CN110597863A - Retrieval system and method for keeping stable performance in control library through dynamic threshold - Google Patents
Retrieval system and method for keeping stable performance in control library through dynamic threshold Download PDFInfo
- Publication number
- CN110597863A CN110597863A CN201910909891.5A CN201910909891A CN110597863A CN 110597863 A CN110597863 A CN 110597863A CN 201910909891 A CN201910909891 A CN 201910909891A CN 110597863 A CN110597863 A CN 110597863A
- Authority
- CN
- China
- Prior art keywords
- retrieval
- database
- dynamic threshold
- test
- module
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/242—Query formulation
- G06F16/2425—Iterative querying; Query formulation based on the results of a preceding query
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24553—Query execution of query operations
- G06F16/24561—Intermediate data storage techniques for performance improvement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/248—Presentation of query results
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Fuzzy Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a retrieval system capable of keeping stable performance in a control library through a dynamic threshold, which comprises: the system comprises a retrieval database, a dynamic threshold testing module and a retrieval module. The dynamic threshold testing module is used for testing the retrieval database and forming a dynamic threshold of the retrieval database. And the retrieval module retrieves the retrieval database according to the retrieval request, and forms a retrieval result according to the dynamic threshold value in the retrieval process. The invention also provides a retrieval method for keeping stable performance in the control library through the dynamic threshold. The invention can improve the retrieval stability.
Description
Technical Field
The invention relates to database retrieval, in particular to a retrieval system, and particularly relates to a retrieval system which keeps stable performance in a deployment control library through a dynamic threshold. The invention also relates to a retrieval method for keeping stable performance in the control library through the dynamic threshold.
Background
The retrieval system is used for retrieving required contents in a retrieval database, generally inputting characteristics related to the contents to be retrieved, then comparing in the retrieval database, listing out data similar to the comparison and obtaining a retrieval result.
With the development of Artificial Intelligence (AI), retrieval has also become important. For example, a control library for face recognition may be formed, a large number of face features may be stored in a retrieval database corresponding to the control library for face recognition, the face features to be retrieved may be acquired by cameras arranged at different region positions, and after a retrieval request is input to the retrieval system, face recognition may be performed and a corresponding retrieval result may be output.
In the prior art, when a control library is searched by using a request, all search results are sorted in a similarity comparison mode, and then post-filtering is performed by using a threshold value. The result above the threshold is output as a positive example of the search.
In the existing method, the threshold value adopted by the retrieval is usually fixed and unchanged, but the structural characteristics of the deployment and control library are changed, so that the retrieval performance is unstable. The reason is that the structural structure of the deployment and control library comprises data distribution and data scale, wherein the semantics represented by the threshold value are unstable under different data distributions, such as different data sources and different data scales. Meaning that the same threshold does not have generalization under different data distributions, if the same threshold is used, the performance of retrieval is unstable.
Disclosure of Invention
The invention aims to provide a retrieval system which keeps stable performance in a control library through a dynamic threshold value, and can improve the retrieval stability. Therefore, the invention also provides a retrieval method for keeping stable performance in the control library through the dynamic threshold.
In order to solve the above technical problem, the present invention provides a retrieval system capable of maintaining stable performance in a deployment control library through a dynamic threshold, comprising: the system comprises a retrieval database, a dynamic threshold testing module and a retrieval module.
The dynamic threshold testing module is used for testing the retrieval database and forming a dynamic threshold of the retrieval database.
And the retrieval module retrieves the retrieval database according to the retrieval request, and forms a retrieval result according to the dynamic threshold value in the retrieval process.
In a further improvement, the search database comprises a plurality of search databases, and the structural characteristics of each search database comprise data distribution and data scale.
The dynamic threshold value of the retrieval database is adaptively and dynamically changed according to the structural feature change of the retrieval database.
In a further refinement, the data distribution of the search database includes a distribution formed from data from different sources.
In a further improvement, the structural characteristics of the data distribution and the data scale of the search database have dynamically changing characteristics.
The dynamic threshold test module obtains the dynamic threshold by detecting structural features of the search database.
The further improvement is that after receiving the retrieval request, the retrieval module firstly detects all preliminary retrieval results by a similarity comparison method and sorts the preliminary retrieval results, and then filters the preliminary retrieval results by adopting the dynamic threshold to obtain the retrieval results, wherein the filtering is to use the preliminary retrieval results with the retrieval values higher than the dynamic threshold as the retrieval results.
The further improvement is that the retrieval database is a control database for face recognition, a plurality of face information are stored in the retrieval database, and the retrieval module receives the retrieved face information while receiving the retrieval request.
