CN110096553A - A kind of the big data analysis system and analysis method of integration across database - Google Patents
A kind of the big data analysis system and analysis method of integration across database Download PDFInfo
- Publication number
- CN110096553A CN110096553A CN201910242794.5A CN201910242794A CN110096553A CN 110096553 A CN110096553 A CN 110096553A CN 201910242794 A CN201910242794 A CN 201910242794A CN 110096553 A CN110096553 A CN 110096553A
- Authority
- CN
- China
- Prior art keywords
- service
- database
- data
- module
- term vector
- 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
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
Abstract
The present invention relates to the big data analysis systems and analysis method of a kind of integration across database.Analysis system includes service Understanding Module, service creation module and service database, and service Understanding Module is used to generate intelligent Service model by deep approach of learning, polymerize provided by each isolated data library and services, and by available service typing service database.Service creation module selects corresponding service to be pushed for responding new service request from service database.Service database is used to store the new service after polymerization.Analysis method of the invention assigns the data in mutual isolated data library with common meaning and similarity, similar services association is condensed together, provides the direct service of integration across database for upper layer application from natural language understanding angle.It solves the problems, such as the logic sexual isolation of data and the reaction to application service, each isolated data library is allowed to be compatible with each other, allow the data in database that mutually there is relevance, more interact, it is more valuable.
Description
Technical field
The present invention relates to big data analysis fields, such as government affairs big data analysis, smart city big data analysis etc., specifically
It is related to the big data analysis system and analysis method of a kind of integration across database.
Background technique
Data silo refer to the data in each database can not (or extremely difficult) connection interaction, i.e., lack between data
Relevance, database are isolated each other, can not be compatible with.Data silo is divided into physical and two kinds of logicality.Physical data are lonely
Island refers to that data store independently of each other in different departments, independent maintenance, to each other independently of each other, forms number physically
According to isolation.The data silo of logicality refers to that different departments are understood and defined to data from the angle of oneself, so that
Some identical data have been assigned different meanings, virtually increase the communication cost of trans-departmental data files, make originally
Can the interrelated, data that interlock each other, become wide of the mark, respectively isolated data.
Data silo problem is generally existing in enterprises, and enterprise development to certain phase multiple division departments occurs, each
There are respective data in division department, the often all respectively storages of the data between division department, each customized.The especially enterprise of grouping of the world economy
With regard to more obvious, the department of most of conglomerate is divided based on functional form, relatively independent between department and department, in enterprise
Each department can generate corresponding data, but each department is different to the understanding angle of data, to data use and definition has
Bigger difference, cause data can not intercommunication, formed isolated island.On the other hand, information departments' construction in many enterprises is compared
In evening, the standard disunity of Information System configuration, there are biggish obstructions for the data interchange for making in the future.
Eliminating data silo is one long-term difficult work, and the most common method is data correlation, makes original phase not
The data of pass, with increasing for associated data, data dimension increases, and the value that can be excavated becomes larger, to make to can't see originally too
The data being worth greatly generate immense value.But often there are numerous obstacles in data correlation, wherein maximum hinder to be every number
According to library, table structure all in close relations, close-coupleds with application program, when the data in database are departing from original existence ring
Border is sent to after other information systems, without corresponding table knot between the data in the information system due to receiving data
Structure, also without corresponding coupled relation, data are just at meaningless, unworthy data.
It is technically more mature to eliminate physical data silo, removing method i.e. will be in these data sets or distributed
Unified management.As shown in Figure 1, measured big data Hadoop platform be selection one of, Hadoop be one can be to sea
The software frame that data carry out distributed treatment is measured, Hadoop is carried out at data in a reliable, efficient and scalable way
Reason, provides distributed memory, distributed computing, distributed data base of data etc., can unify the data of storage, management isolation,
Eliminate the physical isolation between data.But this storage mode can not eliminate the logic sexual isolation of data, in disparate databases
Data understanding and definition it is still different, association, data between database are unable to get satisfaction, data to the reaction of application
It can not become valuable, the application data of service can be directly provided.
Summary of the invention
Against the above deficiency, it the present invention provides the big data analysis system and analysis method of a kind of integration across database, is used for
Solve the problems, such as the logic sexual isolation of data and the reaction to application service.The present invention assigns mutual from natural language understanding angle
Data in isolated data library by similar services association fusion, are condensed together with common meaning and similarity, are upper layer application
The direct service of integration across database is provided.
The technical solution of the present invention is as follows:
A kind of big data analysis system of integration across database, including service Understanding Module, service creation module and service data
Library, the service Understanding Module are used to generate intelligent Service model by deep approach of learning, polymerize each isolated data library and mentioned
The service of confession, by service database described in available service typing, the service creation module is asked for responding new service
It asks, corresponding service is selected to be pushed from service database, the service database is used to store the new clothes after polymerization
Business.
It further include manual intervention module, the manual intervention module is used to carry out the selected service of service creation module
Manual amendment, and the content of modification is supplied to service Understanding Module, intelligent Service model is updated.
The deep approach of learning is the method for machine learning natural language, the clothes that this method provides all isolated data libraries
Business is clustered semantically, together by close service aggregating, provides the service of integration across database.
The intelligent Service model is indicated using term vector embedding inlay technique.
The intelligent Service model is by natural language processing deep learning model, service word model and service Clustering Model group
At.
A kind of big data analysis method of integration across database, comprising the following steps:
S1: service available initialization list is established in initialization service；
S2: major key, the keyword in multiple isolated data libraries are included in by the deep learning dictionary of initialization service Understanding Module
In dictionary, extension dictionary is formed；
S3: service Understanding Module analyzes multiple databases operation syntactic analysis, part of speech, forms service available key
Term vector data list；
S4: service Understanding Module establishes intelligent Service model using term vector embedding inlay technique；
S5: service Understanding Module calculates the distance between each crucial term vector, and measuring similarity is established between term vector；
S6: service Understanding Module is according to preset similarity threshold, by the similarity and threshold value between each term vector
It makes comparisons, the similar services key assignments that similarity is greater than threshold value is merged；
S7: service Understanding Module carries out data analysis to database corresponding to each similar services, will be new after merging
Information on services be stored in service database；
S8: intelligent Service model receives the new service request of upper layer application sending, passes through syntactic analysis, part of speech point
Analysis, crucial word segmentation, obtain the keyword vector lists of service request；
S9: service creation module makees the key assignments term vector serviced in the crucial term vector of service request and service database
Compare, calculates similarity between the two；
S10: service creation module sorts from large to small similarity, the multiple clothes for the setting number for selecting similarity forward
Business is used as candidate service；
S11: candidate service is pushed to upper layer application by service creation module.
The comparison procedure of the step S9 specifically: by all clothes in the crucial term vector of service request and service database
The key assignments term vector of business is made comparisons.
The comparison procedure of the step S9 specifically: establish service database key assignments search tree, retrieved and taken using search tree
The service key assignments being engaged in database, the crucial term vector of service request is made comparisons with the key assignments term vector retrieved.
Further comprising the steps of between the step S10 and S11: the manual intervention module is to selected by service creation module
The candidate service selected carries out manual amendment, regard modified service as candidate service.
It, will be in modification after the manual intervention module carries out manual amendment to the selected candidate service of service creation module
Appearance is supplied to service Understanding Module, is updated to intelligent Service model.
The big data analysis system and analysis method of integration across database of the invention, generate intelligent Service by deep approach of learning
Model, intelligent Service model handle each isolated data library by modes such as machine learning, data mining, statistical analysis, retrievals
In service data, by it is new, valuable, can directly provide service can service data storage to service database.It utilizes
Intelligent Service model handles service request, by the coherence measurement to service key term vector, makes service key word, data
Library key assignments etc. obtains semantic consistency on vector similarity, and similar service is pushed to upper layer application, breaks each only
It is uncorrelated between vertical database, establish the relevance between data in service.
Big data analysis system of the invention and analysis method also have the function of manual intervention, right according to the feedback of user
The service of selection carries out manual amendment, optimization, makes the service more exchange premium user deep layer provided, implicit needs.It is artificial dry
Pre- result is iterated update to intelligent Service model simultaneously, make comprising can the service database of service data constantly agree with user
Actual demand, while be also the association of bottom data, extract can service data provide relevant cue and direction.
Detailed description of the invention
Fig. 1 is Hadoop distributed system infrastructure figure；
Fig. 2 is the big data analysis system architecture diagram of integration across database of the present invention；
Fig. 3 is the work relationship figure of present invention service Understanding Module；
Fig. 4 is the work step figure of present invention service Understanding Module；
Fig. 5 is the work step figure of service creation module of the present invention；
Fig. 6 is service key word and search schematic diagram of the present invention.
Specific embodiment
It is described further below with reference to technical effect of the attached drawing to design of the invention, specific structure and generation, with
It is fully understood from the purpose of the present invention, feature and effect.
With reference to Fig. 2, the big data analysis system of integration across database of the invention, including service Understanding Module, service creation mould
Block, service database and manual intervention module.It services Understanding Module to be used to generate intelligent Service model by deep approach of learning, gather
Close service provided by each isolated data library, and by available service typing service database.Service creation module is used for
New service request is responded, corresponding service is selected to be pushed from service database.Service database is for storing polymerization
New service afterwards.Manual intervention module is used to carry out the selected service of service creation module manual amendment, and will modification
Content be supplied to service Understanding Module, intelligent Service model is updated.
As shown in Fig. 2, database 1, database 2 ..., database n are each isolated database, provided information clothes
Business content is not quite similar, and each database includes major key and provided stand-alone service, inputs in the form of text-string.?
In the expression way of text, even if the manifestation mode of text still may different from comprising same meaning.Such as " gas
As ", " weather ", similar key assignments word, the provided information service such as " weather forecast " be " weather data ".
With reference to Fig. 3 and Fig. 4, the deep approach of learning for servicing Understanding Module is the method for machine learning natural language, utilizes this
Method will be serviced come keyword, the database key assignments etc. for handling each isolated data library service provided by similarity measurement
It is clustered semantically, together by close service aggregating.Break uncorrelated between each self contained data base, establishes between data
The association in service pass through machine learning, data mining, statistical analysis, retrieval etc. on the basis of this associated data set
Method handle natural language, generate it is new, valuable can service data.
Intelligent Service model is made of natural language processing deep learning model, service word model and service Clustering Model,
And it is indicated using term vector embedding inlay technique.For lteral data, if using label coder mode (Label Encoder)
It encodes, if the ID value of different vocabulary is very close, can not there is an actual meaning representation.If using discrete type feature
Coding mode (One hot) encodes, then it is excessively high to will lead to vector dimension, excessively sparse, while being also still difficult to numerically
Represent the relationship between different words.
Term vector embedding inlay technique can read word from urtext (corpus) and then generate term vector, find a kind of word
With the mapping relations of vector, so that vector dimension does not need excessive, and term vector point represented in vector space has
Actual meaning, that is, the word distance in space of similar meaning are closer.For the present invention, term vector embedding inlay technique can be with
Service key word distance in space is preferably calculated, mutual distance is closer, similar several services to find.
Service Understanding Module of the invention intelligent Service model generated is applied not only to mention multiple isolated data libraries
The similar services of confession condense together, and the similar services after polymerization are included in service database, merge service database storage
Afterwards, it is valuable, service can be directly provided can service data.Meanwhile intelligent Service model is also used to answer reception upper layer
Service key term vector column are obtained by syntactic analysis, part of speech analysis, crucial word segmentation with sending, new service request
Table.
With reference to Fig. 5, service creation module of the invention is used for will be in the crucial term vector of service request and service database
The key assignments term vector of the service of storage is made comparisons, and space length between the two is calculated, and it is closer, more similar to find space length
Service, similarity is sorted from large to small, according to preset service push number, match forward multiple service conducts
Candidate service is pushed to upper layer application.
As shown in fig. 6, in the matching process, the lists of keywords of service request is not necessarily completely right in service database
It answers, such as service request keyword is " building price, building market value ", and without corresponding key assignments in service database, but have
Key assignments such as relevant " building assets ", it is therefore desirable to which " building price, building market value " is transformed to key assignments " building by service retrieval tree
Assets " could select corresponding service from service database.
If service database scale is smaller, key assignments small scale can take full matching process, calculate in service database
The similarity of all service key assignments words and service request key term vector.If service database is larger, key assignments scale is big,
It can establish service database key assignments search tree, using search tree retrieval service key assignments, the service key assignments that then will only retrieve
Word is made comparisons with service request keyword, is accelerated retrieval and is compared speed.
Manual intervention module can carry out manual amendment, optimization to the service that service creation module selects, and module is equipped with letter
Interactive interface is ceased, information exchange is manually carried out by information interactive interface and system.Data source is modified in upper layer application user
Feedback, modified according to the feedback opinion of user to selection result, make big data analysis system provide service more stick on
Into user's deep layer, implicit needs.Manual intervention result is iterated update to intelligent Service model simultaneously, makes comprising that can take
The service database of business data constantly agrees with the actual demand of user, while being also the association of bottom data, extracting and can service number
According to providing relevant cue and direction.
Service database is used to store the service after polymerizeing, service request and can be between service data multi-to-multi pass
System, a service request may need it is a plurality of can service data, one can service data can be a variety of different application scenarios
Service is provided.Can service data have the characteristics that instantaneity, reusable, can Rapid Combination, switching, instant sex service is provided,
The requirement of real-time of the Site Services (such as emergency command) such as satisfaction.
One can service data example it is as follows:
Service database supports the requirement description of a variety of users, including bottom data statistic analysis result, business to certain
The requirement description of data and the management data managed data itself especially support the natural language description of data.Passing through will
The Demand mapping of user has got through consumption of the user to data to service database.By mapping can service data be pushed to
Layer applies user, and upper layer application includes assisting the plurality of application scenes such as commander's detection, demonstration report, production run, emergency command,
User is all presented to by visual mode, user is allowed to see data, understand data, gets through data silo, maximizes and plays
The effect of data makes meaningless data originally, becomes significant, valuable.
The big data analysis method of integration across database of the invention, comprising the following steps:
S1: service available initialization list, operation experience generally previous according to operator are established in initialization service
To determine the service that can be provided；
S2: major key, the keyword in multiple isolated data libraries are included in by the deep learning dictionary of initialization service Understanding Module
In dictionary, extension dictionary is formed；
S3: service Understanding Module analyzes multiple databases operation syntactic analysis, part of speech, forms service available key
Term vector；
S4: service Understanding Module establishes intelligent Service model using term vector embedding inlay technique；
S5: service Understanding Module calculates the distance between each crucial term vector, and measuring similarity, word are established between term vector
Vector supports various distances, such as Euclidean distance, manhatton distance, Chebyshev's distance, Minkowski Distance, mahalanobis distance
With included angle cosine distance etc.；
S6: service Understanding Module is according to preset similarity threshold, by the similarity and threshold value between each term vector
It makes comparisons, the similar services key assignments that similarity is greater than threshold value is merged；If the quantity in isolated data library is more, service key assignments number
Amount is also more, and term vector similarity calculates the clustering algorithm that can use machine learning, completes the polymerization of same class service；
S7: service Understanding Module carries out data analysis to database corresponding to each similar services, will be new after merging
Service be stored in service database；If the initial data that some service relies on is after service Understanding Module analysis, polymerization
The new data set of integration across database, new data set are extracted the content of initial data, are directly stored on service database；Such as
The data that fruit new demand servicing relies on are initial data, and service database then provides raw data base path, guide service generation module
Find corresponding service；
S8: intelligent Service model receives the new service request of upper layer application sending, passes through syntactic analysis, part of speech point
Analysis, crucial word segmentation, obtain service key term vector list；
S9: service creation module makees the service key assignments term vector in the crucial term vector of service request and service database
Compare, calculates similarity between the two；Service database lesser for scale takes full matching process, calculates service number
According to the similarity of service key assignments words and service request key term vector all in library；For larger service database, build
Vertical service database key assignments search tree accelerates retrieval rate using search tree retrieval service key assignments；
S10: service creation module sorts from large to small calculated similarity, according to preset push number,
The multiple services for selecting similarity forward are as candidate service；Meanwhile service creation module also receives the dry of artificial intervention module
In advance, manual amendment is carried out to the selected candidate service of service creation module, regard modified service as candidate service；Manually
After intervention module carries out manual amendment to the selected candidate service of service creation module, modification content is supplied to service and is understood
Module is updated intelligent Service model；
S11: candidate service is pushed to upper layer application by service creation module.
The big data analysis system and analysis method of integration across database of the invention assign phase from natural language understanding angle
Data in mutual isolated data library by similar services association fusion, are condensed together, are answered for upper layer with common meaning and similarity
With the direct service of offer integration across database.The present invention solves the problems, such as the logic sexual isolation of data and the reaction to application service,
It allows each isolated data library to be compatible with each other, allows the data in database that mutually there is relevance, more interact, it is more valuable.
Disclosed above is only the embodiment of the present invention, and still, the present invention is not limited to this, the technology of any this field
What personnel can think variation should all fall into protection scope of the present invention.
Claims (10)
1. a kind of big data analysis system of integration across database, which is characterized in that including service Understanding Module, service creation module and
Service database, the service Understanding Module are used to generate intelligent Service model by deep approach of learning, polymerize each isolated number
It is serviced according to provided by library, by service database described in available service typing, the service creation module is new for responding
Service request, select corresponding service to be pushed from service database, the service database is for after storing and polymerizeing
New service.
2. big data analysis system according to claim 1, which is characterized in that further include manual intervention module, the people
Work intervention module is used to carry out manual amendment to the selected service of service creation module, and the content of modification is supplied to service
Understanding Module is updated intelligent Service model.
3. big data analysis system according to claim 1, which is characterized in that the deep approach of learning be machine learning from
The method of right language, this method clusters the service that all isolated data libraries provide semantically, by close service aggregating one
It rises, the service of integration across database is provided.
4. big data analysis system according to claim 1, which is characterized in that the intelligent Service model uses term vector
Embedding inlay technique indicates.
5. big data analysis system according to claim 1, which is characterized in that the intelligent Service model is by natural language
Handle deep learning model, service word model and service Clustering Model composition.
6. a kind of big data analysis method of integration across database, which comprises the following steps:
S1: service available initialization list is established in initialization service；
S2: major key, the keyword in multiple isolated data libraries are included in dictionary by the deep learning dictionary of initialization service Understanding Module
In, form extension dictionary；
S3: service Understanding Module to multiple databases operation syntactic analysis, part of speech analyze, formed service available keyword to
Measure data list；
S4: service Understanding Module establishes intelligent Service model using term vector embedding inlay technique；
S5: service Understanding Module calculates the distance between each crucial term vector, and measuring similarity is established between term vector；
S6: service Understanding Module makees ratio with threshold value according to preset similarity threshold, by the similarity between each term vector
Compared with the similar services key assignments by similarity greater than threshold value merges；
S7: service Understanding Module carries out data analysis to database corresponding to each similar services, by the new clothes after merging
Information of being engaged in is stored in service database；
S8: intelligent Service model receives the new service request of upper layer application sending, by syntactic analysis, part of speech analysis, closes
Keyword cutting obtains the keyword vector lists of service request；
S9: the key assignments term vector serviced in the crucial term vector of service request and service database is made ratio by service creation module
Compared with calculating similarity between the two；
S10: service creation module sorts from large to small similarity, and the multiple services for the setting number for selecting similarity forward are made
For candidate service；
S11: candidate service is pushed to upper layer application by service creation module.
7. big data analysis method according to claim 6, which is characterized in that the comparison procedure of the step S9 is specific
Are as follows: the crucial term vector of service request is made comparisons with the key assignments term vector of services all in service database.
8. big data analysis method according to claim 6, which is characterized in that the comparison procedure of the step S9 is specific
Are as follows: service database key assignments search tree is established, using the service key assignments in search tree retrieval service database, by service request
Crucial term vector is made comparisons with the key assignments term vector retrieved.
9. big data analysis method according to claim 6, which is characterized in that further include between the step S10 and S11
Following steps: the manual intervention module carries out manual amendment to the selected candidate service of service creation module, after modification
Service as candidate service.
10. big data analysis method according to claim 9, which is characterized in that the manual intervention module is to waiter
After carrying out manual amendment at the selected candidate service of module, modification content is supplied to service Understanding Module, to intelligent Service
Model is updated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910242794.5A CN110096553B (en) | 2019-03-28 | 2019-03-28 | Cross-database big data analysis system and analysis method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910242794.5A CN110096553B (en) | 2019-03-28 | 2019-03-28 | Cross-database big data analysis system and analysis method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110096553A true CN110096553A (en) | 2019-08-06 |
| CN110096553B CN110096553B (en) | 2021-05-18 |
Family
ID=67444063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910242794.5A Active CN110096553B (en) | 2019-03-28 | 2019-03-28 | Cross-database big data analysis system and analysis method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110096553B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765121A (en) * | 2019-10-31 | 2020-02-07 | 太原蓝知科技有限公司 | Intelligent big data acquisition and processing method and system |
| CN112199424A (en) * | 2020-09-07 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Method for pulling communication information island through RPA, readable storage medium and RPA robot |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957968A (en) * | 2010-08-31 | 2011-01-26 | 南京财经大学 | Online transaction service aggregation method based on Hadoop |
| CN103136364A (en) * | 2013-03-14 | 2013-06-05 | 曙光信息产业(北京)有限公司 | Cluster database system and data query processing method thereof |
| EP2722753A1 (en) * | 2012-10-19 | 2014-04-23 | Alcatel Lucent | Method and device for improving the building of compositions of services |
| CN107066534A (en) * | 2017-03-02 | 2017-08-18 | 人谷科技（北京）有限责任公司 | Multi-source data polymerization and system |
| CN107451208A (en) * | 2017-07-12 | 2017-12-08 | 北京潘达互娱科技有限公司 | A kind of data search method and device |
| CN107491510A (en) * | 2017-08-03 | 2017-12-19 | 国网江苏省电力公司信息通信分公司 | One kind mixing heterogeneous data source unified query system and distributed enquiring method |
-
2019
- 2019-03-28 CN CN201910242794.5A patent/CN110096553B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957968A (en) * | 2010-08-31 | 2011-01-26 | 南京财经大学 | Online transaction service aggregation method based on Hadoop |
| EP2722753A1 (en) * | 2012-10-19 | 2014-04-23 | Alcatel Lucent | Method and device for improving the building of compositions of services |
| CN103136364A (en) * | 2013-03-14 | 2013-06-05 | 曙光信息产业(北京)有限公司 | Cluster database system and data query processing method thereof |
| CN107066534A (en) * | 2017-03-02 | 2017-08-18 | 人谷科技（北京）有限责任公司 | Multi-source data polymerization and system |
| CN107451208A (en) * | 2017-07-12 | 2017-12-08 | 北京潘达互娱科技有限公司 | A kind of data search method and device |
| CN107491510A (en) * | 2017-08-03 | 2017-12-19 | 国网江苏省电力公司信息通信分公司 | One kind mixing heterogeneous data source unified query system and distributed enquiring method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110765121A (en) * | 2019-10-31 | 2020-02-07 | 太原蓝知科技有限公司 | Intelligent big data acquisition and processing method and system |
| CN110765121B (en) * | 2019-10-31 | 2022-09-20 | 上海意臣信息科技有限公司 | Intelligent big data acquisition and processing method and system |
| CN112199424A (en) * | 2020-09-07 | 2021-01-08 | 深圳市法本信息技术股份有限公司 | Method for pulling communication information island through RPA, readable storage medium and RPA robot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110096553B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8005815B2 (en) | Search engine | |
| CN111753060A (en) | Information retrieval method, device, equipment and computer readable storage medium | |
| CN108460136A (en) | Electric power O＆M information knowledge map construction method | |
| CN109033284A (en) | The power information operational system database construction method of knowledge based map | |
| CN111339421A (en) | Information search method, device, equipment and storage medium based on cloud technology | |
| CN110059160A (en) | A kind of knowledge base answering method and device based on context end to end | |
| TW200809549A (en) | System and method for searching and matching data having ideogrammatic content | |
| TWI743623B (en) | Artificial intelligence-based business intelligence system and its analysis method | |
| CN117290489B (en) | Method and system for quickly constructing industry question-answer knowledge base | |
| CN116991869A (en) | Method for automatically generating database query statement based on NLP language model | |
| JP2019040598A5 (en) | ||
| CN110096553A (en) | A kind of the big data analysis system and analysis method of integration across database | |
| CN113221562A (en) | Method and system for improving document file retrieval efficiency based on knowledge graph | |
| CN110413882B (en) | Information pushing method, device and equipment | |
| CN111563097A (en) | Unsupervised topic aggregation method and device, electronic equipment and storage medium | |
| CN117076636A (en) | Information query method, system and equipment for intelligent customer service | |
| CN111460114A (en) | Retrieval method, device, equipment and computer readable storage medium | |
| CN116610810A (en) | Intelligent searching method and system based on regulation and control of cloud knowledge graph blood relationship | |
| CN115982322A (en) | Water conservancy industry design field knowledge graph retrieval method and retrieval system | |
| CN115269862A (en) | Electric power question-answering and visualization system based on knowledge graph | |
| CN115658856A (en) | Intelligent question-answering system and method based on polymorphic document views | |
| CN111753861B (en) | Automatic image labeling system and method for active learning | |
| JP2006072787A (en) | Automatic question answering method, model production method therefor and computer program | |
| CN117390497B (en) | Category prediction method, device and equipment based on large language model | |
| CN114385823A (en) | System evaluation method based on knowledge graph and network analysis 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 |