CN114238664A - Overseas trademark retrieval method, equipment, medium and product - Google Patents
Overseas trademark retrieval method, equipment, medium and product Download PDFInfo
- Publication number
- CN114238664A CN114238664A CN202111306150.1A CN202111306150A CN114238664A CN 114238664 A CN114238664 A CN 114238664A CN 202111306150 A CN202111306150 A CN 202111306150A CN 114238664 A CN114238664 A CN 114238664A
- Authority
- CN
- China
- Prior art keywords
- trademark
- target
- retrieval
- overseas
- sample
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/381—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using identifiers, e.g. barcodes, RFIDs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/38—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/383—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
Abstract
The invention provides a overseas trademark retrieval method, which comprises the steps of receiving trademark English keywords input by a user for overseas trademark retrieval and retrieval rules appointed by the user; screening out a corresponding target trademark name from an operation memory database of the local server according to the trademark English keyword and the retrieval rule, and screening out corresponding target trademark associated information from a distributed database on a hard disk of the local server according to the target trademark name; and outputting the target trademark name and the corresponding target trademark association information to a user side. In the whole calculation process, the target trademark name and the target trademark associated information are respectively placed in a hard disk and a memory of the local server for screening and matching, so that the memory pressure of the local server is greatly reduced, and a large amount of calculation requirements are met.
Description
Technical Field
The invention relates to the field of trademark retrieval, in particular to a method, equipment, a medium and a method for retrieving overseas trademarks.
Background
Most of traditional database query modes are like keyword query of a relational database MySQL (MySQL is a relational database management system) and non-relational nosql (generally refers to a non-relational database) database query, the like keyword query of the relational database MySQL is not suitable for trademark data services, and trademark data is an incremental data system. The non-relational nosql database faces the difficulty of data updating, real-time updating of trademark incremental data can hardly be met, and the approximate query service still depends on the database, so that larger machine memory and disk space are required and the target requirement of the approximate query service is far different. In summary, the two database query methods cannot meet the requirement of foreign trademark retrieval.
Disclosure of Invention
In order to overcome the defects of the prior art, one of the purposes of the invention is to provide a overseas trademark retrieval method, which can solve the problem that the traditional database query mode meets the huge retrieval requirement in the overseas trademark retrieval process.
The second objective of the present invention is to provide a computer-readable storage medium, which can solve the problem that the traditional database query method can satisfy the huge search requirement in the overseas trademark search process.
The invention also aims to provide a computer program product which can solve the problem that the traditional database query mode meets the huge search requirement in the overseas trademark search process.
One of the purposes of the invention is realized by adopting the following technical scheme:
a overseas trademark retrieval method comprises the following steps:
receiving information to be retrieved, and receiving trademark English keywords input by a user for overseas trademark retrieval and retrieval rules specified by the user;
searching overseas trademarks, screening out corresponding target trademark names from an operating memory database of a local server according to the trademark English keywords and the search rules, and screening out corresponding target trademark associated information from a distributed database on a hard disk of the local server according to the target trademark names;
and outputting a retrieval result, and outputting the target trademark name and the corresponding target trademark associated information to a user side.
Further, the retrieving of the overseas trademark specifically comprises the following substeps:
performing word segmentation processing, namely judging whether the trademark English keyword contains at least two key words or not, if so, performing word segmentation processing on the trademark English keyword to obtain a plurality of trademark English key words to be retrieved, and if not, taking the trademark English keyword as the trademark English key word to be retrieved without word segmentation processing;
screening a target trademark name, screening a sample trademark name corresponding to the English key vocabulary of the trademark to be searched from a plurality of sample trademark names stored in an operating memory database according to the search rule as the target trademark name, and obtaining a target trademark number corresponding to the target trademark name;
and screening target trademark associated information, and screening the corresponding target trademark associated information according to the distributed database of the target trademark number on the hard disk of the local server.