In a further improvement, the retrieved face information is input through a controlled camera.
In a further improvement, the dynamic threshold test module forms the dynamic threshold by:
and querying the structural feature information of the retrieval database.
And carrying out retrieval test on the retrieval database based on the structural characteristic information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates.
And calculating the required accuracy rate when searching is carried out based on the structural characteristics and calculating the corresponding required false alarm rate based on the accuracy rate.
And taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
In order to solve the technical problem, the retrieval method for maintaining stable performance in the control library through the dynamic threshold comprises the following steps:
step one, providing a dynamic threshold value testing module, testing a retrieval database by using the dynamic threshold value testing module and forming a dynamic threshold value table of the retrieval database.
And step two, the retrieval module retrieves the retrieval database according to the retrieval request, and a retrieval result is formed according to the dynamic threshold value in the retrieval process.
In a further improvement, the search database comprises a plurality of search databases, and the structural characteristics of each search database comprise data distribution and data scale.
The dynamic threshold value of the retrieval database is adaptively and dynamically changed according to the structural feature change of the retrieval database.
In a further refinement, the data distribution of the search database includes a distribution formed from data from different sources.
In a further improvement, the structural characteristics of the data distribution and the data scale of the search database have dynamically changing characteristics.
The dynamic threshold test module obtains the dynamic threshold by detecting structural features of the search database.
The further improvement is that after receiving the retrieval request, the retrieval module firstly detects all preliminary retrieval results by a similarity comparison method and sorts the preliminary retrieval results, and then filters the preliminary retrieval results by adopting the dynamic threshold to obtain the retrieval results, wherein the filtering is to use the preliminary retrieval results with the retrieval values higher than the dynamic threshold as the retrieval results.
The further improvement is that the retrieval database is a control database for face recognition, a plurality of face information are stored in the retrieval database, and the retrieval module receives the retrieved face information while receiving the retrieval request.
In a further improvement, the retrieved face information is input through a controlled camera.
In a further improvement, the dynamic threshold test module forms the dynamic threshold by:
and querying the structural feature information of the retrieval database.
And carrying out retrieval test on the retrieval database based on the structural characteristic information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates.
And calculating the required accuracy rate when searching is carried out based on the structural characteristics and calculating the corresponding required false alarm rate based on the accuracy rate.
And taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
The invention sets dynamic threshold test module in the search system, which can detect the dynamic threshold of the search database, and uses dynamic threshold to form and output the search result in the search process, therefore, the invention can realize the dynamic change of threshold along with the structure characteristic of the search database, when the structure characteristic of the search database changes, such as the data distribution and the data scale, the invention can still realize accurate output, thereby improving the stability of search.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is a block diagram of a retrieval system in accordance with an embodiment of the present invention;
FIG. 2 is a flow chart of the dynamic threshold test module in the retrieval system for calculating the dynamic threshold according to the embodiment of the present invention.
Detailed Description
FIG. 1 is a block diagram of a search system according to an embodiment of the present invention; the retrieval system capable of keeping stable performance in a control library through a dynamic threshold comprises the following steps: the system comprises a retrieval database 1, a dynamic threshold testing module 2 and a retrieval module 3.
The dynamic threshold test module 2 is configured to test the search database 1 and form a dynamic threshold of the search database 1.
And the retrieval module 3 retrieves the retrieval database 1 according to the retrieval request, and forms a retrieval result according to the dynamic threshold value in the retrieval process.
The search database 1 includes a plurality of search databases, and the structural characteristics of each search database 1 include data distribution and data size. The data distribution of the search database 1 includes a distribution formed by data from different sources, for example, data from different provinces. The retrieval database 1 can be used as a control library for face recognition, a plurality of face information are stored in the retrieval database 1, and the retrieval module 3 receives the retrieved face information while receiving the retrieval request. The retrieved face information is input through a controlled camera, and the retrieved face information can also be input manually. Wherein, the manual input can input text information; the input of the camera through control is to input graphic information, and the graphic information needs to be subjected to feature extraction through an identification module and converted into text information.
The structural features of the data distribution and data scale of the search database 1 have dynamically changing features, such as: the data source corresponding to the data distribution can be continuously expanded, so that the data distribution is wider and wider; the data scale is also expanded continuously, so that the data scale is larger and larger; and vice versa.
The dynamic threshold of the search database 1 is adaptively and dynamically changed according to the structural feature change of the search database 1.
The dynamic threshold test module 2 obtains the dynamic threshold by detecting the structural features of the search database 1.
