happiness "; " Li Licheng/nr " is modified as to " Li Li/nr becomes "; And upgrade name frequency meter, the name frequency meter after renewal is as shown in table 3:
Table 3:
Name | Number of times |
Zeng Yike | 10 |
Bao little Bai | 1 |
Li Li | 9 |
Once be lost | 1 |
Sequence after the name identifying is marked is:
" water.Li Li/nr gains the 13rd, my elegant whole nation of my type of 2007 Sprites happily.Li Li/nr side is won.Once be lost/nr can a ticket difference be defeated by Liu Xijun "
The method that the present embodiment provides is set up name frequency meter to the name of initiation sequence and this name occurrence number, then the name the first two word in triliteral name entry being formed in candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, by this candidate's name sign, and upgrade this name frequency meter, because the method be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it can solve emphatically the traditional difficult problem in name identification: without the identification of surname name and name identification ambiguity, thereby can improve the advantage of name recognition accuracy.
The present invention also provides a kind of recognition system of name, and this system as shown in Figure 6, comprising:
Storage unit 61 is stored in the number of times of the name identifying in initiation sequence and the appearance of this name in name frequency meter;
Candidate's name determined in entry in determining unit 62 these initiation sequences;
Recognition unit 63 appears in this name frequency meter at this candidate's name, and occurrence number is while surpassing preset times threshold value, using this candidate's name as the name mark identifying.
The definition of above-mentioned initiation sequence can be referring to the associated description in embodiment of the method.
Optionally, said system can also comprise:
Mark updating block 64 these names that identify of mark, and the number of times occurring in initiation sequence according to this candidate's name upgrades name frequency meter.
Optionally, above-mentioned determining unit 62 can comprise in following module any or a plurality of:
Composite module 621 is combined into candidate's name by two or more entries continuous in this initiation sequence continuously;
Composite module 622 forms candidate's name by first Chinese character of a rear entry of the name entry of two words in this initiation sequence and this entry or the first two Chinese character;
Form module 623 the first two word of triliteral name entry in this initiation sequence is formed to candidate's name.
The system that the present embodiment provides is set up name frequency meter to the name of initiation sequence and this name occurrence number, then according to the entry of this initiation sequence, determine candidate's name, and the name in this candidate's name and this name frequency meter is compared, as appear in this name frequency meter, and when the number of times in this name frequency meter surpasses frequency threshold value, determine this candidate name that leaks identification by name, by this candidate's name sign, and upgrade this name frequency meter, because this system be take initiation sequence and is carried out error correction as basis, so it has when cutting word result appearance mistake, can carry out the processing of error correction to the recognition result of prior art (being initiation sequence), so it can solve emphatically the traditional difficult problem in name identification: without the identification of surname name and name identification ambiguity, thereby can improve the advantage of name recognition accuracy.
It should be noted that in said system embodiment, included unit is just divided according to function logic, but is not limited to above-mentioned division, as long as can realize corresponding function; In addition, the concrete title of each functional unit also, just in being convenient to mutual differentiation, is not limited to protection scope of the present invention.
In addition, one of ordinary skill in the art will appreciate that all or part of step realizing in above-described embodiment method is to come the hardware that instruction is relevant to complete by program, corresponding program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium of mentioning can be ROM (read-only memory), disk or CD etc.
In sum, technical scheme provided by the invention has advantages of difficult to name identification error.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (4)
1. a recognition methods for name, is characterized in that, described method comprises the steps:
The number of times that the name identifying in initiation sequence and this name are occurred in described initiation sequence is stored in name frequency meter,
First Chinese character of a rear entry of the name entry of two words in this initiation sequence and this entry or the first two Chinese character are formed to candidate's name, as as described in form candidate's name appear at as described in name frequency meter, and when occurrence number is greater than frequency threshold value, using described composition candidate name as the name mark identifying, and upgrade described name frequency meter;
Maybe the first two word of triliteral name entry in this initiation sequence is formed to candidate's name, as as described in form candidate's name appear at as described in name frequency meter, and when occurrence number is greater than frequency threshold value, using described composition candidate name as the name mark identifying, and upgrade described name frequency.
2. method according to claim 1, is characterized in that, described method also comprises the steps: this candidate's name as the name identifying afterwards
Mark the name that this identifies, and the number of times occurring according to this candidate's name upgrades name frequency meter in initiation sequence.
3. a recognition system for name, is characterized in that, described system comprises:
Storage unit, the number of times occurring in described initiation sequence for name that initiation sequence is identified and this name is stored in name frequency meter;
Composite module, for first Chinese character of a rear entry of the name entry of two words of this initiation sequence and this entry or the first two Chinese character are formed to candidate's name,
Or for the first two word of the triliteral name entry of this initiation sequence is formed to candidate's name;
Recognition unit, for appearing at this name frequency meter at this candidate's name, and occurrence number is while surpassing preset times threshold value, using this candidate's name as the name identifying, and upgrades described name frequency.