Further, before the target trademark associated information is screened, precise screening is further included, a first vector score corresponding to the target trademark name is calculated according to a first preset weight and a first occurrence frequency ratio corresponding to the target trademark name, the first occurrence frequency ratio is a ratio of the occurrence frequency of each English letter in the target trademark name in all sample trademark names in the operating memory database to the sum of all English letters in the target trademark name, a second vector score corresponding to the trademark English keyword is calculated according to a second preset weight and a second occurrence frequency ratio corresponding to the trademark English keyword, and the second occurrence frequency ratio is a ratio of the occurrence frequency of each letter in the trademark English in all sample trademark names in the operating memory database to the sum of all English letters in the target trademark name, and calculating a vector score difference value of the first vector score and the second vector score, and reserving the target trademark name with the vector score difference value smaller than a preset vector score difference value threshold value.
Further, if the number of the target trademark names with the vector score difference smaller than the preset vector score difference threshold is larger than ten, sorting the target trademark names with the vector score difference smaller than the preset vector score difference threshold from small to large according to the vector score difference, and reserving the first ten target names in the sorting.
Further, before the information to be retrieved is received, a sample database is created, data cleaning processing, user-defined word segmentation processing and segmentation combination processing are performed on sample trademark data containing sample trademark names and sample trademark associated information to obtain a plurality of sample trademark names and corresponding sample trademark associated information, a sample trademark number is set for each sample trademark name and corresponding sample trademark associated information, all the sample trademark names and corresponding sample trademark numbers are stored in an operating memory database in the local server, and all the sample trademark associated information and corresponding sample trademark numbers are stored in a distributed database on a hard disk of the local server.
Further, in the step of receiving the information to be retrieved, the step of retrieving overseas trademarks is performed on the batch of information to be retrieved, which is received each time and contains a plurality of trademark English keywords, so that a corresponding batch query result containing a plurality of target trademark associated information is obtained.
Further, the retrieval rule is a homophone retrieval rule or an add-subtract word retrieval rule or an inclusion/included retrieval rule or an order-changing retrieval rule.
And further, the method also comprises approximate retrieval, wherein the trademark English keyword is input into a preset examination mechanism model for approximate examination, and an approximate examination result is obtained.
The second purpose of the invention is realized by adopting the following technical scheme:
a computer-readable storage medium having stored thereon a computer program for execution by a processor of a method of overseas brand retrieval as described in the present application.
The third purpose of the invention is realized by adopting the following technical scheme:
a computer program product comprising a computer program which, when executed by a processor, implements a method of overseas brand retrieval as described in the present application.
Compared with the prior art, the invention has the beneficial effects that: the oversea trademark retrieval method comprises the steps of receiving trademark English keywords input by a user and used for oversea trademark retrieval and retrieval rules appointed by the user, screening out corresponding target trademark names from an operation memory database of a local server according to the trademark English keywords and the retrieval rules, screening out corresponding target trademark associated information from a distributed database of a hard disk of the local server according to the target trademark names, finally outputting the target trademark names and the corresponding target trademark associated information to a user terminal, respectively placing the target trademark names and the target trademark associated information in the hard disk and the memory of the local server for screening and matching in the whole calculation process, greatly reducing the memory pressure of the local server, and simultaneously directly dropping the target trademark associated information serving as non-word index fields into a distributed database, the distributed database realizes second-level query of hundred million data on the non-index field, the target trademark name serving as the index field can be quickly hit in the memory after word segmentation, and the whole query process can be completed by directly taking the main key of the hit data.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood and to implement them in accordance with the contents of the description, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings. The detailed description of the present invention is given in detail by the following examples and the accompanying drawings.
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 application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 is a flow chart of the overseas trademark retrieval method of the present invention.
Detailed Description
The present invention will be further described with reference to the accompanying drawings and the detailed description, and it should be noted that any combination of the embodiments or technical features described below can be used to form a new embodiment without conflict.