After receiving the retrieval request, the retrieval module 3 firstly detects all preliminary retrieval results by a similarity comparison method and sorts the preliminary retrieval results, and then filters the preliminary retrieval results by using the dynamic threshold to obtain the retrieval results, wherein the filtering is to use the preliminary retrieval results with retrieval values higher than the dynamic threshold as the retrieval results.
Fig. 2 is a flowchart illustrating the dynamic threshold value test module in the search system according to an embodiment of the present invention calculating the dynamic threshold value. The dynamic threshold testing module 2 forms the dynamic threshold by adopting the following steps:
and inquiring the structural feature information of the retrieval database 1.
And carrying out retrieval test on the retrieval database 1 based on the structural characteristic information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates.
And calculating the required accuracy rate when searching is carried out based on the structural characteristics and calculating the corresponding required false alarm rate based on the accuracy rate.
And taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
In other embodiments can also be: providing the retrieval database corresponding to a series of structural features in advance, and then testing the dynamic threshold corresponding to each structural feature according to the above process to form a dynamic threshold table; then, in the actual retrieval process, the dynamic threshold test module 2 only needs to detect the structural feature information of the retrieval database, and select the corresponding dynamic threshold from the dynamic threshold table according to the structural feature obtained by detection.
The embodiment of the invention is provided with the dynamic threshold testing module 2 in the retrieval system, the dynamic threshold of the retrieval database 1 can be detected by the dynamic threshold testing module 2, and the retrieval result is formed and output by adopting the dynamic threshold in the retrieval process, so that the embodiment of the invention can realize that the threshold dynamically follows the structural characteristic change of the retrieval database 1, and when the structural characteristic of the retrieval database 1 changes, such as the data distribution, the data scale and the like, the embodiment of the invention can still realize accurate output, thereby improving the retrieval stability.
The retrieval method provided by the embodiment of the invention comprises the following steps:
step one, providing a dynamic threshold value testing module 2, testing a retrieval database 1 by using the dynamic threshold value testing module 2 and forming a dynamic threshold value table of the retrieval database 1.
The search database 1 includes a plurality of search databases, and the structural characteristics of each search database 1 include data distribution and data size. The data distribution of the search database 1 includes a distribution formed by data from different sources, for example, data from different provinces. The retrieval database 1 can be used as a control library for face recognition, a plurality of face information are stored in the retrieval database 1, and the retrieval module 3 receives the retrieved face information while receiving the retrieval request. The retrieved face information is input through a controlled camera, and the retrieved face information can also be input manually. Wherein, the manual input can input text information; the input of the camera through control is to input graphic information, and the graphic information needs to be subjected to feature extraction through an identification module and converted into text information.
The structural features of the data distribution and data scale of the search database 1 have dynamically changing features, such as: the data source corresponding to the data distribution can be continuously expanded, so that the data distribution is wider and wider; the data scale is also expanded continuously, so that the data scale is larger and larger; and vice versa.
The dynamic threshold of the search database 1 is adaptively and dynamically changed according to the structural feature change of the search database 1.
The dynamic threshold test module 2 obtains the dynamic threshold by detecting the structural features of the search database 1.
Fig. 2 is a flowchart illustrating the dynamic threshold value test module in the search system according to an embodiment of the present invention calculating the dynamic threshold value. The dynamic threshold testing module 2 forms the dynamic threshold by adopting the following steps:
and inquiring the structural feature information of the retrieval database 1.
And carrying out retrieval test on the retrieval database 1 based on the structural characteristic information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates.
And calculating the required accuracy rate when searching is carried out based on the structural characteristics and calculating the corresponding required false alarm rate based on the accuracy rate.
And taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
In other embodiments can also be: providing the retrieval database corresponding to a series of structural features in advance, and then testing the dynamic threshold corresponding to each structural feature according to the above process to form a dynamic threshold table; then, in the actual retrieval process, the dynamic threshold test module 2 only needs to detect the structural feature information of the retrieval database, and select the corresponding dynamic threshold from the dynamic threshold table according to the structural feature obtained by detection.
And step two, the retrieval module 3 retrieves the retrieval database 1 according to the retrieval request, and forms a retrieval result according to the dynamic threshold value in the retrieval process.
Preferably, after receiving the search request, the search module 3 first detects all preliminary search results by a similarity comparison method and sorts the preliminary search results, and then filters the preliminary search results by using the dynamic threshold to obtain the search results, where the filtering is to use the preliminary search results with a search value higher than the dynamic threshold as the search results.
The present invention has been described in detail with reference to the specific embodiments, but these should not be construed as limitations of the present invention. Many variations and modifications may be made by one of ordinary skill in the art without departing from the principles of the present invention, which should also be considered as within the scope of the present invention.
