CN107025509A - Decision system and method based on business model - Google Patents
Decision system and method based on business model Download PDFInfo
- Publication number
- CN107025509A CN107025509A CN201610070829.8A CN201610070829A CN107025509A CN 107025509 A CN107025509 A CN 107025509A CN 201610070829 A CN201610070829 A CN 201610070829A CN 107025509 A CN107025509 A CN 107025509A
- Authority
- CN
- China
- Prior art keywords
- sample
- model
- business
- user data
- modeling
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of operational decision making system and method based on business model, belong to computer and Internet technical field.The system includes：Customer information system, modeling and decision system；Customer information system, which is provided, to be used to model the various user data with decision-making；Modeling obtains the second user data of sample of users from customer information system on the basis of the first user data of the sample of users related to business objective provided by business side；Modeling sample is generated according to the above-mentioned data of sample of users；Obtain the model formulated by business root according to business objective；Model is trained using modeling sample, and the model for completing training is deployed to decision system；Decision system is used to carry out operational decision making using model.By said system customized development model, help to save development cost, improve development efficiency and improve model accuracy.
Description
Technical field
The present invention relates to computer and Internet technical field, more particularly to a kind of decision-making based on business model
System and method.
Background technology
Business side is generally by developing business model to help to carry out operational decision making.Business model refers to be used to help
Business side is helped to carry out the mathematical modeling of operational decision making, hereinafter referred to as " model ".
In the prior art, business root factually border business demand, determines business objective.Afterwards, business side
Model for realizing above-mentioned business objective is developed according to existing business datum, then entered using the model of exploitation
Row operational decision making.
However, at least there are the following problems for prior art：On the one hand, business side is limited to itself development ability
Limited, the development cost of model is higher, and development efficiency is relatively low；On the other hand, business side is limited to itself institute
The modeling data that can be obtained is limited, and the accuracy for the model developed is poor.
The content of the invention
In order to solve in the prior art, development cost height, exploitation effect present in business side itself development model
The problem of rate is low and model accuracy is poor, the embodiments of the invention provide a kind of decision-making based on business model
System and method.The technical scheme is as follows：
First aspect includes there is provided a kind of operational decision making system based on business model, the system：With
Family information system, modeling and decision system；
The modeling, the of the sample of users related to business objective provided for obtaining by business side
One user data；The second user data of the sample of users are obtained from the customer information system；According to institute
State the first user data and second user data generation modeling sample of sample of users；Obtain by the business side
The model formulated according to the business objective；The model is trained using the modeling sample, and will
The model for completing training is deployed to the decision system；
The decision system, for carrying out operational decision making using the model for completing training, the business is determined
Plan refers to be predicted the relevant information of targeted customer for the business side.
Second aspect includes there is provided a kind of operational decision making method based on business model, methods described：
First user data of the sample of users related to business objective provided by business side is provided；
The second user data of the sample of users are obtained from customer information system；
According to the first user data of the sample of users and second user data generation modeling sample；
Obtain the model formulated by the business root according to the business objective；
The model is trained using the modeling sample；
Operational decision making is carried out using the model for completing training, the operational decision making refers to for the business side
Relevant information to targeted customer is predicted.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought includes：
By modeling the sample of users related to business objective provided by business side the first number of users
On the basis of, the second user data of sample of users are obtained from customer information system, and then according to above-mentioned number
According to generation modeling sample, the model formulated by business root according to business objective is also obtained by modeling, and
Model is trained using above-mentioned modeling sample, it is real that the model for completing training then is deployed into decision system
Apply, to carry out operational decision making；Solve in the prior art, the exploitation present in business side itself development model
The problem of cost height, low development efficiency and poor model accuracy；Business side is provided by the embodiment of the present invention
System customization development model, contribute to save development cost, improve development efficiency and improve model it is accurate
Property.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, institute in being described below to embodiment
The accompanying drawing needed to use is briefly described, it should be apparent that, drawings in the following description are only the present invention
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work,
Other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is the block diagram for the operational decision making system based on business model that one embodiment of the invention is provided；
Fig. 2 is the schematic diagram for the server architecture that one embodiment of the invention is provided；
Fig. 3 is the structural representation for the server that one embodiment of the invention is provided；
Fig. 4 is the software architecture for the operational decision making system based on business model that one embodiment of the invention is provided
Figure；
Fig. 5 is the schematic diagram for the model development flow that one embodiment of the invention is provided；
Fig. 6 is the flow chart for the operational decision making method based on business model that one embodiment of the invention is provided.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to the present invention
Embodiment is described in further detail.
Technical scheme provided in an embodiment of the present invention, by being determined to each business side offer model development and business
Plan platform, the platform is provided with the user data of magnanimity, big data calculating platform, visual model editing
Manage deployment tool and efficient decision system, business root is according to business objective, in the user data of magnanimity
On the basis of carry out the design of model, exploitation and verify, for the specific crowd related to business objective and specific
Business, targetedly formulates model, and carries out operational decision making using the model formulated, with meet high accuracy,
The customization model development demand of high efficiency and low cost.
" model " being referred to herein is business model, and business model refers to that being used to help business side carries out business
The mathematical modeling of decision-making, operational decision making refers to be predicted the relevant information of targeted customer for business side.
Below, the technical scheme provided by several embodiments the present invention is introduced and illustrated.
Fig. 1 is refer to, it illustrates the operational decision making system based on business model that one embodiment of the invention is provided
The block diagram of system 10, the operational decision making system 10 based on business model of being somebody's turn to do includes：Customer information system 11, build
Modular system 12 and decision system 13.
Modeling 12, the first of the sample of users related to business objective provided for obtaining by business side
User data；The second user data of sample of users are obtained from customer information system 11；According to sample of users
First user data and second user data generation modeling sample；Acquisition is formulated by business root according to business objective
Model；Model is trained using modeling sample, and the model for completing training is deployed to decision system
13。
Business side refers to the mechanism for providing a user business.In embodiments of the present invention, to business side
The type for the business that type and business side are provided is not construed as limiting.Business objective is business root factually border industry
Business demand, the function to be realized of identified model.
Sample of users is business root according to existing business datum, the selected use for being used to build modeling sample
Family.First user data (alternatively referred to as " user shows data ") includes the number corresponding to the output variable of model
According to.Business side is formulated after business objective, and the applicable people's group object of model, and root are determined according to business objective
The people's group object being applicable according to model and existing business datum choose sample of users, and mix the sample with family first is used
User data is supplied to the system.The the first user data storage for the sample of users that the system can provide business side
Into customer information system 11.
Customer information system 11 is used for each item data for storing user, the data access function with mass users.
In customer information system 11, using user as kernel object, it is stored with for modeling the second user with decision-making
Data (alternatively referred to as " user attribute data ").Second user data are included corresponding to the input variable of model
Data.Second user data include but is not limited to：User basic information, network application data, user's finance
Data etc. are used for each item data for reflecting user characteristics.It should be noted is that, second user data include
The user data obtained can be inquired about from common platform, and second user data are before user authorizes and obtained
Put what is just obtained from common platform or other approach.By the identity of user in customer information system 11
Storage corresponding with each item data of user, builds and enriches comprehensive whole user view.Customer information system 11
It can provide with bodies such as the identification card number of user, phone number, social networking application account number, instant messaging application account numbers
Part is designated the information retrieval function of index.The data source of customer information system 11 includes but is not limited to：Should
In the internal data of system provider, such as people's row reference, public security household register, student status, social security office, common reserve fund
The data that the outside public organizations such as heart provide, the subscriber data crawled from network, such as payment data, consumption number
According to, finance data, social data etc..