As shown in fig. 1, the method for searching for overseas trademarks in the present application specifically includes the following steps:
the method comprises the steps of creating a sample database, carrying out data cleaning processing, user-defined word segmentation processing and segmentation combination processing on sample trademark data containing sample trademark names and sample trademark associated information to obtain a plurality of sample trademark names and corresponding sample trademark associated information, setting a sample trademark number for each sample trademark name and corresponding sample trademark associated information, storing all the sample trademark names and corresponding sample trademark numbers in an operating memory database in a local server, and storing all the sample trademark associated information and corresponding sample trademark numbers in a distributed database on a hard disk of the local server. The data processing specifically includes: the sample trademark data is normalized and feature extracted, a data feature model is built in advance, multi-dimensional query can be rapidly achieved, and similar vector features can be directly searched in the features during multi-dimensional query. The user-defined word segmentation processing specifically comprises the following steps: through a preset custom word segmentation device, the sample trademark data is segmented according to business characteristics, for example, similar query of homophones needs to extract vowel parts in the trademark name, and a plurality of phrases are combined to flow into the next stage. The segment combination specifically comprises: in the link, the word segmentation phrases and the trademark data in the previous steps are received, sample trademark associated information of the non-index query field in the sample trademark data is stored in a distributed database, and the stored index is the unique number of the trademark; meanwhile, the word segmentation result and the unique number can reside in the memory; and directly storing the sample trademark name serving as the index query field into an operating memory database. The trademark-related data in the present embodiment is information related to a corresponding trademark, for example, bibliographic information such as the applicant and the applicant address of the trademark.
Receiving information to be searched, and receiving trademark English keywords input by a user for overseas trademark search and search rules specified by the user. The search rule is a homophone search rule, an add-subtract word search rule, an inclusion/included search rule or an order-changing search rule.
Searching overseas trademarks, screening out corresponding target trademark names from an operating memory database of the local server according to the trademark English keywords and the search rules, and screening out corresponding target trademark associated information from a distributed database on a hard disk of the local server according to the target trademark names. The method specifically comprises the following steps:
performing word segmentation processing, namely judging whether the trademark English keyword contains at least two key words or not, if so, performing word segmentation processing on the trademark English keyword to obtain a plurality of trademark English key words to be retrieved, and if not, taking the trademark English keyword as the trademark English key word to be retrieved without word segmentation processing;
screening a target trademark name, screening a sample trademark name corresponding to the English key vocabulary of the trademark to be searched from a plurality of sample trademark names stored in an operating memory database according to the search rule as the target trademark name, and obtaining a target trademark number corresponding to the target trademark name;
and screening target trademark associated information, and screening the corresponding target trademark associated information according to the distributed database of the target trademark number on the hard disk of the local server.
Preferably, the method further comprises a precise screening step before the screening of the target trademark associated information, wherein a first vector score corresponding to the target trademark name is calculated according to a first preset weight corresponding to the target trademark name and a first occurrence ratio, the first number of occurrences is compared to the number of occurrences of each english letter in the target brand name in all sample brand names in the operating memory database, calculating a second vector score corresponding to the trademark English keyword according to a second preset weight and a second occurrence frequency ratio corresponding to the trademark English keyword, and the second occurrence frequency is the frequency of occurrence of each letter in the trademark English keyword in all sample trademark names in the operating memory database, the vector score difference value of the first vector score and the second vector score is calculated, and the target trademark name with the vector score difference value smaller than a preset vector score difference value threshold value is reserved. If the number of the target trademark names with the vector score difference smaller than the preset vector score difference threshold value is larger than ten, sorting the target trademark names with the vector score difference smaller than the preset vector score difference threshold value according to the vector score difference from small to large, and reserving the first ten target names in the sorting. For the above calculation process of the vector fraction, the following is exemplified:
assuming that the trademark english keyword inputted by the user for overseas trademark search is "THIS IS NIKI" and the designated search rule is a homophone search rule, "THIS IS NIKI" is first participled as: the three key words, namely THIS, IS and NIKI are used as English key words of the trademark to be retrieved, and a first target trademark name corresponding to THIS, a second target trademark name corresponding to IS and a third target trademark name corresponding to NIKI are respectively screened from the operation memory database according to retrieval rules. Calculating a second vector vocabulary of the three vocabularies of the THIS, the IS and the NIKI respectively, wherein each vector score IS 1/(1+ math.exp (-2 × second preset weight-second occurrence time ratio), and further obtaining a vector score of the THIS, the IS and the NIKI, for example, when calculating the vector score of the NIKI, assuming that the sample tradenames stored in the operation memory database according to the NIKI are NIKE, NGAO and NIKI, the number of occurrences of the letters IS N (3) G (1) K (2) O (1) I (3) E (1) a (1), the corresponding vector score IS 1/1+ math.exp (-2 ═ 0.8- (1 × 0.25+ 1.25 +0.166+0.083)) (N (3) G (1) K (2) O (1) a (1), and the corresponding vector score IS 32 + math.2 ═ 0.8- (1 × 0.25 × 0.539 + 2.36) (+ 3) G + 2.36.36) (+ 3) NIKE) (+ 368) G + 3) G + 2.36.2) N + 3) G + 3) b) (368 (9) G + 3) N) (9) b) (9) G + 2) N + 3) 3 x 0.083)) ═ 1.1225789459298634; assuming the user input is an NIKI, the NIKI output is hit.