Claims (16)
1. A retrieval system for maintaining stable performance in a deployed library through dynamic thresholds, comprising: the system comprises a retrieval database, a dynamic threshold testing module and a retrieval module;
the dynamic threshold test module is used for testing the retrieval database and forming a dynamic threshold of the retrieval database;
and the retrieval module retrieves the retrieval database according to the retrieval request, and forms a retrieval result according to the dynamic threshold value in the retrieval process.
2. The retrieval system of claim 1, wherein: the retrieval database comprises a plurality of retrieval databases, and the structural characteristics of each retrieval database comprise data distribution and data scale;
the dynamic threshold value of the retrieval database is adaptively and dynamically changed according to the structural feature change of the retrieval database.
3. The retrieval system of claim 2, wherein: the data distribution of the search database includes a distribution formed by data from different sources.
4. The retrieval system of claim 2, wherein: the structural characteristics of the data distribution and the data scale of the retrieval database have the characteristics of dynamic change;
the dynamic threshold test module obtains the dynamic threshold by detecting structural features of the search database.
5. A retrieval system as recited in claim 3, wherein: after receiving the retrieval request, the retrieval module firstly detects all preliminary retrieval results through a similarity comparison method and sorts the preliminary retrieval results, and then filters the preliminary retrieval results by adopting the dynamic threshold to obtain the retrieval results, wherein the filtering is to take the preliminary retrieval results with retrieval values higher than the dynamic threshold as the retrieval results.
6. The retrieval system of claim 1, wherein: the retrieval database is a control database for face recognition, a plurality of face information are stored in the retrieval database, and the retrieval module receives the retrieval request and the retrieved face information at the same time.
7. The retrieval system of claim 6, wherein: and the retrieved face information is input through the controlled camera.
8. The retrieval system of claim 4, wherein: the dynamic threshold testing module forms the dynamic threshold by adopting the following steps:
inquiring the structural feature information of the retrieval database;
carrying out retrieval test on the retrieval database based on the structural feature information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates;
calculating the required accuracy rate when searching is carried out based on the structural features and calculating the corresponding required false alarm rate based on the accuracy rate;
and taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
9. A retrieval method for keeping stable performance in a deployment control library through a dynamic threshold is characterized by comprising the following steps:
providing a dynamic threshold testing module, testing a retrieval database by using the dynamic threshold testing module and forming a dynamic threshold table of the retrieval database;
and step two, the retrieval module retrieves the retrieval database according to the retrieval request, and a retrieval result is formed according to the dynamic threshold value in the retrieval process.
10. The retrieval method of claim 9, wherein: the retrieval database comprises a plurality of retrieval databases, and the structural characteristics of each retrieval database comprise data distribution and data scale;
the dynamic threshold value of the retrieval database is adaptively and dynamically changed according to the structural feature change of the retrieval database.
11. The retrieval method of claim 10, wherein: the data distribution of the search database includes a distribution formed by data from different sources.
12. The retrieval method of claim 10, wherein: the structural characteristics of the data distribution and the data scale of the retrieval database have the characteristics of dynamic change;
the dynamic threshold test module obtains the dynamic threshold by detecting structural features of the search database.
13. The retrieval method of claim 10, wherein: after receiving the retrieval request, the retrieval module firstly detects all preliminary retrieval results through a similarity comparison method and sorts the preliminary retrieval results, and then filters the preliminary retrieval results by adopting the dynamic threshold to obtain the retrieval results, wherein the filtering is to take the preliminary retrieval results with retrieval values higher than the dynamic threshold as the retrieval results.
14. The retrieval method of claim 9, wherein: the retrieval database is a control database for face recognition, a plurality of face information are stored in the retrieval database, and the retrieval module receives the retrieval request and the retrieved face information at the same time.
15. The retrieval method of claim 14, wherein: and the retrieved face information is input through the controlled camera.