4. system according to claim 3, is characterized in that, described system also comprises:
Mark updating block, the name identifying for marking this, and the number of times occurring in initiation sequence according to this candidate's name upgrades name frequency meter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010270770.XA CN102385587B (en) | 2010-08-27 | 2010-08-27 | Method and system for identifying name |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010270770.XA CN102385587B (en) | 2010-08-27 | 2010-08-27 | Method and system for identifying name |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102385587A CN102385587A (en) | 2012-03-21 |
CN102385587B true CN102385587B (en) | 2014-07-30 |
Family
ID=45825008
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201010270770.XA Active CN102385587B (en) | 2010-08-27 | 2010-08-27 | Method and system for identifying name |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102385587B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103823859B (en) * | 2014-02-21 | 2017-02-22 | 安徽博约信息科技股份有限公司 | Name recognition algorithm based on combination of decision-making tree rules and multiple statistic models |
CN105373530A (en) * | 2015-12-03 | 2016-03-02 | 北京锐安科技有限公司 | Chinese name identification method and apparatus |
CN112016272A (en) * | 2019-10-29 | 2020-12-01 | 河南拓普计算机网络工程有限公司 | Bidding information review expert identification system and method |
CN113792186B (en) * | 2021-08-16 | 2023-07-11 | 青岛海尔科技有限公司 | Method, device, electronic equipment and storage medium for name retrieval |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101088082A (en) * | 2004-10-25 | 2007-12-12 | 英孚威尔公司 | Full text query and search systems and methods of use |
CN101645134A (en) * | 2005-07-29 | 2010-02-10 | 富士通株式会社 | Integral place name recognition method and integral place name recognition device |
-
2010
- 2010-08-27 CN CN201010270770.XA patent/CN102385587B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101088082A (en) * | 2004-10-25 | 2007-12-12 | 英孚威尔公司 | Full text query and search systems and methods of use |
CN101645134A (en) * | 2005-07-29 | 2010-02-10 | 富士通株式会社 | Integral place name recognition method and integral place name recognition device |
Non-Patent Citations (1)
Title |
---|
和雪娟等.基于统计和规则混合策略的中国人名识别研究.《云南民族大学学报(自然科学版)》.2009,第18卷(第1期),第70-72页. * |
Also Published As
Publication number | Publication date |
---|---|
CN102385587A (en) | 2012-03-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10783171B2 (en) | Address search method and device | |
WO2016037519A1 (en) | Input method and apparatus and electronic device | |
CN103123618B (en) | Text similarity acquisition methods and device | |
CN105005577A (en) | Address matching method | |
CN102385587B (en) | Method and system for identifying name | |
CN102693279B (en) | Method, device and system for fast calculating comment similarity | |
US20120330990A1 (en) | Evaluating query translations for cross-language query suggestion | |
CN106202028B (en) | A kind of address information recognition methods and device | |
CN108287843A (en) | A kind of method and apparatus and navigation equipment of interest point information retrieval | |
CN106537370A (en) | Method and system for robust tagging of named entities in the presence of source or translation errors | |
WO2011140766A1 (en) | Method and terminal device for updating word stock | |
CN102750282B (en) | Synonym template mining method and device as well as synonym mining method and device | |
JP2015506515A (en) | Method, apparatus and computer storage medium for automatically adding tags to a document | |
CN104199965A (en) | Semantic information retrieval method | |
CN101853292A (en) | Method and system for constructing business social network | |
CN109597895B (en) | Knowledge graph-based official document searching method | |
CN110705292B (en) | Entity name extraction method based on knowledge base and deep learning | |
JP5934749B2 (en) | Method and apparatus for journal generation | |
CN104615782B (en) | Address matching process based on sliding window maximum matching algorithm | |
CN106326206B (en) | Entity extraction method based on grammar template | |
CN102193920A (en) | Name word stock generating method and device as well as text input system | |
CN102033891B (en) | Retrieval method and device for Chinese information | |
CN102567365A (en) | Input method and input system based on labeling specific to a keyword | |
CN103500163A (en) | Method and device for recognizing event key progress | |
CN111611793B (en) | Data processing method, device, equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
C41 | Transfer of patent application or patent right or utility model | ||
TR01 | Transfer of patent right |
Effective date of registration: 20151230 Address after: The South Road in Guangdong province Shenzhen city Fiyta building 518057 floor 5-10 Nanshan District high tech Zone Patentee after: Shenzhen Tencent Computer System Co., Ltd. Address before: Shenzhen Futian District City, Guangdong province 518044 Zhenxing Road, SEG Science Park 2 East Room 403 Patentee before: Tencent Technology (Shenzhen) Co., Ltd. |