And outputting a retrieval result, and outputting the target trademark name and the corresponding target trademark associated information to a user side.
In this embodiment, in the step of receiving information to be retrieved, the step of retrieving overseas trademarks is performed on batch information to be retrieved, which is received each time and contains a plurality of trademark english keywords, to obtain a corresponding batch query result containing a plurality of target trademark associated information; the method specifically comprises the following steps: manually and once importing an xls or txt file by a user, wherein the file comprises about 1w or more brand names to be applied (namely the brand English keywords) which are arranged in a line-changing mode; and (4) carrying out approximate query on the imported trademark English keywords and all overseas trademarks in the database in sequence, and sequencing according to the query quantity results. When the trademark database has the similarity value of completely identical trademark names as k, and when the trademark database has the similarity value of approximate trademark names as m, the total similarity value z is k + m. And discarding trademarks with z exceeding 100, sequencing each trademark name z to be inquired from small to large, and then sequentially exporting the sequenced trademark name z to a user for viewing, wherein the trademarks in front of the sequencing are the trademark names which can be applied by the user and have higher success rate, and meanwhile, the trademark z is marked as a similar value, and generally, trademarks with z below 10 can be manually checked again.
In addition, the method also comprises an approximate query step, wherein the trademark English keyword is input into a preset examination mechanism model for approximate examination, and an approximate examination result is obtained. The method specifically comprises the following steps: in addition to the conventional word segmentation and inclusion relation, in the embodiment, a search with higher accuracy can be performed according to a manual search formula of overseas trademark examiners, such as American trademark examiners, and all history files in a trademark library, and a search suggestion that a user is closer to the judgment of the examiners is given. Take the united states as an example: each inquiry record code of the American examiner is public and is stored in the American trademark office database, and all the examination records are trained through a machine learning algorithm to accurately learn the examination rules. The us trademark review rules specifically include the following: 1. the applied trademark is contained by the current living trademark 2, the applied trademark is the current living trademark 3, the applied trademark is completely the same as the current living trademark after adding or subtracting words 4, the pronunciation of the applied trademark is the same as the pronunciation of the current living trademark 5, the applied trademark is completely the same as the current living trademark after exchanging the word sequence 6, the applied trademark contains sensitive words, and sensitive words such as celebrities, place names, political religions and the like are required to be removed (the situation that a black name single library cannot be applied is required to be established). And (3) acquiring the rule by using XSearch Search Summary data, automatically querying 1-6 by applying an approximate query rule, completing a blacklist related to the American trademark in the rule 7, and automatically prompting that the trademark applied by the user belongs to the xx category blacklist if the trademark applied by the user is in the blacklist. In addition, the newly applied trademark can be avoided being rejected according to the trademark rejection records in the overseas database. For example: the user wants to apply for a trademark: DREAMO, which can be registered by general trademark inquiry, but inquires about the trademark of the prior application DREAMO because of the approximate rejection of a certain trademark through a database, and then considers that the trademark can cause the rejection for the same reason; the censorship of dead trademarks and all prior approximate trademarks is also taken as a reference basis for trademark inquiry to improve the accuracy of trademark inquiry.
The present invention also provides a computer-readable storage medium having stored thereon a computer program for execution by a processor of a method of overseas trademark retrieval as described in the present application.
The invention also provides a computer program product comprising a computer program which, when executed by a processor, implements a method of overseas brand retrieval as described in the present application.