16. The retrieval method of claim 12, wherein: the dynamic threshold testing module forms the dynamic threshold by adopting the following steps:
inquiring the structural feature information of the retrieval database;
carrying out retrieval test on the retrieval database based on the structural feature information, wherein a series of test thresholds are adopted in the retrieval test, and different test thresholds obtain corresponding false alarm rates to form a one-to-one corresponding relation between the test thresholds and the false alarm rates;
calculating the required accuracy rate when searching is carried out based on the structural features and calculating the corresponding required false alarm rate based on the accuracy rate;
and taking the test threshold corresponding to the required false alarm rate as the dynamic threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910909891.5A CN110597863B (en) | 2019-09-25 | 2019-09-25 | Retrieval system and method for keeping stable performance in control library through dynamic threshold |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910909891.5A CN110597863B (en) | 2019-09-25 | 2019-09-25 | Retrieval system and method for keeping stable performance in control library through dynamic threshold |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110597863A true CN110597863A (en) | 2019-12-20 |
| CN110597863B CN110597863B (en) | 2023-01-24 |
Family
ID=68863099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910909891.5A Active CN110597863B (en) | 2019-09-25 | 2019-09-25 | Retrieval system and method for keeping stable performance in control library through dynamic threshold |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110597863B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104503920A (en) * | 2015-01-23 | 2015-04-08 | 桂林电子科技大学 | Software system fault detection method based on dynamic description logic and case-based reasoning |
| CN106295672A (en) * | 2015-06-12 | 2017-01-04 | 中国移动（深圳）有限公司 | A kind of face identification method and device |
| CN106372226A (en) * | 2016-09-07 | 2017-02-01 | 知识产权出版社有限责任公司 | Information retrieval device and method |
| CN108241836A (en) * | 2016-12-23 | 2018-07-03 | 同方威视技术股份有限公司 | For the method and device of safety check |
| US20180333855A1 (en) * | 2017-05-19 | 2018-11-22 | Stc.Unm | System and methods for multiple-place swarm foraging with dynamic depots |
| CN109711062A (en) * | 2018-12-28 | 2019-05-03 | 广东电网有限责任公司 | A kind of equipment fault diagnosis method and device based on cloud service |
-
2019
- 2019-09-25 CN CN201910909891.5A patent/CN110597863B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104503920A (en) * | 2015-01-23 | 2015-04-08 | 桂林电子科技大学 | Software system fault detection method based on dynamic description logic and case-based reasoning |
| CN106295672A (en) * | 2015-06-12 | 2017-01-04 | 中国移动（深圳）有限公司 | A kind of face identification method and device |
| CN106372226A (en) * | 2016-09-07 | 2017-02-01 | 知识产权出版社有限责任公司 | Information retrieval device and method |
| CN108241836A (en) * | 2016-12-23 | 2018-07-03 | 同方威视技术股份有限公司 | For the method and device of safety check |
| US20180333855A1 (en) * | 2017-05-19 | 2018-11-22 | Stc.Unm | System and methods for multiple-place swarm foraging with dynamic depots |
| CN109711062A (en) * | 2018-12-28 | 2019-05-03 | 广东电网有限责任公司 | A kind of equipment fault diagnosis method and device based on cloud service |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110597863B (en) | 2023-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111324784B (en) | Character string processing method and device | |
| CN103207898B (en) | A kind of similar face method for quickly retrieving based on local sensitivity Hash | |
| CN104572717B (en) | Information searching method and device | |
| CN111368867B (en) | File classifying method and system and computer readable storage medium | |
| US11450107B1 (en) | Dynamic detection and recognition of media subjects | |
| US20030033138A1 (en) | Method for partitioning a data set into frequency vectors for clustering | |
| CN102314464A (en) | Lyrics searching method and lyrics searching engine | |
| CN110413997B (en) | New word discovery method, system and readable storage medium for power industry | |
| CN110688572A (en) | Method for identifying search intention in cold starting state | |
| CN111984673B (en) | Fuzzy retrieval method and device for tree structure of power grid electric energy metering system | |
| CN110597863B (en) | Retrieval system and method for keeping stable performance in control library through dynamic threshold | |
| CN105512270B (en) | Method and device for determining related objects | |
| CN110019637A (en) | A kind of sort algorithm of normative document retrieval | |
| CN114511027B (en) | Method for extracting English remote data through big data network | |
| CN112965998B (en) | Method and system for establishing and retrieving compound database | |
| US11386340B2 (en) | Method and apparatus for performing block retrieval on block to be processed of urine sediment image | |
| Pardede et al. | Comparison of similarity measures in HSV quantization for CBIR | |
| CN111539576B (en) | Risk identification model optimization method and device | |
| CN113326688A (en) | Ideological and political theory word duplication checking processing method and device | |
| US20050198059A1 (en) | Database and database management system | |
| KR102655234B1 (en) | Method and apparatus for retrieving packet at high-speed | |
| CN116860977B (en) | Abnormality detection system and method for contradiction dispute mediation | |
| CN110633430A (en) | Event discovery method, device, equipment and computer readable storage medium | |
| CN117252183B (en) | Semantic-based multi-source table automatic matching method, device and storage medium | |
| CN103348348A (en) | Information search apparatus and information search method |
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 |