CN109947782A - A kind of update method of big data real-time application system, apparatus and system - Google Patents
A kind of update method of big data real-time application system, apparatus and system Download PDFInfo
- Publication number
- CN109947782A CN109947782A CN201711069874.2A CN201711069874A CN109947782A CN 109947782 A CN109947782 A CN 109947782A CN 201711069874 A CN201711069874 A CN 201711069874A CN 109947782 A CN109947782 A CN 109947782A
- Authority
- CN
- China
- Prior art keywords
- data
- real
- updated
- application system
- preset model
- 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
Abstract
The present invention provides a kind of update methods of big data real-time application system, apparatus and system, wherein the update method of big data real-time application system, comprising: obtains prediction data using the preset model in big data real-time application system；Obtain real data corresponding with the prediction data；According to the prediction data and the real data, the preset model is updated.This programme is by obtaining prediction data using the preset model in big data real-time application system；Obtain real data corresponding with the prediction data；According to the prediction data and the real data, the preset model is updated；Automatically big data real-time application system can be adjusted, to adapt to production procedure, creation data and the variation of production law；Solve the problems, such as that update scheme low efficiency, the accuracy of big data application system in the prior art are poor.
Description
Technical field
The present invention relates to big data technical fields, particularly relate to update method, the dress of a kind of big data real-time application system
It sets and system.
Background technique
Existing big data application system, can be divided into three parts function: data synchronous (this step is omitted in some systems),
Data Integration, data analysis.Wherein Data Integration is that the data that will be generated in production process are integrated, and generation can be used for big number
According to the data set of analysis；Data analysis is to be analyzed using model data set, and model used is by machine learning
Method analyzes historical data.
The data generated in production process are synchronized and integrated in real time, need first to conclude production procedure
And summary.Existing big data application system is that, by application system and production procedure close coupling, had based on fixed production procedure
Process has been made into configurable by a little application systems, can manually modify to production procedure data.If production procedure is sent out
Changing is needed first to analyze new production procedure, then be modified accordingly system；If procedure information is can
Configuration, then it needs to modify to the configuration data of process.These work are all manually to be handled at present.
It is, the data generated in process of production may change, for example occur occurring in certain characteristic items new
Characteristic value or the values of certain characteristic items become empty.The variation of these characteristic values may make the data set and machine generated
Learning model mismatches, and is not used to big data analysis.Existing technology can be filtered these data, or to system into
Row modification or re -training model, but latter two operation is all manually to be handled, not only low efficiency, accuracy are poor, also
It will cause data waste.
Machine learning model used in big data application system is obtained by the analysis to historical data, is using machine
The method of device study excavates the production law contained in historical data.But these rules may occur in process of production
At this moment variation reuses that original model is analyzed the result is that inaccurate.When existing technology is by one section
Between after historical data is integrated again, generate new model, by way of machine learning then for replacing application system
In old model, the work of this part is also manually performed.But manually big data application system is intervened, time-consuming consumption
Power cannot grasp the real-time change of environment, be also easy to cause to fail to judge, judge by accident, influence the accuracy of system.
Summary of the invention
The purpose of the present invention is to provide a kind of update methods of big data real-time application system, apparatus and system, solve
The update scheme low efficiency of big data application system, the problem of accuracy difference in the prior art.
In order to solve the above-mentioned technical problem, the embodiment of the present invention provides a kind of update side of big data real-time application system
Method, comprising:
Prediction data is obtained using the preset model in big data real-time application system；
Obtain real data corresponding with the prediction data；
According to the prediction data and the real data, the preset model is updated.
Optionally, the step of preset model using in big data real-time application system obtains prediction data include:
Synchrodata is obtained according to real-time production data；
Integral data is obtained according to the synchrodata；
The integral data is analyzed using the preset model, obtains prediction data.
Optionally, described the step of obtaining synchrodata according to real-time production data, includes:
Extract the characteristic item information of the real-time production data；
According to the characteristic item information, the real-time production data is stored as synchrodata.
Optionally, described the step of obtaining integral data according to the synchrodata, includes:
It, will synchrodata corresponding with the production procedure after real-time production data corresponding production procedure
It is integrated, obtains the integral data.
Optionally, described to integrate synchrodata corresponding with the production procedure, obtain the integral data
Step includes:
Will synchrodata corresponding with the production procedure be successively filtered, clean, table structure conversion, duplicate removal, obtain
Preliminary data that treated；
The preliminary data is associated, the integral data is obtained.
Optionally, described according to the prediction data and the real data, the step that the preset model is updated
Suddenly include:
According to the prediction data and the real data, historical data is updated；
Using updated historical data, the preset model is updated.
Optionally, described the step of utilizing updated historical data, being updated to the preset model, includes:
The preset model is updated using updated historical data according to timing information；Or
When meeting default update condition, using updated historical data, the preset model is updated.
Optionally, described when meeting default update condition, using updated historical data, to the preset model into
Row update the step of include:
It is corresponding lower than original characteristic item in first threshold, the historical data in the predictablity rate of the prediction data
The quantity of blank information is more than when there is newly-increased feature item in second threshold and/or the historical data, to be gone through using updated
History data are updated the preset model.
The embodiment of the invention also provides a kind of updating devices of big data real-time application system, comprising:
First obtains module, for obtaining prediction data using the preset model in big data real-time application system；
Second obtains module, for obtaining real data corresponding with the prediction data；
First update module, for being carried out more to the preset model according to the prediction data and the real data
Newly.
Optionally, the first acquisition module includes:
First processing submodule, for obtaining synchrodata according to real-time production data；
Second processing submodule, for obtaining integral data according to the synchrodata；
Third handles submodule, for analyzing using the preset model the integral data, obtains prediction number
According to.
Optionally, the first processing submodule includes:
First extraction unit, for extracting the characteristic item information of the real-time production data；
First storage unit, for according to the characteristic item information, the real-time production data to be stored as synchrodata.
Optionally, the second processing submodule includes:
First processing units are used for after the real-time production data corresponding production procedure, will be with the production
The corresponding synchrodata of process is integrated, and the integral data is obtained.
Optionally, the first processing units include:
First processing subelement, for synchrodata corresponding with the production procedure to be successively filtered, cleans, table
Structure conversion, duplicate removal, the preliminary data that obtains that treated；
Second processing subelement obtains the integral data for the preliminary data to be associated.
Optionally, first update module includes:
First updates submodule, for being updated to historical data according to the prediction data and the real data；
Second updates submodule, for utilizing updated historical data, is updated to the preset model.
Optionally, the second update submodule includes:
First updating unit, for being carried out using updated historical data to the preset model according to timing information
It updates；Or
Second updating unit, for using updated historical data, being preset to described when meeting default update condition
Model is updated.
Optionally, second updating unit includes:
First updates subelement, is lower than first threshold, the history number for the predictablity rate in the prediction data
The quantity of the corresponding blank information of original characteristic item is more than newly-increased feature occur in second threshold and/or the historical data in
Xiang Shi is updated the preset model using updated historical data.
The embodiment of the invention also provides a kind of big data real-time application system, including memory, processor and it is stored in
On the memory and the computer program that can run on the processor；The processor is realized when executing described program
The update method for the big data real-time application system stated.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, the journey
The step in the update method of above-mentioned big data real-time application system is realized when sequence is executed by processor.
The advantageous effects of the above technical solutions of the present invention are as follows:
In above scheme, the update method of the big data real-time application system is by utilizing big data real-time application system
In preset model obtain prediction data；Obtain real data corresponding with the prediction data；According to the prediction data and
The real data is updated the preset model；Automatically big data real-time application system can be adjusted, with suitable
Answer the variation of production procedure, creation data and production law；Solve the update scheme effect of big data application system in the prior art
The problem that rate is low, accuracy is poor.
Detailed description of the invention
Fig. 1 is the update method flow diagram of the big data real-time application system of the embodiment of the present invention；
Fig. 2 is the update method concrete application block schematic illustration of the big data real-time application system of the embodiment of the present invention；
Fig. 3 is the update method concrete application example schematic diagram of the big data real-time application system of the embodiment of the present invention；
Fig. 4 is the updating device structural schematic diagram of the big data real-time application system of the embodiment of the present invention.
Specific embodiment
To keep the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.
The present invention in view of the prior art in big data application system update scheme low efficiency, the problem of accuracy difference,
A kind of update method of big data real-time application system is provided, as shown in Figure 1, comprising:
Step 11: obtaining prediction data using the preset model in big data real-time application system；
Step 12: obtaining real data corresponding with the prediction data；
Step 13: according to the prediction data and the real data, the preset model being updated.
The update method of the big data real-time application system provided in an embodiment of the present invention passes through real-time using big data
Preset model in application system obtains prediction data；Obtain real data corresponding with the prediction data；According to described pre-
Measured data and the real data, are updated the preset model；Automatically big data real-time application system can be carried out
Adjustment, to adapt to production procedure, creation data and the variation of production law；Solve in the prior art that big data application system is more
The problem of new departure low efficiency, accuracy difference.
Wherein, the step of preset model using in big data real-time application system obtains prediction data includes: root
Synchrodata is obtained according to real-time production data；Integral data is obtained according to the synchrodata；Using the preset model to institute
It states integral data to be analyzed, obtains prediction data.
Specifically, described the step of obtaining synchrodata according to real-time production data includes: to extract the real-time production number
According to characteristic item information；According to the characteristic item information, the real-time production data is stored as synchrodata.
Described the step of obtaining integral data according to the synchrodata includes: life corresponding in the real-time production data
After producing process, synchrodata corresponding with the production procedure is integrated, the integral data is obtained.
More specifically, described to integrate synchrodata corresponding with the production procedure, obtain the integral data
The step of include: synchrodata corresponding with the production procedure is successively filtered, is cleaned, table structure conversion, duplicate removal, obtain
To treated preliminary data；The preliminary data is associated, the integral data is obtained.
It is described according to the prediction data and institute in the present embodiment in order to improve the updated precision of prediction of preset model
The step of stating real data, the preset model be updated include: according to the prediction data and the real data, it is right
Historical data is updated；Using updated historical data, the preset model is updated.
In view of opportunity for being updated about preset model can there are many, in the present embodiment, it is preferred that described using updating
Historical data afterwards, the step of being updated to the preset model include: to utilize updated history number according to timing information
According to being updated to the preset model；Or when meeting default update condition, using updated historical data, to institute
Preset model is stated to be updated.
Specifically, described when meeting default update condition, using updated historical data, to the preset model into
The step of row updates includes: the predictablity rate in the prediction data lower than original spy in first threshold, the historical data
The quantity of the corresponding blank information of sign item is more than when there is newly-increased feature item in second threshold and/or the historical data, to utilize
Updated historical data is updated the preset model.
From the foregoing, it will be observed that above scheme provided in an embodiment of the present invention can be adaptive update big data real-time application system
In preset model, improve precision of prediction.
The update method of the big data real-time application system provided in an embodiment of the present invention is carried out furtherly below
It is bright.
In view of the above technical problems, the embodiment of the invention provides a kind of update method of big data real-time application system,
Automatically big data real-time application system can be adjusted, to adapt to production procedure, creation data and the variation of production law；
The processing frame of the update method can be as shown in Fig. 2, corresponding process flow be as follows:
(1) to newly-generated creation data carry out real-time synchronization (such as: user register and diagnostic message, there are hospitals
HIS in, need to be synchronized to these data in big data real-time application system.Big data real-time application system and user information
Generation system be separation), and save as synchrodata, at the same by the newly-generated affiliated flow nodes information of creation data and
Characteristic information (can be used for the value of the characteristic item of big data analysis, such as register affiliated department, diagnostic message, medical history taking, payment
Information etc.) it extracts and saves as flow data；Wherein, HIS be carried out in hospital management and curative activity information management and
The computer application system of on-line operation；
(2) using the preset model generated based on machine learning method, model analysis is carried out to newly-generated flow data
(judging whether current operation process terminates)；
(3) it at the end of the result of process analysis is process, extracts relevant synchrodata and (has recorded use in procedure information
Flow nodes information belonging to the every single stepping in family and database index information, can be the whole of user according to these information
Data caused by a operating process extract), and progress Data Integration (such as: synchrodata is filtered, is cleaned, table
The new data of generation, is then associated by structure conversion, duplicate removal etc. again, generates a big data set), result is saved
For data set；
(4) above-mentioned preset model is used, model analysis is carried out (such as to some aspect of user to newly-generated data set
It is analyzed and predicted, such as probability of illness, further consultation probability etc., is all to be carried out using corresponding preset model to data set
Analysis, finally obtains prediction result), analysis result is saved as into result set；
(5) timing or meet trigger condition (such as: predictablity rate occurs new lower than threshold value, in data set features item
There is a large amount of null value etc. in characteristic value, data set features item) when, extract the history procedure information and/or number in flow data
According to the history data set of concentration, model training is carried out using the method for machine learning, corresponding new preset model is generated, to original
Some preset models are updated.
The update method of big data real-time application system provided in an embodiment of the present invention is applied to hospital below to lift
Example explanation, as shown in Figure 3.
Real time data deposit HIS DB (database) is constituted into new data；
By new data after data synchronization engine, data filtering engine, data cleansing engine and data transformation engine shape
At pretreated new data, it is stored in rear DB (being referred to as big data application system DB)；
By historical data after data pick-up engine, data filtering engine, data cleansing engine and data transformation engine,
It denoises again, duplicate removal, forms pretreated historical data, be stored in CRMI DB；
New data comparison data with existing after data synchronization engine obtains patient's states (y value), is stored in CRMI
DB；
Extract pretreated new data and pretreated historical data, carry out Java modeling, including creation model and
More new model；
Using the model and pretreated new data of creation, carries out Java and calculate probability, obtain prediction result, be stored in
CRMI DB, and showed.
From the foregoing, it will be observed that scheme provided in an embodiment of the present invention can be assisted greatly by the big data analysis to production procedure
Data real-time application system carries out Data Integration, obtains the data set that can be used for big data applied analysis；And it can be automatic right
Big data real-time application system is adjusted, to adapt to the variation of production procedure and data content:
During the real-time synchronization of creation data, extracts production procedure nodal information belonging to data and data characteristics is raw
It is analyzed it at flow data collection, and using the method for machine learning, to obtain complete production procedure belonging to this data
Data, improve work efficiency and accuracy rate；
According to process analysis result guide data integration work, in real time, it is complete, accurately generate and can be used for big data application
The data set of analysis；
During the real-time synchronization of creation data, according to service logic and causality, relevant go through is concentrated to data
History data are updated, this partial data can be used for training new model；
The mould for automatically extracting the historical data of certain period of time in data set, and calling the method training of machine learning new
Type, for being updated to original model, to adapt to the variation of production procedure and creation data.
The embodiment of the invention also provides a kind of updating devices of big data real-time application system, as shown in Figure 4, comprising:
First obtains module 41, for obtaining prediction data using the preset model in big data real-time application system；
Second obtains module 42, for obtaining real data corresponding with the prediction data；
First update module 43, for being carried out to the preset model according to the prediction data and the real data
It updates.
The updating device of the big data real-time application system provided in an embodiment of the present invention passes through real-time using big data
Preset model in application system obtains prediction data；Obtain real data corresponding with the prediction data；According to described pre-
Measured data and the real data, are updated the preset model；Automatically big data real-time application system can be carried out
Adjustment, to adapt to production procedure, creation data and the variation of production law；Solve in the prior art that big data application system is more
The problem of new departure low efficiency, accuracy difference.
Wherein, the first acquisition module includes: the first processing submodule, for being synchronized according to real-time production data
Data；Second processing submodule, for obtaining integral data according to the synchrodata；Third handles submodule, for utilizing
The preset model analyzes the integral data, obtains prediction data.
Specifically, the first processing submodule includes: the first extraction unit, for extracting the real-time production data
Characteristic item information；First storage unit, for according to the characteristic item information, the real-time production data to be stored as same step number
According to.
The second processing submodule includes: first processing units, in the corresponding production of the real-time production data
After process, synchrodata corresponding with the production procedure is integrated, the integral data is obtained.
More specifically, the first processing units include: the first processing subelement, and being used for will be corresponding with the production procedure
Synchrodata be successively filtered, clean, table structure conversion, duplicate removal, the preliminary data that obtains that treated；Second processing is single
Member obtains the integral data for the preliminary data to be associated.
In order to improve the updated precision of prediction of preset model, in the present embodiment, first update module includes: first
Submodule is updated, for being updated to historical data according to the prediction data and the real data；Second updates submodule
Block is updated the preset model for utilizing updated historical data.
In view of opportunity for being updated about preset model can there are many, in the present embodiment, it is preferred that described second updates
Submodule includes: the first updating unit, is used for according to timing information, using updated historical data, to the preset model
It is updated；Or second updating unit, for when meeting default update condition, using updated historical data, to institute
Preset model is stated to be updated.
Specifically, second updating unit includes: the first update subelement, it is quasi- for the prediction in the prediction data
The quantity of true rate blank information corresponding lower than characteristic item original in first threshold, the historical data be more than second threshold and/
Or when occurring newly-increased feature item in the historical data, using updated historical data, the preset model is updated.
Wherein, the realization embodiment of the update method of above-mentioned big data real-time application system is suitable for the big data
In the embodiment of the updating device of real-time application system, it can also reach identical technical effect.
From the foregoing, it will be observed that above scheme provided in an embodiment of the present invention can be adaptive update big data real-time application system
In preset model, improve precision of prediction.
The embodiment of the invention also provides a kind of big data real-time application system, including memory, processor and it is stored in
On the memory and the computer program that can run on the processor；The processor is realized when executing described program
The update method for the big data real-time application system stated.
Wherein, the realization embodiment of the update method of above-mentioned big data real-time application system is suitable for the big data
In the embodiment of real-time application system, it can also reach identical technical effect.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, the journey
The step in the update method of above-mentioned big data real-time application system is realized when sequence is executed by processor.
Wherein, the realization embodiment of the update method of above-mentioned big data real-time application system is suitable for the computer
In the embodiment of readable storage medium storing program for executing, it can also reach identical technical effect.
It should be noted that this many functional component described in this description all be referred to as module/submodule/unit/
Subelement, specifically to emphasize the independence of its implementation.
In the embodiment of the present invention, module/submodule/unit/subelement can use software realization, so as to by various types of
Processor executes.For example, the executable code module of a mark may include one or more objects of computer instruction
Reason or logical block, for example, it can be built as object, process or function.Nevertheless, institute's mark module is held
Line code needs not be physically located together, but may include the different instructions being stored in different positions, when these instructions
When being combined together in logic, constitutes module and realize the regulation purpose of the module.
In fact, executable code module can be the either many item instructions of individual instructions, and can even be distributed
It on multiple and different code segments, is distributed in distinct program, and is distributed across multiple memory devices.Similarly, it grasps
Making data can be identified in module, and can realize according to any form appropriate and be organized in any appropriate class
In the data structure of type.The operation data can be used as individual data collection and be collected, or can be distributed on different location
(including in different storage device), and at least partly can only be present in system or network as electronic signal.
When module can use software realization, it is contemplated that the level of existing hardware technique, it is possible to implemented in software
Module, without considering the cost, those skilled in the art can build corresponding hardware circuit to realize correspondence
Function, the hardware circuit includes conventional ultra-large integrated (VLSI) circuit or gate array and such as logic core
The existing semiconductor of piece, transistor etc either other discrete elements.Module can also use programmable hardware device, such as
Field programmable gate array, programmable logic array, programmable logic device etc. are realized.
Above-described is the preferred embodiment of the present invention, it should be pointed out that the ordinary person of the art is come
It says, under the premise of not departing from principle of the present invention, can also make several improvements and retouch, these improvements and modifications should also regard
For protection scope of the present invention.
Claims (11)
1. a kind of update method of big data real-time application system characterized by comprising
Prediction data is obtained using the preset model in big data real-time application system；
Obtain real data corresponding with the prediction data；
According to the prediction data and the real data, the preset model is updated.
2. update method according to claim 1, which is characterized in that described using pre- in big data real-time application system
If model obtains the step of prediction data and includes:
Synchrodata is obtained according to real-time production data；
Integral data is obtained according to the synchrodata；
The integral data is analyzed using the preset model, obtains prediction data.
3. update method according to claim 2, which is characterized in that described to obtain synchrodata according to real-time production data
The step of include:
Extract the characteristic item information of the real-time production data；
According to the characteristic item information, the real-time production data is stored as synchrodata.
4. update method according to claim 2, which is characterized in that described to obtain integral data according to the synchrodata
The step of include:
After the real-time production data corresponding production procedure, synchrodata corresponding with the production procedure is carried out
Integration, obtains the integral data.
5. update method according to claim 4, which is characterized in that it is described will same step number corresponding with the production procedure
Include: according to the step of being integrated, obtaining the integral data
Will synchrodata corresponding with the production procedure be successively filtered, clean, table structure conversion, duplicate removal, handled
Preliminary data afterwards；
The preliminary data is associated, the integral data is obtained.
6. update method according to claim 1, which is characterized in that described according to the prediction data and the actual number
Include: according to, the step of being updated to the preset model
According to the prediction data and the real data, historical data is updated；
Using updated historical data, the preset model is updated.
7. update method according to claim 6, which is characterized in that it is described to utilize updated historical data, to described
The step of preset model is updated include:
The preset model is updated using updated historical data according to timing information；Or
When meeting default update condition, using updated historical data, the preset model is updated.
8. update method according to claim 7, which is characterized in that it is described when meeting default update condition, using more
Historical data after new, the step of being updated to the preset model include:
In the predictablity rate of prediction data blank corresponding lower than characteristic item original in first threshold, the historical data
The quantity of information is more than when there is newly-increased feature item in second threshold and/or the historical data, to utilize updated history number
According to being updated to the preset model.
9. a kind of updating device of big data real-time application system characterized by comprising
First obtains module, for obtaining prediction data using the preset model in big data real-time application system；
Second obtains module, for obtaining real data corresponding with the prediction data；
First update module, for being updated to the preset model according to the prediction data and the real data.
10. a kind of big data real-time application system, including memory, processor and it is stored on the memory and can be described
The computer program run on processor；It is characterized in that, the processor realized when executing described program as claim 1 to
The update method of big data real-time application system described in any one of 8.
11. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The step in the update method such as big data real-time application system described in any item of the claim 1 to 8 is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711069874.2A CN109947782A (en) | 2017-11-03 | 2017-11-03 | A kind of update method of big data real-time application system, apparatus and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711069874.2A CN109947782A (en) | 2017-11-03 | 2017-11-03 | A kind of update method of big data real-time application system, apparatus and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109947782A true CN109947782A (en) | 2019-06-28 |
Family
ID=67003958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711069874.2A Pending CN109947782A (en) | 2017-11-03 | 2017-11-03 | A kind of update method of big data real-time application system, apparatus and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947782A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598338A (en) * | 2020-05-18 | 2020-08-28 | 贝壳技术有限公司 | Method, apparatus, medium, and electronic device for updating prediction model |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130346044A1 (en) * | 2012-06-25 | 2013-12-26 | International Business Machines Corporation | Methods and apparatus for data collection |
| CN104573000A (en) * | 2015-01-07 | 2015-04-29 | 北京云知声信息技术有限公司 | Sequential learning based automatic questions and answers device and method |
| CN104866727A (en) * | 2015-06-02 | 2015-08-26 | 陈宽 | Deep learning-based method for analyzing medical data and intelligent analyzer thereof |
| CN106600356A (en) * | 2016-10-27 | 2017-04-26 | 杭州王道科技有限公司 | Multi-platform electronic commerce information aggregation method and system |
-
2017
- 2017-11-03 CN CN201711069874.2A patent/CN109947782A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130346044A1 (en) * | 2012-06-25 | 2013-12-26 | International Business Machines Corporation | Methods and apparatus for data collection |
| CN104573000A (en) * | 2015-01-07 | 2015-04-29 | 北京云知声信息技术有限公司 | Sequential learning based automatic questions and answers device and method |
| CN104866727A (en) * | 2015-06-02 | 2015-08-26 | 陈宽 | Deep learning-based method for analyzing medical data and intelligent analyzer thereof |
| CN106600356A (en) * | 2016-10-27 | 2017-04-26 | 杭州王道科技有限公司 | Multi-platform electronic commerce information aggregation method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598338A (en) * | 2020-05-18 | 2020-08-28 | 贝壳技术有限公司 | Method, apparatus, medium, and electronic device for updating prediction model |
| CN111598338B (en) * | 2020-05-18 | 2021-08-31 | 贝壳找房(北京)科技有限公司 | Method, apparatus, medium, and electronic device for updating prediction model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106295807B (en) | A kind of method and device of information processing | |
| CN107888397A (en) | The method and apparatus for determining fault type | |
| CN109686441A (en) | A kind of big data medical data feature extraction and intellectual analysis prediction technique | |
| CN107463661A (en) | The introduction method and device of data | |
| CN106503043B (en) | A kind of interaction data processing method for intelligent robot | |
| CN106951499A (en) | A kind of knowledge mapping method for expressing based on translation model | |
| CN108108426A (en) | Understanding method, device and the electronic equipment that natural language is putd question to | |
| CN106021422B (en) | A kind of method and system forming Hive data warehouse based on relevant database | |
| CN109033356A (en) | The method and customer service system to label for customer service system log | |
| CN108846695A (en) | The prediction technique and device of terminal replacement cycle | |
| CN110110095A (en) | A kind of power command text matching technique based on shot and long term memory Recognition with Recurrent Neural Network | |
| CN110070929A (en) | A kind of acquisition and cleaning method for atrial fibrillation Single diseases data | |
| CN106407096A (en) | Time series based log analysis method and log analysis apparatus | |
| CN101482897A (en) | Intelligent skirt style and pattern design system | |
| CN106264545A (en) | Step recognition method and device | |
| CN109947782A (en) | A kind of update method of big data real-time application system, apparatus and system | |
| CN109324264B (en) | Identification method and device for abnormal value of power distribution network line impedance data | |
| CN105117980B (en) | A kind of automatic evaluation method of grid equipment state | |
| CN107845042A (en) | Supervisory systems is imposed in house in a kind of state-owned land | |
| CN115062674B (en) | Tool arrangement and tool changing method and device based on deep learning and storage medium | |
| CN107908697A (en) | The automatic acquiring method and device of host batch processing job result | |
| CN115579150A (en) | Method and device for creating disease-specific database, electronic equipment and computer medium | |
| CN110147403A (en) | Agriculture big data fusion method, device, equipment and storage medium | |
| CN114840507A (en) | Data governance method and device, electronic equipment and storage medium | |
| CN113888231A (en) | Marketing data screening analysis system based on artificial intelligence |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190628 |
|
| RJ01 | Rejection of invention patent application after publication |