The oversea trademark retrieval method comprises the steps of receiving trademark English keywords input by a user and used for oversea trademark retrieval and retrieval rules appointed by the user, screening out corresponding target trademark names from an operation memory database of a local server according to the trademark English keywords and the retrieval rules, screening out corresponding target trademark associated information from a distributed database of a hard disk of the local server according to the target trademark names, finally outputting the target trademark names and the corresponding target trademark associated information to a user terminal, respectively placing the target trademark names and the target trademark associated information in the hard disk and the memory of the local server for screening and matching in the whole calculation process, greatly reducing the memory pressure of the local server, and simultaneously directly dropping the target trademark associated information serving as non-word index fields into a distributed database, the distributed database realizes second-level query of hundred million data on the non-index field, the target trademark name serving as the index field can be quickly hit in the memory after word segmentation, and the whole query process can be completed by directly taking the main key of the hit data.
The foregoing is merely a preferred embodiment of the invention and is not intended to limit the invention in any manner; those skilled in the art can readily practice the invention as shown and described in the drawings and detailed description herein; however, those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiments as a basis for designing or modifying other structures for carrying out the same purposes of the present invention without departing from the scope of the invention as defined by the appended claims; meanwhile, any changes, modifications, and evolutions of the equivalent changes of the above embodiments according to the actual techniques of the present invention are still within the protection scope of the technical solution of the present invention.
Claims (10)
1. A overseas trademark retrieval method is characterized in that: the method comprises the following steps:
receiving information to be retrieved, and receiving trademark English keywords input by a user for overseas trademark retrieval and retrieval rules specified by the user;
searching overseas trademarks, screening out corresponding target trademark names from an operating memory database of a local server according to the trademark English keywords and the search rules, and screening out corresponding target trademark associated information from a distributed database on a hard disk of the local server according to the target trademark names;
and outputting a retrieval result, and outputting the target trademark name and the corresponding target trademark associated information to a user side.
2. The overseas trademark retrieval method of claim 1, wherein: the retrieval of the overseas trademark specifically comprises the following substeps:
performing word segmentation processing, namely judging whether the trademark English keyword contains at least two key words or not, if so, performing word segmentation processing on the trademark English keyword to obtain a plurality of trademark English key words to be retrieved, and if not, taking the trademark English keyword as the trademark English key word to be retrieved without word segmentation processing;
screening a target trademark name, screening a sample trademark name corresponding to the English key vocabulary of the trademark to be searched from a plurality of sample trademark names stored in an operating memory database according to the search rule as the target trademark name, and obtaining a target trademark number corresponding to the target trademark name;
and screening target trademark associated information, and screening the corresponding target trademark associated information according to the distributed database of the target trademark number on the hard disk of the local server.
3. A method of overseas trademark retrieval as recited in claim 2, wherein: before the target trademark associated information is screened, the method further comprises accurate screening, a first vector score corresponding to the target trademark name is calculated according to a first preset weight and a first occurrence frequency ratio corresponding to the target trademark name, the first occurrence frequency ratio is a ratio of the occurrence frequency of each English letter in the target trademark name in all sample trademark names in the operating memory database to the sum of all English letters in the target trademark name, a second vector score corresponding to the trademark English keyword is calculated according to a second preset weight and a second occurrence frequency ratio corresponding to the trademark English keyword, and the second occurrence frequency ratio is a ratio of the occurrence frequency of each letter in the trademark English keyword in all sample trademark names in the operating memory database to the sum of all English letters in the target trademark name, and calculating a vector score difference value of the first vector score and the second vector score, and reserving the target trademark name with the vector score difference value smaller than a preset vector score difference value threshold value.
4. A method of overseas trademark retrieval as claimed in claim 3, wherein: if the number of the target trademark names with the vector score difference smaller than the preset vector score difference threshold value is larger than ten, sorting the target trademark names with the vector score difference smaller than the preset vector score difference threshold value according to the vector score difference from small to large, and reserving the first ten target names in the sorting.