Modeling 12 is obtained after the first user data of the sample of users that business side is provided, and is used according to sample
The identity at family, matching obtains the second user data of sample of users from customer information system 11.
Modeling 12 generates modeling sample according to the first user data and second user data of sample of users.
Specifically, modeling 12 specifically for：To the first user data and second user data of sample of users
Handled, set up basic data collection, basic data, which is concentrated, includes the sample variable institute related to business objective
Corresponding data；Pair sample variable related to business objective is handled, and chooses target sample variable, mesh
Mark sample variable refers to the sample variable with strong predictive ability；Data according to corresponding to target sample variable,
Generate modeling sample.Modeling 12 is according to business objective, data character, structure and internal logic, logarithm
According to being extracted, being cleared up and being arranged, basic data collection is set up.Basic data collection can be one and include and industry
The wide table that data corresponding to the related sample variable of business target are spliced to form.For example, basic data collection can
To be an identity using user as index, age, household register, student status, working condition comprising user,
The wide table that some items such as Income situation sample variable related to business objective is spliced to form.Modeling 12
The analysis of data information is carried out on the basis of basic data collection, its internal association is found out, and by sample
The packet, merging, conversion and segmentation of this variable, reject the weaker sample variable of predictive ability, and selection meets
Modeling conditions and the sample variable (i.e. target sample variable) with stronger predictive ability.For example, modeling
The data mart modeling modes such as 12 can be merged using variable partitions, data, variable derivative, and using ETL (Extract
Transform Load, data warehouse technology) cleaning conversion is carried out to data, finally give target sample variable.
Modeling 12 can provide the variable deriving method and outstanding variable segmentation tool of specialty, help to be rapidly completed
The ETL processes of data.Modeling sample at least includes the sample for being used to be trained model, optional including using
In the sample tested to model.
Alternatively, modeling 12 also provides for data analysis description instrument, in order to which business side is to data
Explored and described, checked data and understand its feature, the value distribution of such as data, deletion condition, no
The features such as the correlation between the influence with value to target variable, variable, in order to the quick desampling in business side
The distribution of notebook data and quality.Wherein, target variable is the output variable of model.
Modeling 12 also provides for all kinds of modeling languages and instrument, in order to which business root is according to business objective,
The appropriate model method of selection, determines variable combination and the weight of model, and it is legal, reasonable, soluble,
Final mask is determined on the basis of enforceable.Built for example, modeling 12 can provide SAS, R, SPM etc.
Mould language and instrument, and provide decision tree, TreeNet, logistic regression, random forest, neutral net,
The model methods such as deep learning select for business side.Under normal conditions, business can be attempted to choose a variety of
Model, then therefrom the optimal model of selection performance is used as final mask.
After the formulation of model is completed, modeling 12 is trained using modeling sample to model, and in mould
Type is reached after target, is delivered decision system 13 and is implemented.Alternatively, when modeling sample also includes being used for
During the sample tested to model, modeling 12 can be additionally used in the mould to completion training using modeling sample
Type is tested, to ensure the precision of prediction of model.
Decision system 13, the model for being trained using above-mentioned completion carries out operational decision making.Operational decision making refers to
The relevant information of targeted customer is predicted for business side.Specifically, decision system 13 specifically for：
Obtain the operational decision making request for carrying the identity of targeted customer；According to the identity from user profile
System 11 obtains the second user data of targeted customer；By completing the model of training according to the of targeted customer
Two user data, calculate operational decision making result.
Alternatively, modeling 12, are additionally operable to：Visual model editing management deployment tool is provided；Adopt
Deployment tool is managed with the model editing, is rule language by the model conversion for completing training；By rule language
Executable module is compiled into, decision system 13 is deployed to.Model editing management deployment tool is provided with visualization
Interface, and with by the function that model conversion is rule language, so that business side turns the model for completing training
Rule language is changed to, executable module is compiled into.For example, the module that modeling 12 is provided includes but not limited
In decision tree, rule set, tables of data, scorecard, data method etc..Compared to using such as C++ sentences
Carry out descriptive model, by the above-mentioned means, using rule language descriptive model, the industry to business side can be reduced
The technical requirements of business personnel so that general service personnel model can be also designed, tested, run and
Monitoring, to reduce the dependence to professional IT (Information Technology, information technology) technical staff.
Modeling 12, is additionally operable to：Obtain by the business root targeted customer's that factually border state of affairs is provided
First user data；According to the first user data of targeted customer and second user data generation optimization sample；
Model is optimized using optimization sample.
In summary, the system that the present embodiment is provided, by modeling in provided by business side and business
On the basis of first user data of the related sample of users of target, sample of users is obtained from customer information system
Second user data, and then according to above-mentioned data generate modeling sample, also by modeling obtain by industry
The model that business root is formulated according to business objective, and model is trained using above-mentioned modeling sample, then will
The model for completing training is deployed to decision system implementation, to carry out operational decision making；Solve in the prior art,
Development cost present in business side itself development model is high, development efficiency is low and model accuracy is poor asks
Topic；The system customization development model that business side is provided by the present embodiment, contribute to save development cost,
Improve development efficiency and improve model accuracy.
In addition, the system that the present embodiment is provided, also passes through first user data of the modeling to sample of users
Handled with second user data, set up basic data collection, a pair sample variable related to business objective enters
Row processing, chooses the target sample variable with strong predictive ability, the number according to corresponding to target sample variable
According to generation modeling sample has substantially ensured that sample quality, has been that the predictive ability of model provides safeguard.
In addition, the system that the present embodiment is provided, is also provided by visual model editing management deployment tool,
Implemented so that model conversion is deployed into decision system as rule language and being compiled into after executable module, fully
Reduce the technical requirements to the business personnel of business side so that general service personnel can also be carried out to model
Design, test, operation and monitoring, to reduce the dependence to professional IT technical staff.
In addition, the system that the embodiment of the present disclosure is provided, it is adaptable to all kinds of industries and field with modeling requirement,
The disclosure is not construed as limiting to the type of business side.For example, business side can be credit agency, credit agency leads to
Said system customized development model is crossed, helps to lift the air control level of decision-making of credit agency.
Fig. 2 is refer to, the schematic diagram of the server architecture provided it illustrates one embodiment of the invention, the clothes
Business device framework is used for the operational decision making system based on business model for disposing above-described embodiment offer.Such as Fig. 2 institutes
Show, the server architecture includes：Database server 21, upload server 22, sample server 23, build
Mould server 24, rule management server 25, policy server 26 and access server 27.
Customer information system includes database server 21.Database server 21 respectively with upload server
22nd, sample server 23 and policy server 26 set up communication connection.Database server 21 is used to store
Each item data of user, the data access function with mass users.In database server 21, with
Family is kernel object, is stored with for modeling the second user data with decision-making.
Modeling includes upload server 22, sample server 23, Modeling Server 24 and regulation management
Server 25.
Upload server 22 is set up with business method, apparatus, database server 21 and sample server 23 respectively
Communication connection.Upload server 22 is used for the sample of users related to business objective for receiving the upload of business side
First user data, and mix the sample with first user data at family and be sent to database server 21 and stored.
The first user data that upload server 22 is additionally operable to mix the sample with family is sent to sample server 23.
Sample server 23 is built with database server 21, upload server 22 and Modeling Server 24 respectively
Vertical communication connection.Sample server 23 is used for the first number of users that sample of users is received from upload server 22
According to, and from the second user data of the acquisition sample of users of database server 21, according to the first of sample of users
User data and second user data generation modeling sample.Sample server 23 is additionally operable to send modeling sample
To Modeling Server 24.
Modeling Server 24 is set up with sample server 23 and rule management server 25 communicate to connect respectively.
Modeling Server 24 is used to receive modeling sample from sample server 23, obtains the modeling personnel by business side
The model formulated according to business objective, and model is trained using modeling sample, obtain completing training
Model, for operational decision making.Modeling Server 24 is additionally operable to the model for completing training being sent to regular pipe
Manage server 25.
Rule management server 25 is set up with Modeling Server 24 and policy server 26 communicate to connect respectively.
Rule management server 25 is used for the model that training is finished receiving from Modeling Server 24, regulation management service
Device 25 is provided with visual model editing to the rules administrator of business side and manages deployment tool, regulation management
Member can manage deployment tool using above-mentioned model editing, be rule language by the model conversion for completing training, and
Rule language is compiled into executable module, policy server 26 is deployed to.
Decision system includes policy server 26 and access server 27.
Policy server 26 respectively with rule management server 25, access server 27 and database server
21 set up communication connection, and access server 27 is set up with business method, apparatus and policy server 26 communicate respectively
Connection.Access server 27 is used for the identity for carrying targeted customer for receiving the transmission of business method, apparatus
Operational decision making is asked, and operational decision making request is transmitted into policy server 26.Policy server 26 is used for
Operational decision making request is received from access server 27, mesh is obtained from database server 21 according to identity
The second user data of user are marked, and by second user data of the above-mentioned model according to targeted customer, are calculated
The result of decision.Policy server 26 is additionally operable to the result of decision being sent to access server 27.Access server
27 are additionally operable to receive the result of decision that policy server 26 is sent, and the result of decision is transmitted into business method, apparatus.
Certainly, in other possible embodiments, Modeling Server 24 sets up logical with policy server 26
The model for completing training is directly deployed to policy server 26 by letter connection, Modeling Server 24.
Alternatively, upload server 22 is additionally operable to receive by the business root target that factually border state of affairs is provided
The first user data of user, and the first user data of targeted customer is sent to database server 21 entered
Row storage.Upload server 22 is additionally operable to the first user data of targeted customer being sent to sample server 23.
Sample server 23 is additionally operable to receive the first user data of targeted customer from upload server 22, and from number
The second user data of targeted customer are obtained according to storehouse server 21, according to the first user data of targeted customer and
Second user data generation optimization sample.Sample server 23 is additionally operable to optimization sample being sent to modeling service
Device 24.Modeling Server 24 is additionally operable to receive optimization sample from sample server 23, using optimization sample pair
The model of above-mentioned completion training is optimized.
Alternatively, the server architecture also includes：Monitoring management server 28.28 points of monitoring management server
Do not set up and communicate to connect with upload server 22, policy server 26 and access server 27.Monitoring management
Server 28 is used for when traffic direction upload server 22 uploads the first user data of sample of users, right
Business side carries out rights management.Monitoring management server 28 is additionally operable to determine in the carry out of policy server 26 business
During plan, business monitoring is carried out.Monitoring management server 28 is additionally operable to obtain employment in access server 27
When business side receives operational decision making request, authentication management is carried out to business side.By disposing monitoring management server
28, it is favorably improved the safety and stability of system.
In one example, exemplified by being credit agency by business side, credit agency provides credit operation.Assuming that
Business objective is to provide installment credit to the university student client of high-quality, identity fraud is taken precautions against, to obtain through commercial
Profit.Correspondingly, the function for the model to be built is to predict that promise breaking, the possibility of fraud occurs in targeted customer
Property.Business root is according to above-mentioned business objective, and it is the universities of Guang Shen 211 that go up north to determine the applicable people's group object of model
It is full-time, and there is no within past 1 year promise breaking, the student of fraud occur.Afterwards, business side from
In business existing business datum at initial stage, sample of users is chosen, and accordingly obtains the first user of sample of users
Data.For example, sample of users, which is Shenzhen, has handled the university student of installment credit business, the of sample of users
One user data may include whether record sample of users the feelings such as credit, refund, promise breaking, fraud and lost contact occurs
The data of condition.Business side mixes the sample with first user data upload at family to upload server 22.Upload service
The first user data that device 22 mixes the sample with family is sent to sample server 23.Sample server 23 is from data
Storehouse server 21 obtain sample of users second user data (such as user basic information, network application data,
User's financial data etc.), and the first user data according to sample of users and the generation modeling of second user data
Sample.Modeling sample is sent to Modeling Server 24 by sample server 23.Modeling Server 24 obtain by
The model that the modeling personnel of business side formulate according to business objective, and model is trained using modeling sample,
Obtain completing the model of training, and the model that the completion is trained is deployed to policy server 26.When business side
When needing to predict that promise breaking, the possibility of fraud occurs in targeted customer, business can be sent to policy server 26
Carry the operational decision making request of the identity (such as identification card number) of targeted customer, policy server 26
According to the identity of targeted customer, the second user data for obtaining targeted customer are matched from database server 21,
And it is right according to the input variable institute of input variable Selection Model from the second user data of targeted customer of model
The data answered, then calculate operational decision making result by model, and the operational decision making result is used to reflect target
There is promise breaking, the possibility of fraud in user.Operational decision making result is then fed back to business by policy server 26
Side.In addition, business side can be also provided for reflecting target according to practical business situation to upload server 22
Whether user actually there is fraud, the first user data of violation of agreement, in order to sample server 23 accordingly
Build optimization sample to optimize model, improve the applicability and accuracy of model.
When business side carries out model development using system provided in an embodiment of the present invention, it is only necessary to according to existing industry
Business data, the first user data of sample of users are provided to system, and select suitably according to business objective
Model method.Training and inspection on the structure, model of modeling sample can be completed by system automation,
Compared to prior art, the requirement to the development ability of business side is substantially reduced, and substantially increases model
Development efficiency.Further, since customer information system is provided with the user data of magnanimity, compared to prior art
Middle business side is modeled according only to itself can be obtained a small amount of modeling data, and business side is opened using the system
Model is sent out, the modeling data of more horn of plenty can be obtained, the reliability of modeling sample is favorably improved, so that
Improve the accuracy for the model developed.
In addition, for each server included in the server architecture shown in Fig. 2, its function can be with
Realized by a server, or the server cluster that its function can be also made up of multiple servers is realized, or
The function of a variety of servers of person is integrated in a server realization.Fig. 3 shows that one embodiment of the invention is carried
The structural representation of the server of confession.The server can be implemented as in the server architecture shown in Fig. 2
Any one server.Specifically：
The server 300 includes CPU (CPU) 301 including random access memory (RAM)
302 and the system storage 304 of read-only storage (ROM) 303, and the connection He of system storage 304
The system bus 305 of CPU 301.The server 300 also includes helping each in computer
The basic input/output (I/O systems) 306 of information is transmitted between device, and for storage program area
313rd, the mass-memory unit 307 of application program 314 and other program modules 315.
The basic input/output 306 includes for the display 308 of display information and for user
Input the input equipment 309 of such as mouse, keyboard etc of information.Wherein described display 308 and input are set
Standby 309 are all connected to CPU by being connected to the IOC 310 of system bus 305
301.The basic input/output 306 can also include IOC 310 for receive and
Handle the input from multiple other equipments such as keyboard, mouse or electronic touch pen.Similarly, input defeated
Go out controller 310 and also provide output to display screen, printer or other kinds of output equipment.
The mass-memory unit 307 is by being connected to the bulk memory controller of system bus 305 (not
Show) it is connected to CPU 301.The mass-memory unit 307 and its associated computer
Computer-readable recording medium is that server 300 provides non-volatile memories.That is, the mass-memory unit 307
The computer-readable medium (not shown) of such as hard disk or CD-ROM drive etc can be included.
Without loss of generality, the computer-readable medium can include computer-readable storage medium and communication media.
Computer-readable storage medium include for storage such as computer-readable instruction, data structure, program module or
Volatibility that any methods or techniques of the information such as other data is realized and non-volatile, removable and not removable
Dynamic medium.Computer-readable storage medium include RAM, ROM, EPROM, EEPROM, flash memory or other
Its technology of solid-state storage, CD-ROM, DVD or other optical storages, cassette, tape, disk storage
Or other magnetic storage apparatus.Certainly, skilled person will appreciate that the computer-readable storage medium does not limit to
In above-mentioned several.Above-mentioned system storage 304 and mass-memory unit 307 may be collectively referred to as memory.
According to various embodiments of the present invention, the server 300 can also pass through the networks such as internet
It is connected to the remote computer operation on network.Namely server 300 can be total by being connected to the system
NIU 311 on line 305 is connected to network 312, in other words, can also use network interface list
Member 311 is connected to other kinds of network or remote computer system (not shown).
The memory also includes one or more than one program, one or more than one program
It is stored in memory, and is configured to by one or more than one computing device.Said one or
More than one program bag contains the instruction for being used for realizing the various functions of server 300.
Below, it is based on business there is provided one kind for credit operation (correspondingly, business side is credit agency)
The operational decision making system of model.Fig. 4 is refer to, it illustrates the architecture diagram of the system, the system can be with
Including：Customer information system 41, data collecting system (not shown), air control policy system 43, rule
Then automotive engine system 44 and rating engine system 45.
Customer information system 41 is used for each item data for storing user, the data access function with mass users.
Customer information system 41 is the distribution centre of user profile, and the user profile that upper-layer service needs is all from user's letter
Breath system 41；Upper strata produce data, such as user's scorecard, user's rating system, user's Classification Management,
The information such as quota control also land customer information system 41.Alternatively, customer information system 41 includes：User
Information bank, risk data fairground and product database.User information database is used for the second use for storing each user
User data, the first user data that business side is provided can be also stored into user information database.Risk data fairground
For storing the user profile related to financial risks.Product database is used for the data for storing upper strata generation,
Such as user's scorecard, user's rating system, user's Classification Management, quota control information.
Data collecting system, which is included, is used for the data with being communicated inside and outside the model development and operational decision making system
Interface.On the one hand, data collecting system is according in business demand, with the model development and operational decision making system
Portion and outside are communicated, and realization is horizontal computerized with internal data and external data, collects subscriber data,
To build modeling sample and modeling.On the other hand, data collecting system is additionally operable to be communicated with business side,
The operational decision making request that reception business side is sent, to business side's feedback traffic result of decision.Wherein, data connect
The agreement that mouth layer is used includes but is not limited to TCP (Transmission Control Protocol, transmission control
Agreement), UDP (User Datagram Protocol, UDP), HTTP (HyperText
Transfer Protocol, HTTP) etc. agreement.
It is deployed with all kinds of universal models related to credit operation in air control policy system 43 and business side formulates
Customize model.Above-mentioned model includes but is not limited to：Anti- fraud management model, quota control model, price
Model, borrow before client's screening model and credit risk management model, borrow in tune volume model and by stages model, borrow
Risk Monitoring model and collection administrative model afterwards etc..
It is rule language that rule engine system 44, which is used for model conversion,.Rule engine system 44 includes rule
Engine, regulation engine is a kind of reusable computation module of insertion in the application, it by domain knowledge and
Business rule set is stored and managed as knowledge base, receives data input, according to data-oriented and is known
Know storehouse to make inferences, parsing performs respective rule, and makes operational decision making according to rule.Regulation engine is by pushing away
Reason engine is developed.Regulation engine is realized separates operational decision making logic from application code,
And write business rule logic using predefined semantic modules.Business rule logic is run in regulation engine,
Regulation engine receives data input, explains business rule and make operational decision making according to business rule.By making
It can effectively support this changeable operation flow to realize with regulation engine dynamic configuration air control strategy, also simultaneously
The development time of project can effectively be shortened.Rule engine system 44 can include following three part：First
It is IDE (Integrated Development Environment, IDE) to divide, for IT personnel
Required framework is created, to create business and management rule；Part II is regulation management application program (Rule
Management Application), for providing the user interface for creating and changing rule to business side；
Part III is regulation engine (Rules Engine) itself, and regulation engine is connected with decision engine, with based on
The rule created in rule base is come the operational decision making that is automated.
Rating engine system 45 provides various conventional scoring modeling tools and template, supports rapid modeling；There is provided
Simulation Test Environment, can be verified to the quick test of model；And Distributed Computing Platform, support pair are provided
Mass data carries out computing.
In one example, Fig. 5 is refer to, is determined it illustrates a kind of using the above-mentioned business based on business model
Plan system carries out the schematic diagram of the development process of model development.As shown in figure 5, the development process can be divided into mould
Type prepares, model development and model dispose three phases.
The step of model preparation stage is related to includes：
1st, business objective is determined：It is described using the business objective to be realized to model of the language being easily understood.
Exemplified by being credit agency by business side, business objective can provide installment credit to the university student client of high-quality,
Identity fraud is taken precautions against, to obtain Operating profit.
2nd, applicable object is determined：The applicable people's group object of model is specified, the target variable of accurate description model
Definition, explicitly defines the performance time window and observing time window of target variable.For example, the applicable crowd of model
Object is the full-time of the universities of Guang Shen 211 of going up north, and without appearance promise breaking, fraud within past 1 year
Student.
3rd, data acquisition and integration：The source of identification data and scope, it is ensured that can effectively obtain these data,
And these Data Integrations are processed as being adapted to the form that further data prepare, such as variable partitions, data merge,
Variable derivative etc..In this stage, business side provides existing business datum, and the student at such as business initial stage is by stages
Loan credit, refund, promise breaking, fraud, the first user data of lost contact；And system can be according to the body of student
Part mark, matches household register, student status, hobby, working condition, payment data, the consumption number of user
According to, second user data such as finance data, social data, and it is integrated into unified modeling sample.
4th, Data Mining and description：Check data and arrange its feature, the value distribution of data, deletion condition,
Correlation between influence of the different values to target variable, variable etc..System provides complete data analysis and retouched
Instrument is stated, in order to which business side quickly understands distribution and the quality of sample data.
5th, ETL data prepare：By carrying out cleaning conversion to data, the variable with stronger predictive power is created,
Reduce radix and optimal segmentation is carried out to continuous variable.System provides the variable deriving method and outstanding of specialty
Variable segmentation tool, help to be rapidly completed the ETL processes of data.
The step of model development stage is related to includes：
6th, variables choice：In the data of various dimensions of comforming, select the variable useful to model, can by by
Step recurrence, value of information IV, PCA, Variable cluster etc. are picked out in statistical significance significantly, and with preferably solution
The variable for the property released.The method that system provides various variables choices, helps business side quickly to select optimum prediction power
Variable.
7th, model is selected：According to the characteristics of business datum, the suitable business model of selection, the mould that system is provided
Type includes：Decision tree, TreeNet, logistic regression, random forest, neutral net, deep learning etc. are common
Model is available.Business side can typically attempt a variety of models, and then therefrom selection shows optimal model.
8th, model is set up：The standard of model optimization is explicitly defined, the parameter of model is determined；From numerous candidates
Final mask is selected in model；And judgment models whether there is poor fitting or over-fitting and multicollinearity etc.
Problem.
9th, model is verified：Whether testing model reaches acceptable levels of accuracy, and model is in different pieces of information collection
Whether middle performance is sane, and whether model result can parse.
The step of model deployment phase is related to includes：
10th, model configuration management：Visual model editing management deployment tool in, by model translation into
Rule language, is compiled into executable module, and dispose arrive operating environment.It is common that regulation engine is provided
Module includes decision tree, rule set, tables of data, scorecard, data method etc..
11st, model calculation engine：Send request data arrive decision engine, decision engine from user information database with
User related data is extracted in risk data fairground and submits to regulation engine, regulation engine is according to the model of business
Formula rule is calculated, and returns to operational decision making result.
12nd, model running is monitored：The implementation status of monitoring model, modelling effect monitoring analysis, model is related to
Variable changes in distribution, the stability of model, accuracy monitoring.
Following is the inventive method embodiment, for the details not disclosed in the inventive method embodiment, please be joined
According to present system embodiment.
Fig. 6 is refer to, it illustrates the operational decision making side based on business model that one embodiment of the invention is provided
The flow chart of method, this method may include steps of：
Step 601, the first number of users of the sample of users related to business objective provided by business side is provided
According to.
Step 602, the second user data of sample of users are obtained from customer information system.
Step 603, modeling sample is generated according to the first user data of sample of users and second user data.
Step 604, the model formulated by business root according to business objective is obtained.
Step 605, model is trained using modeling sample.
Step 606, operational decision making is carried out using the model for completing training, operational decision making refers to be business side to mesh
The relevant information of mark user is predicted.
Alternatively, above-mentioned steps 603 can include following several sub-steps：
The first user data and second user data of sample of users are handled, basic data collection is set up,
Basic data is concentrated comprising the data corresponding to the sample variable related to business objective；
Pair sample variable related to business objective is handled, and chooses target sample variable, and target sample becomes
Amount refers to the sample variable with strong predictive ability；
Data according to corresponding to target sample variable, generate modeling sample.
Alternatively, above-mentioned steps 606 can include following several sub-steps：
Obtain the operational decision making request for carrying the user identity information of targeted customer；
The second user data of targeted customer are obtained from customer information system according to user identity information；
Model by completing training calculates operational decision making result according to the second user data of targeted customer.
Alternatively, also include after above-mentioned steps 606：
Obtain by business root factually border state of affairs provide targeted customer the first user data；
According to the first user data of targeted customer and second user data generation optimization sample；
Model is optimized using optimization sample.
Alternatively, also include after above-mentioned steps 605：
Deployment tool is managed using visual model editing, is rule language by the model conversion for completing training,
And rule language is compiled into executable module.
In summary, the method that the present embodiment is provided, by modeling in provided by business side and business
On the basis of first user data of the related sample of users of target, sample of users is obtained from customer information system
Second user data, and then according to above-mentioned data generate modeling sample, also by modeling obtain by industry
The model that business root is formulated according to business objective, and model is trained using above-mentioned modeling sample, then will
The model for completing training is deployed to decision system implementation, to carry out operational decision making；Solve in the prior art,
Development cost present in business side itself development model is high, development efficiency is low and model accuracy is poor asks
Topic；The system customization development model that business side is provided by the present embodiment, contribute to save development cost,
Improve development efficiency and improve model accuracy.
It should be appreciated that referenced herein " multiple " refer to two or more."and/or", is retouched
The incidence relation of affiliated partner is stated, expression may have three kinds of relations, for example, A and/or B, can be represented：
Individualism A, while there is A and B, these three situations of individualism B.Before and after character "/" is typicallyed represent
Affiliated partner is a kind of relation of "or".
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
One of ordinary skill in the art will appreciate that realizing all or part of step of above-described embodiment can pass through
Hardware is completed, and the hardware of correlation can also be instructed to complete by program, described program can be stored in
In a kind of computer-readable recording medium, storage medium mentioned above can be read-only storage, disk or
CD etc..
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all the present invention's
Within spirit and principle, any modification, equivalent substitution and improvements made etc. should be included in the present invention's
Within protection domain.
Claims (10)
1. a kind of operational decision making system based on business model, it is characterised in that the system includes：User
Information system, modeling and decision system；
The modeling, the of the sample of users related to business objective provided for obtaining by business side
One user data；The second user data of the sample of users are obtained from the customer information system；According to institute
State the first user data and second user data generation modeling sample of sample of users；Obtain by the business side
The model formulated according to the business objective；The model is trained using the modeling sample, and will
The model for completing training is deployed to the decision system；
The decision system, for carrying out operational decision making using the model for completing training, the business is determined
Plan refers to be predicted the relevant information of targeted customer for the business side.
2. according to the method described in claim 1, it is characterised in that the modeling, specifically for：
The first user data and second user data of the sample of users are handled, basic data is set up
Collection, the basic data is concentrated comprising the data corresponding to the sample variable related to the business objective；
Pair sample variable related to the business objective is handled, and chooses target sample variable, the mesh
Mark sample variable refers to the sample variable with strong predictive ability；
Data according to corresponding to the target sample variable, generate the modeling sample.
3. system according to claim 1, it is characterised in that the decision system, specifically for：
Obtain the operational decision making request for carrying the identity of the targeted customer；
The second user number of the targeted customer is obtained from the customer information system according to the identity
According to；
By second user data of the model according to the targeted customer, operational decision making result is calculated.
4. system according to claim 3, it is characterised in that the modeling, is additionally operable to：
Obtain by the business root factually border state of affairs provide the targeted customer the first number of users
According to；
According to the first user data of the targeted customer and second user data generation optimization sample；
The model is optimized using the optimization sample.
5. the system according to any one of Claims 1-4, it is characterised in that the modeling,
It is additionally operable to：
Visual model editing management deployment tool is provided；
Deployment tool is managed using the model editing, is rule language by the model conversion for completing training；
The rule language is compiled into executable module, the decision system is deployed to.
6. a kind of operational decision making method based on business model, it is characterised in that methods described includes：
First user data of the sample of users related to business objective provided by business side is provided；
The second user data of the sample of users are obtained from customer information system；
According to the first user data of the sample of users and second user data generation modeling sample；
Obtain the model formulated by the business root according to the business objective；
The model is trained using the modeling sample；
Operational decision making is carried out using the model for completing training, the operational decision making refers to for the business side
Relevant information to targeted customer is predicted.
7. method according to claim 6, it is characterised in that described according to the of the sample of users
One user data and second user data generation modeling sample, including：
The first user data and second user data of the sample of users are handled, basic data is set up
Collection, the basic data is concentrated comprising the data corresponding to the sample variable related to the business objective；
Pair sample variable related to the business objective is handled, and chooses target sample variable, the mesh
Mark sample variable refers to the sample variable with strong predictive ability；
Data according to corresponding to the target sample variable, generate the modeling sample.
8. method according to claim 6, it is characterised in that described using the mould for completing training
Type carries out operational decision making, including：
Obtain the operational decision making request for carrying the identity of the targeted customer；
The second user number of the targeted customer is obtained from the customer information system according to the identity
According to；
By completing the model of training according to the second user data of the targeted customer, calculating business is determined
Plan result.
9. method according to claim 8, it is characterised in that the mould by completing training
Type is according to the second user data of the targeted customer, after calculating operational decision making result, in addition to：
Obtain by the business root factually border state of affairs provide the targeted customer the first number of users
According to；
According to the first user data of the targeted customer and second user data generation optimization sample；
The model is optimized using the optimization sample.
10. the method according to any one of claim 6 to 9, it is characterised in that described using described
After modeling sample is trained to the model, in addition to：
Deployment tool is managed using visual model editing, is rule by the model conversion for completing training
Language, and the rule language is compiled into executable module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610070829.8A CN107025509B (en) | 2016-02-01 | 2016-02-01 | Decision making system and method based on business model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610070829.8A CN107025509B (en) | 2016-02-01 | 2016-02-01 | Decision making system and method based on business model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107025509A true CN107025509A (en) | 2017-08-08 |
| CN107025509B CN107025509B (en) | 2021-06-18 |
Family
ID=59523978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610070829.8A Active CN107025509B (en) | 2016-02-01 | 2016-02-01 | Decision making system and method based on business model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107025509B (en) |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107480725A (en) * | 2017-08-23 | 2017-12-15 | 京东方科技集团股份有限公司 | Image-recognizing method, device and computer equipment based on deep learning |
| CN107808339A (en) * | 2017-09-07 | 2018-03-16 | 广西钱盆科技股份有限公司 | A kind of data acquisition and analysis system |
| CN108536941A (en) * | 2018-03-29 | 2018-09-14 | 上海嘉银金融科技股份有限公司 | The training of air control model and strategy execution system |
| CN108764960A (en) * | 2018-04-12 | 2018-11-06 | 合肥天源迪科信息技术有限公司 | A kind of IN service decision system and method based on big data analysis |
| CN108830454A (en) * | 2018-05-08 | 2018-11-16 | 阿里巴巴集团控股有限公司 | Decision-making technique, device and the electronic equipment of business |
| CN108874487A (en) * | 2018-06-13 | 2018-11-23 | 北京九章云极科技有限公司 | Data analysis processing method and system based on workflow |
| CN109034660A (en) * | 2018-08-22 | 2018-12-18 | 平安科技（深圳）有限公司 | Based on the determination method and relevant apparatus of the risk control strategy of prediction model |
| CN109299178A (en) * | 2018-09-30 | 2019-02-01 | 北京九章云极科技有限公司 | A kind of application method and data analysis system |
| CN109377333A (en) * | 2018-09-03 | 2019-02-22 | 平安科技（深圳）有限公司 | Electronic device determines method and storage medium based on the collection person of disaggregated model |
| CN109409516A (en) * | 2017-08-11 | 2019-03-01 | 微软技术许可有限责任公司 | Machine learning model for the tool depth and width that position is recommended |
| CN109409922A (en) * | 2018-08-31 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Data aggregate modeling method, device, computer equipment and storage medium |
| WO2019061991A1 (en) * | 2017-09-30 | 2019-04-04 | 平安科技（深圳）有限公司 | Multi-element universal model platform modeling method, electronic device, and computer readable storage medium |
| CN109656914A (en) * | 2018-11-07 | 2019-04-19 | 上海前隆信息科技有限公司 | On-line off-line mixed air control modeling training and production dissemination method and system |
| CN109902831A (en) * | 2018-11-05 | 2019-06-18 | 阿里巴巴集团控股有限公司 | Operational decision making processing method and device |
| CN109949148A (en) * | 2019-02-27 | 2019-06-28 | 四川享宇金信金融科技有限公司 | A kind of automation air control configuration system and method for credit financing business |
| CN110135701A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Control automatic generation method, device, electronic equipment and the readable medium of rule |
| CN110458600A (en) * | 2019-07-08 | 2019-11-15 | 平安科技（深圳）有限公司 | Portrait model training method, device, computer equipment and storage medium |
| CN110489131A (en) * | 2018-05-15 | 2019-11-22 | ***通信集团浙江有限公司 | A kind of gray scale user choosing method and device |
| CN110689134A (en) * | 2018-07-05 | 2020-01-14 | 第四范式（北京）技术有限公司 | Method, apparatus, device and storage medium for performing machine learning process |
| CN110851482A (en) * | 2019-11-07 | 2020-02-28 | 支付宝(杭州)信息技术有限公司 | Method and device for providing data model for multiple data parties |
| CN110874667A (en) * | 2018-08-31 | 2020-03-10 | 重庆小雨点小额贷款有限公司 | Scoring card model optimization method and device, server and storage medium |
| WO2020048056A1 (en) * | 2018-09-06 | 2020-03-12 | 平安科技（深圳）有限公司 | Risk decision method and apparatus |
| CN111104422A (en) * | 2019-12-10 | 2020-05-05 | 北京明略软件***有限公司 | Training method, device, equipment and storage medium of data recommendation model |
| CN111240648A (en) * | 2020-01-20 | 2020-06-05 | 重庆富民银行股份有限公司 | Intelligent management system and method for variables |
| CN111338693A (en) * | 2020-02-22 | 2020-06-26 | 深圳市魔数智擎人工智能有限公司 | Target file generation method based on model construction, server and storage medium |
| CN111382196A (en) * | 2018-12-28 | 2020-07-07 | ***股份有限公司 | Distributed accounting processing method and system |
| CN111401766A (en) * | 2020-03-24 | 2020-07-10 | 支付宝(杭州)信息技术有限公司 | Model, service processing method, device and equipment |
| CN111768291A (en) * | 2020-06-24 | 2020-10-13 | 兴业消费金融股份公司 | Method for information processing, computing device, and computer-readable storage medium |
| CN111861096A (en) * | 2020-05-30 | 2020-10-30 | 上海维信荟智金融科技有限公司 | Method and system for visual configuration management of decision flow |
| CN112767133A (en) * | 2021-01-26 | 2021-05-07 | 北京健康之家科技有限公司 | Service decision method and device, storage medium and computer equipment |
| CN113050955A (en) * | 2019-12-26 | 2021-06-29 | 阿尔法云计算(深圳)有限公司 | Self-adaptive AI model deployment method |
| WO2021169212A1 (en) * | 2020-02-26 | 2021-09-02 | 平安科技（深圳）有限公司 | Risk control model construction method and apparatus, risk control checking method and apparatus, and computer device |
| CN113762688A (en) * | 2021-01-06 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Business analysis system, method and storage medium |
| CN114268661A (en) * | 2021-11-19 | 2022-04-01 | 科大讯飞股份有限公司 | Service scheme deployment method, device, system and equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1967579A (en) * | 2006-02-13 | 2007-05-23 | 湖南大学 | Aid excavating analysis system of financial risk management |
| US7321883B1 (en) * | 2005-08-05 | 2008-01-22 | Perceptronics Solutions, Inc. | Facilitator used in a group decision process to solve a problem according to data provided by users |
| CN101504745A (en) * | 2008-12-04 | 2009-08-12 | 阿里巴巴集团控股有限公司 | Risk regulation/model establishing and optimizing system and method based on network on-line service |
| CN101826184A (en) * | 2010-05-24 | 2010-09-08 | 天津大学 | Method for generating plan based on cognitive model |
| CN102521652A (en) * | 2012-01-10 | 2012-06-27 | 武汉大学 | Evaluation and decision method for operation efficiency of power grid |
| CN103886203A (en) * | 2014-03-24 | 2014-06-25 | 美商天睿信息***（北京）有限公司 | Automatic modeling system and method based on index prediction |
| CN104123395A (en) * | 2014-08-13 | 2014-10-29 | 北京赛科世纪数码科技有限公司 | Decision making method and system based on big data |
| CN105095588A (en) * | 2015-08-05 | 2015-11-25 | 中国联合网络通信集团有限公司 | Mobile internet user complaint prediction method and device |
| CN111339375A (en) * | 2020-03-19 | 2020-06-26 | 中国海洋石油集团有限公司 | Universal big data model configuration and analysis method |
-
2016
- 2016-02-01 CN CN201610070829.8A patent/CN107025509B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7321883B1 (en) * | 2005-08-05 | 2008-01-22 | Perceptronics Solutions, Inc. | Facilitator used in a group decision process to solve a problem according to data provided by users |
| CN1967579A (en) * | 2006-02-13 | 2007-05-23 | 湖南大学 | Aid excavating analysis system of financial risk management |
| CN101504745A (en) * | 2008-12-04 | 2009-08-12 | 阿里巴巴集团控股有限公司 | Risk regulation/model establishing and optimizing system and method based on network on-line service |
| CN101826184A (en) * | 2010-05-24 | 2010-09-08 | 天津大学 | Method for generating plan based on cognitive model |
| CN102521652A (en) * | 2012-01-10 | 2012-06-27 | 武汉大学 | Evaluation and decision method for operation efficiency of power grid |
| CN103886203A (en) * | 2014-03-24 | 2014-06-25 | 美商天睿信息***（北京）有限公司 | Automatic modeling system and method based on index prediction |
| CN104123395A (en) * | 2014-08-13 | 2014-10-29 | 北京赛科世纪数码科技有限公司 | Decision making method and system based on big data |
| CN105095588A (en) * | 2015-08-05 | 2015-11-25 | 中国联合网络通信集团有限公司 | Mobile internet user complaint prediction method and device |
| CN111339375A (en) * | 2020-03-19 | 2020-06-26 | 中国海洋石油集团有限公司 | Universal big data model configuration and analysis method |
Non-Patent Citations (3)
| Title |
|---|
| BREWIS 等: "Decision-making practices in commercial enterprises: A cybernetic intervention into a business model", 《JOURNAL OF ORGANISATIONAL TRANSFORMATION & SOCIAL CHANGE》 * |
| 尹蕊 等: "电网企业数据中心业务模型的研究", 《电子世界》 * |
| 李鹏: "智能电网运营管理风险元传递模型及决策支持***研究", 《中国博士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (45)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109409516A (en) * | 2017-08-11 | 2019-03-01 | 微软技术许可有限责任公司 | Machine learning model for the tool depth and width that position is recommended |
| CN107480725A (en) * | 2017-08-23 | 2017-12-15 | 京东方科技集团股份有限公司 | Image-recognizing method, device and computer equipment based on deep learning |
| CN107808339A (en) * | 2017-09-07 | 2018-03-16 | 广西钱盆科技股份有限公司 | A kind of data acquisition and analysis system |
| WO2019061991A1 (en) * | 2017-09-30 | 2019-04-04 | 平安科技（深圳）有限公司 | Multi-element universal model platform modeling method, electronic device, and computer readable storage medium |
| CN108536941A (en) * | 2018-03-29 | 2018-09-14 | 上海嘉银金融科技股份有限公司 | The training of air control model and strategy execution system |
| CN108764960A (en) * | 2018-04-12 | 2018-11-06 | 合肥天源迪科信息技术有限公司 | A kind of IN service decision system and method based on big data analysis |
| CN108830454B (en) * | 2018-05-08 | 2021-12-28 | 创新先进技术有限公司 | Service decision method and device and electronic equipment |
| CN108830454A (en) * | 2018-05-08 | 2018-11-16 | 阿里巴巴集团控股有限公司 | Decision-making technique, device and the electronic equipment of business |
| CN110489131B (en) * | 2018-05-15 | 2023-04-07 | ***通信集团浙江有限公司 | Gray level user selection method and device |
| CN110489131A (en) * | 2018-05-15 | 2019-11-22 | ***通信集团浙江有限公司 | A kind of gray scale user choosing method and device |
| CN108874487A (en) * | 2018-06-13 | 2018-11-23 | 北京九章云极科技有限公司 | Data analysis processing method and system based on workflow |
| CN110689134A (en) * | 2018-07-05 | 2020-01-14 | 第四范式（北京）技术有限公司 | Method, apparatus, device and storage medium for performing machine learning process |
| CN109034660A (en) * | 2018-08-22 | 2018-12-18 | 平安科技（深圳）有限公司 | Based on the determination method and relevant apparatus of the risk control strategy of prediction model |
| CN109034660B (en) * | 2018-08-22 | 2023-07-14 | 平安科技（深圳）有限公司 | Method and related device for determining risk control strategy based on prediction model |
| CN109409922A (en) * | 2018-08-31 | 2019-03-01 | 深圳壹账通智能科技有限公司 | Data aggregate modeling method, device, computer equipment and storage medium |
| CN110874667A (en) * | 2018-08-31 | 2020-03-10 | 重庆小雨点小额贷款有限公司 | Scoring card model optimization method and device, server and storage medium |
| CN109377333A (en) * | 2018-09-03 | 2019-02-22 | 平安科技（深圳）有限公司 | Electronic device determines method and storage medium based on the collection person of disaggregated model |
| WO2020048056A1 (en) * | 2018-09-06 | 2020-03-12 | 平安科技（深圳）有限公司 | Risk decision method and apparatus |
| CN109299178A (en) * | 2018-09-30 | 2019-02-01 | 北京九章云极科技有限公司 | A kind of application method and data analysis system |
| CN109902831B (en) * | 2018-11-05 | 2023-04-07 | 创新先进技术有限公司 | Service decision processing method and device |
| CN109902831A (en) * | 2018-11-05 | 2019-06-18 | 阿里巴巴集团控股有限公司 | Operational decision making processing method and device |
| CN109656914A (en) * | 2018-11-07 | 2019-04-19 | 上海前隆信息科技有限公司 | On-line off-line mixed air control modeling training and production dissemination method and system |
| CN111382196B (en) * | 2018-12-28 | 2023-10-03 | ***股份有限公司 | Distributed accounting processing method and system |
| CN111382196A (en) * | 2018-12-28 | 2020-07-07 | ***股份有限公司 | Distributed accounting processing method and system |
| CN109949148A (en) * | 2019-02-27 | 2019-06-28 | 四川享宇金信金融科技有限公司 | A kind of automation air control configuration system and method for credit financing business |
| CN110135701A (en) * | 2019-04-23 | 2019-08-16 | 北京淇瑀信息科技有限公司 | Control automatic generation method, device, electronic equipment and the readable medium of rule |
| CN110458600A (en) * | 2019-07-08 | 2019-11-15 | 平安科技（深圳）有限公司 | Portrait model training method, device, computer equipment and storage medium |
| CN110851482A (en) * | 2019-11-07 | 2020-02-28 | 支付宝(杭州)信息技术有限公司 | Method and device for providing data model for multiple data parties |
| CN110851482B (en) * | 2019-11-07 | 2022-02-18 | 支付宝(杭州)信息技术有限公司 | Method and device for providing data model for multiple data parties |
| CN111104422B (en) * | 2019-12-10 | 2023-08-29 | 北京明略软件***有限公司 | Training method, device, equipment and storage medium of data recommendation model |
| CN111104422A (en) * | 2019-12-10 | 2020-05-05 | 北京明略软件***有限公司 | Training method, device, equipment and storage medium of data recommendation model |
| CN113050955A (en) * | 2019-12-26 | 2021-06-29 | 阿尔法云计算(深圳)有限公司 | Self-adaptive AI model deployment method |
| CN111240648B (en) * | 2020-01-20 | 2023-02-28 | 重庆富民银行股份有限公司 | Intelligent management system and method for variables |
| CN111240648A (en) * | 2020-01-20 | 2020-06-05 | 重庆富民银行股份有限公司 | Intelligent management system and method for variables |
| CN111338693B (en) * | 2020-02-22 | 2023-07-14 | 深圳市魔数智擎人工智能有限公司 | Model construction-based target file generation method, server and storage medium |
| CN111338693A (en) * | 2020-02-22 | 2020-06-26 | 深圳市魔数智擎人工智能有限公司 | Target file generation method based on model construction, server and storage medium |
| WO2021169212A1 (en) * | 2020-02-26 | 2021-09-02 | 平安科技（深圳）有限公司 | Risk control model construction method and apparatus, risk control checking method and apparatus, and computer device |
| CN111401766A (en) * | 2020-03-24 | 2020-07-10 | 支付宝(杭州)信息技术有限公司 | Model, service processing method, device and equipment |
| CN111861096A (en) * | 2020-05-30 | 2020-10-30 | 上海维信荟智金融科技有限公司 | Method and system for visual configuration management of decision flow |
| CN111768291A (en) * | 2020-06-24 | 2020-10-13 | 兴业消费金融股份公司 | Method for information processing, computing device, and computer-readable storage medium |
| CN113762688A (en) * | 2021-01-06 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Business analysis system, method and storage medium |
| CN112767133A (en) * | 2021-01-26 | 2021-05-07 | 北京健康之家科技有限公司 | Service decision method and device, storage medium and computer equipment |
| CN112767133B (en) * | 2021-01-26 | 2024-02-27 | 北京水滴科技集团有限公司 | Service decision method and device, storage medium and computer equipment |
| CN114268661A (en) * | 2021-11-19 | 2022-04-01 | 科大讯飞股份有限公司 | Service scheme deployment method, device, system and equipment |
| CN114268661B (en) * | 2021-11-19 | 2024-04-30 | 科大讯飞股份有限公司 | Service scheme deployment method, device, system and equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107025509B (en) | 2021-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025509A (en) | Decision system and method based on business model | |
| US7584160B2 (en) | System and method for optimizing project subdivision using data and requirements focuses subject to multidimensional constraints | |
| Ayag* | An integrated approach to evaluating conceptual design alternatives in a new product development environment | |
| US8275644B2 (en) | Generating an optimized analytical business transformation | |
| CN108021984A (en) | Determine the method and system of the feature importance of machine learning sample | |
| Wen et al. | A knowledge-based intelligent decision support system for national defense budget planning | |
| Nordin | Challenges in the industrial implementation of generative design systems: An exploratory study | |
| AU2019227432A1 (en) | Matching adopting users and contributing users for decentralized software localization | |
| Menassa et al. | Hybrid model incorporating real options with process centric and system dynamics modeling to assess value of investments in alternative dispute resolution techniques | |
| Boose et al. | Knowledge acquisition techniques for group decision support | |
| Soriano | Maximizing benefits from IT project management: from requirements to value delivery | |
| CN110070264A (en) | Contract evaluation method, platform, equipment and readable storage medium storing program for executing | |
| US10846641B1 (en) | Business decision management system to operational decision management adaptor | |
| He et al. | Integrating a case-based reasoning shell and Web 2.0: design recommendations and insights | |
| Feitosa | Soft Systems Thinking and Aspect-Based Sentiment Analysis to support performance management and better-informed decision-making | |
| Darooghe Arefi et al. | System dynamics model for sports entrepreneurship ecosystem (case study: Iran) | |
| US11983162B2 (en) | Change management process for identifying potential regulatory violations for improved processing efficiency | |
| Kchaou et al. | Application of Fuzzy Logic to Evaluate the Performance of Business Process Models | |
| Bailey | Development of an optimal spatial decision-making system using approximate reasoning | |
| Ciurea | Using genetic algorithms for building metrics of collaborative systems | |
| Idoko et al. | A Semantic Portal to Improve Search on Rivers State’s Independent National Electoral Commission | |
| Li et al. | Case-based reasoning approach for decision-making in building retrofit: A review | |
| Javadi et al. | Identification and fixing bottlenecks of a food manufacturing system using a simulation approach | |
| Parsley | A Canonical Model for Comparing Data and Information Management and Capability Models | |
| Zhang et al. | Factors Influencing the Acceptance of BIM-Based Automated Code Compliance Checking in the AEC Industry in China |
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 |