5. The overseas trademark retrieval method of claim 1, wherein: before receiving the information to be retrieved, the method further comprises the steps of creating a sample database, carrying out data cleaning processing, custom word segmentation processing and segmentation combination processing on sample trademark data containing sample trademark names and sample trademark associated information to obtain a plurality of sample trademark names and corresponding sample trademark associated information, setting a sample trademark number for each sample trademark name and corresponding sample trademark associated information, storing all the sample trademark names and corresponding sample trademark numbers in an operating memory database in the local server, and storing all the sample trademark associated information and corresponding sample trademark numbers in a distributed database on a hard disk of the local server.
6. The overseas trademark retrieval method of claim 1, wherein: in the step of receiving the information to be retrieved, the step of retrieving overseas trademarks is carried out on the batch of information to be retrieved each time the batch of information to be retrieved containing a plurality of trademark English keywords is received, and a corresponding batch query result containing a plurality of target trademark associated information is obtained.
7. The overseas trademark retrieval method of claim 1, wherein: the retrieval rule is a homophone retrieval rule, an add-subtract word retrieval rule, an inclusion/included retrieval rule or an order-changing retrieval rule.
8. The overseas trademark retrieval method of claim 1, wherein: and the method also comprises approximate retrieval, namely inputting the trademark English keyword into a preset examination mechanism model for approximate examination to obtain an approximate examination result.
9. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program is executed by a processor to perform a method of overseas trademark retrieval as claimed in any one of claims 1 to 7.
10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out a method of overseas trademark retrieval of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306150.1A CN114238664A (en) | 2021-11-05 | 2021-11-05 | Overseas trademark retrieval method, equipment, medium and product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306150.1A CN114238664A (en) | 2021-11-05 | 2021-11-05 | Overseas trademark retrieval method, equipment, medium and product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114238664A true CN114238664A (en) | 2022-03-25 |
Family
ID=80748515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111306150.1A Pending CN114238664A (en) | 2021-11-05 | 2021-11-05 | Overseas trademark retrieval method, equipment, medium and product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114238664A (en) |
-
2021
- 2021-11-05 CN CN202111306150.1A patent/CN114238664A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107704102B (en) | Text input method and device | |
| US7567906B1 (en) | Systems and methods for generating an annotation guide | |
| CN112667794A (en) | Intelligent question-answer matching method and system based on twin network BERT model | |
| RU2547213C2 (en) | Assigning actionable attributes to data describing personal identity | |
| CN112069298A (en) | Human-computer interaction method, device and medium based on semantic web and intention recognition | |
| CN114911917B (en) | Asset meta-information searching method and device, computer equipment and readable storage medium | |
| JP2006309446A (en) | Classification dictionary updating device, classification dictionary updating program, and classification dictionary updating method | |
| CN110297880B (en) | Corpus product recommendation method, apparatus, device and storage medium | |
| CN110929498A (en) | Short text similarity calculation method and device and readable storage medium | |
| CN110543555A (en) | method and device for question recall in intelligent customer service | |
| CN112395867B (en) | Synonym mining method and device, storage medium and computer equipment | |
| CN110955767A (en) | Algorithm and device for generating intention candidate set list set in robot dialogue system | |
| CN110795942B (en) | Keyword determination method and device based on semantic recognition and storage medium | |
| CN112395881B (en) | Material label construction method and device, readable storage medium and electronic equipment | |
| CN112506864A (en) | File retrieval method and device, electronic equipment and readable storage medium | |
| CN114238664A (en) | Overseas trademark retrieval method, equipment, medium and product | |
| US20220343087A1 (en) | Matching service requester with service providers | |
| CN113722460B (en) | Index data warehousing method, device, equipment and storage medium | |
| CN114595305A (en) | Intention identification method based on semantic index | |
| CN112416754B (en) | Model evaluation method, terminal, system and storage medium | |
| CN114722086A (en) | Method and device for determining search rearrangement model | |
| CN112765311A (en) | Method for searching referee document | |
| CN112182150A (en) | Aggregation retrieval method, device, equipment and storage medium based on multivariate data | |
| CN112559550A (en) | Multi-data-source NL2SQL system based on semantic rules and multi-dimensional model | |
| CN111984776B (en) | Mechanism name standardization method based on word vector model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |