CN109858927A - A kind of trade company's checking method, device, computer readable storage medium and server - Google Patents
A kind of trade company's checking method, device, computer readable storage medium and server Download PDFInfo
- Publication number
- CN109858927A CN109858927A CN201910041444.2A CN201910041444A CN109858927A CN 109858927 A CN109858927 A CN 109858927A CN 201910041444 A CN201910041444 A CN 201910041444A CN 109858927 A CN109858927 A CN 109858927A
- Authority
- CN
- China
- Prior art keywords
- trade company
- audit
- sample
- merchant information
- dimension
- 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 invention belongs to field of computer technology more particularly to a kind of trade company's checking method, device, computer readable storage medium and servers.The trade company that the terminal device that the method receives trade company is sent audits application, and the merchant identification of the trade company is extracted from trade company audit application；According to the merchant identification respectively from the trade company is obtained in each preset data source in each preset merchant information audited in dimension, wherein each data source corresponds to the merchant information at least one audit dimension；The merchant information vector of the trade company is constructed according to the merchant information in each audit dimension；The merchant information vector of the trade company is calculated using preset audit model, obtain the auditing result of the trade company, in the case where intervening without auditor, both trade company can be audited automatically, substantially increase review efficiency, and the personal experience without relying on auditor, it ensure that the just objective of auditing result.
Description
Technical field
The invention belongs to field of computer technology more particularly to a kind of trade company's checking method, device, computer-readable storages
Medium and server.
Background technique
With the continuous development of development of Mobile Internet technology, the various application platforms for providing service for trade company emerge one after another, example
Such as, the promising trade company take-away platform that take away service available services on Vehicles Collected from Market, also promising trade company provide the payment platform of gathering service
Etc..It is added in application platform to prevent the trade company of irregularity, causes undesirable social influence to application platform, these are answered
The trade company to be added can generally be audited with platform, only auditing the trade company passed through can just be added in application platform.But
It is current trade company's audit, generally requires the experience that special auditor is set according to oneself and manual examination and verification are carried out to trade company,
Review efficiency is low, and the personal experience of heavy dependence auditor, it is difficult to which it is just objective to guarantee.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of trade company's checking method, device, computer readable storage medium and
Server, to solve low efficiency in a manner of existing manual examination and verification, and the personal experience of heavy dependence auditor, it is difficult to guarantee public
Just objective problem.
The first aspect of the embodiment of the present invention provides a kind of trade company's checking method, may include:
Trade company's audit application that the terminal device of trade company is sent is received, and extracts the quotient from trade company audit application
The merchant identification at family；
The trade company is obtained from each preset data source respectively according to the merchant identification in each preset audit
Merchant information in dimension, wherein each data source corresponds to the merchant information at least one audit dimension；
The merchant information vector of the trade company is constructed according to the merchant information in each audit dimension；
The merchant information vector of the trade company is calculated using preset audit model, obtains the audit of the trade company
As a result, the audit model is the machine learning model trained by preset sample set, it include examining in the sample set
Core result be by positive sample and auditing result be unacceptable negative sample, each sample standard deviation corresponds to preset database
In a history trade company audit logging.
The second aspect of the embodiment of the present invention provides a kind of trade company's audit device, may include:
Trade company's audit application receiving module, trade company's audit application that the terminal device for receiving trade company is sent, and from institute
State the merchant identification that the trade company is extracted in trade company's audit application；
Merchant information obtains module, described in being obtained from each preset data source respectively according to the merchant identification
Merchant information of the trade company in each preset audit dimension, wherein each data source corresponds at least one audit dimension
Merchant information；
Merchant information vector constructing module, for constructing the quotient of the trade company according to the merchant information in each audit dimension
Family information vector；
Model computation module is audited, based on carrying out using preset audit model to the merchant information vector of the trade company
It calculates, obtains the auditing result of the trade company, the audit model is the machine learning model trained by preset sample set,
In the sample set include auditing result be by positive sample and auditing result be unacceptable negative sample, each sample
Both correspond to a history trade company audit logging in preset database.
The third aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has computer-readable instruction, and the computer-readable instruction realizes following steps when being executed by processor:
Trade company's audit application that the terminal device of trade company is sent is received, and extracts the quotient from trade company audit application
The merchant identification at family；
The trade company is obtained from each preset data source respectively according to the merchant identification in each preset audit
Merchant information in dimension, wherein each data source corresponds to the merchant information at least one audit dimension；
The merchant information vector of the trade company is constructed according to the merchant information in each audit dimension；
The merchant information vector of the trade company is calculated using preset audit model, obtains the audit of the trade company
As a result, the audit model is the machine learning model trained by preset sample set, it include examining in the sample set
Core result be by positive sample and auditing result be unacceptable negative sample, each sample standard deviation corresponds to preset database
In a history trade company audit logging.
The fourth aspect of the embodiment of the present invention provides a kind of server, including memory, processor and is stored in institute
The computer-readable instruction that can be run in memory and on the processor is stated, the processor executes described computer-readable
Following steps are realized when instruction:
Trade company's audit application that the terminal device of trade company is sent is received, and extracts the quotient from trade company audit application
The merchant identification at family；
The trade company is obtained from each preset data source respectively according to the merchant identification in each preset audit
Merchant information in dimension, wherein each data source corresponds to the merchant information at least one audit dimension；
The merchant information vector of the trade company is constructed according to the merchant information in each audit dimension；
The merchant information vector of the trade company is calculated using preset audit model, obtains the audit of the trade company
As a result, the audit model is the machine learning model trained by preset sample set, it include examining in the sample set
Core result be by positive sample and auditing result be unacceptable negative sample, each sample standard deviation corresponds to preset database
In a history trade company audit logging.
Existing beneficial effect is the embodiment of the present invention compared with prior art: the embodiment of the present invention is receiving trade company
After trade company's audit application that terminal device is sent, the merchant identification of the trade company is extracted from trade company audit application, according to
The merchant identification is respectively from the trade company is obtained in each preset data source in each preset trade company audited in dimension
Information, and the merchant information vector of the trade company is constructed according to the merchant information in each audit dimension, finally using preset
Audit model calculates the merchant information vector of the trade company, obtains the auditing result of the trade company.Due to the audit
Model have passed through a large amount of sample training in advance, obtain general audit from history trade company audit logging by machine learning
Mode both can automatically audit trade company in the case where intervening without auditor, and substantially increase review efficiency, and
Without relying on the personal experience of auditor, the just objective of auditing result ensure that.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of one embodiment flow chart of trade company's checking method in the embodiment of the present invention；
Fig. 2 is that trade company is obtained from each preset data source in each preset audit dimension respectively according to merchant identification
On merchant information schematic flow diagram；
A kind of one embodiment structure chart of trade company's audit device in Fig. 3 embodiment of the present invention；
A kind of schematic block diagram of server in Fig. 4 embodiment of the present invention.
Specific embodiment
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that disclosed below
Embodiment be only a part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
Those of ordinary skill's all other embodiment obtained without making creative work, belongs to protection of the present invention
Range.
Referring to Fig. 1, a kind of one embodiment of trade company's checking method may include: in the embodiment of the present invention
Step S101, trade company's audit application that the terminal device of trade company is sent is received, and from trade company audit application
Extract the merchant identification of the trade company.
When trade company wants to be added in application platform, can be set by terminals such as mobile phone, tablet computer, laptops
It is standby to send trade company's audit application to the responsible execute server for carrying out trade company's audit, its quotient is carry in trade company audit application
Family mark, the merchant identification can be the industrial and commercial registration number of trade company, status of a legal person card number or other can uniquely represent
The mark of the trade company.
The execute server can therefrom extract trade company's mark of the trade company after receiving trade company audit application
Know.
Step S102, the trade company is obtained from each preset data source respectively according to the merchant identification each pre-
If audit dimension on merchant information.
Wherein, each data source corresponds to the merchant information at least one audit dimension.
These merchant informations include but is not limited to: industrial and commercial registration information, legal person's credit information, tax record information, major
The positive and negative face information of website and forum about the trade company etc. audits dimension.
These merchant informations can be from multiple and different data source acquisitions, for example, the server of administrative organization for industry and commerce
Legal person's credit information, the tax row of the trade company stored in the industrial and commercial registration information of the trade company of middle storage, the server of bank
The tax of the trade company stored in the server of political affairs organ records the quotient stored in information, major website and the server of forum
The positive and negative face information at family etc..
As shown in Fig. 2, step S102 can specifically include following process:
Step S1021, the terminal device of Xiang Suoshu trade company sends data grant request.
It include the device identification of the execute server in the data grant request, the device identification is unique represents
The mark of the execute server.
Step S1022, the authorization message of the trade company of the terminal device feedback of the trade company is received.
The terminal device of the trade company is after receiving data grant request, the equipment that will record the lower execute server
It identifies, and feeds back the authorization message of the trade company to the execute server.
Step S1023, server corresponding with s-th of data source is chosen from preset server list to take as data
Business device.
Wherein, 1≤s≤SourceNum, SourceNum are the sum of data source.
Corresponding relationship of the execute server list records between each data source and each data server, specifically
It is as shown in the table:
Step S1024, Xiang Suoshu data server sends request of data.
Include the merchant identification and authorization message of the trade company in the request of data, further includes the execute server
Device identification.
Step S1025, the merchant information for the trade company that the data server is sent is received.
The data server checks the authorization message after receiving the request of data, if verification nothing
Accidentally, then the merchant information of the trade company is found in the local database according to the merchant identification, and execute service to described
Device sends the merchant information of the trade company.
By above procedure, under the premise of obtaining trade company's authorization, then the letter of client is obtained from each data server
Breath, ensure that the safety of merchant information.After obtaining merchant information of the trade company in each preset audit dimension,
It can then be audited according to these merchant informations.
Step S103, the merchant information vector of the trade company is constructed according to the merchant information in each audit dimension.
Since the merchant information obtained in step S102 is not the information of numeralization, it is unfavorable for carrying out analytical calculation, because
This, needs to carry out numeralization processing to the merchant information of each audit dimension first.
For example, being stepped on for this audit dimension of industrial and commercial registration information by what is obtained from the server of administrative organization for industry and commerce
Note information is compared with the information that trade company provides, for example, industrial and commercial registration license can be compared by OCR identification,
Similarity between the two is calculated, the value interval of the similarity is [0,1], is finally stepped on using the similarity as trade company in industry and commerce
Remember the value in this audit dimension of information.
For this audit dimension of legal person's credit information, record of bad behavior therein records such as (for example) breaking one's promise, break a contract is counted
Number, and the threshold value of record of bad behavior is set, it is 1 by the numeralization of this audit dimension of legal person's credit information if being more than the threshold value,
If being less than the threshold value, the ratio of record of bad behavior number Yu the threshold value is calculated, using the ratio as treated the result of quantizing.
It distinguishingly, is 0 by the numeralization of this audit dimension of legal person's credit information if record of bad behavior number is 0.
This audit dimension of information is recorded for the tax, counts record of bad behavior therein (for example, tax evasion etc. records)
Number, and the threshold value of record of bad behavior is set, it is 1 by tax record this audit dimension of information numeralization if being more than the threshold value,
If being less than the threshold value, the ratio of record of bad behavior number Yu the threshold value is calculated, using the ratio as treated the result of quantizing.
It distinguishingly, is 0 by tax record this audit dimension of information numeralization if record of bad behavior number is 0.
For major website and forum about this audit dimension of the positive and negative face information of the trade company, can count according to the following formula
Count the result after value:
Wherein, w is the serial number of website (forum), and 1≤w≤WN, WN are the sum of website (forum), WeightwIt is w-th
The weight of website (forum), can be configured according to the actual situation, generally, will affect the biggish website of power (forum) setting
Higher weight will affect the lesser website of power (forum) and lower weight, Pos be arrangedwIt is w-th of website (forum) about this
The number of the positive information of trade company, NegwNumber for w-th of website (forum) about the negative information of the trade company, NeuwFor w
The number of neutral information of a website (forum) about the trade company, Val are the numeralization result of this audit dimension.
After having carried out numeralization processing, these merchant informations can be constructed in shown in merchant information to
Amount:
InfoVec=(InfoEm1,InfoEm2,...,InfoEmdn,...,InfoEmDN)
Wherein, dn is the serial number for auditing dimension, and 1≤dn≤DN, DN are the sum for auditing dimension, InfoEmdnIt is dn
The merchant information in dimension is audited, InfoVec is the merchant information vector of the trade company.
Step S104, the merchant information vector of the trade company is calculated using preset audit model, is obtained described
The auditing result of trade company.
The audit model is the machine learning model trained by preset sample set, includes in the sample set
Auditing result be by positive sample and auditing result be unacceptable negative sample, each sample standard deviation corresponds to preset data
A history trade company audit logging in library.
It is described audit model building process may include:
Firstly, obtaining each history trade company audit logging from the database, and is audited and remembered according to the history trade company
Record construction sample set.
Wherein, any sample standard deviation in the sample set includes merchant information vector sum auditing result.In order to guarantee to tie
The accuracy of fruit should increase the number of samples in the sample set as far as possible, for example, number of samples can be set to 5000,
10000,50000 or other values.
The number of positive sample in the sample set and the number of negative sample should meet certain restrictive condition, such as:
Wherein, PN is the sum of the positive sample in the sample set, and NgN is the total of the negative sample in the sample set
Number, Thresh are preset threshold value, which can be configured according to the actual situation, for example, can be set to 0.1,
0.01,0.001 or other values.
Distinguishingly, the number of positive sample can be kept consistent with the number of negative sample.For example, if choosing 10000 samples altogether
This, then wherein positive sample and negative sample each 5000, guarantee the balance for analyzing result with this.
Then, audit model as follows is constructed:
Wherein, WtEmdnThe weight coefficient of the merchant information in dimension is audited for the dn, Result is examining for model output
Core result.
Finally, being trained using the sample set to the audit model, until error amount is less than preset error
Until threshold value.
Specifically training process may include:
The tn times sample training is carried out to the audit model according to the following formula:
Wherein, sn is the sample serial number in the sample set, and 1≤sn≤SN, SN are the sample in the sample set
Sum, SampleVecsnFor the merchant information vector of the sn sample, and SampleVecsn=(SpEmsn,1,SpEmsn,2,...,
SpEmsn,dn,...,SpEmsn,DN), SpEmsn,dnTaking in dimension is audited at the dn for the merchant information vector of the sn sample
Value, tn are the number serial number of sample training, WeightVectnFor the weight vectors of the tn times sample training, and WeightVectn
=(WtEmtn,1,WtEmtn,2,...,WtEmtn,dn,...,WtEmtn,DN), WtEmtn,dnFor WeightVectnIt is audited at the dn
Value in dimension, T are transposition symbol, EsResultVectnFor the result vector of the tn times sample training, and
EsResultVectn=(EsValtn,1,EsValtn,2,...,EsValtn,sn,...,EsValtn,SN),Distinguishingly, the weight vectors of the 1st sample training can be according to the actual situation
It is configured, for example, complete 1 vector, i.e. WeightVec can be set to1=(1,1 ..., 1 ..., 1).
The error vector of the tn times sample training is calculated according to the following formula:
Errortn=EsResultVectn-SpResultVec
Wherein, SpResultVec is objective result vector, and SpResultVec=(SpVal1,SpVal2,...,
SpValsn,...,SpValSN), SpValsnFor the sn sample auditing result (for example, if auditing result be pass through, with 1
Indicate, if auditing result is not pass through, indicated with 0), ErrortnFor the error vector of the tn times sample training, and
Errortn=(ErrEmtn,1,ErrEmtn,2,...,ErrEmtn,sn,...,ErrEmtn,SN), ErrEmtn,sn=SpValsn-
EsValtn,sn。
On this basis, the error amount of the tn times sample training can also be further calculated according to the following formula:
Wherein, ErrValtnThe error amount of as the tn times sample training.
If the error amount of the tn times sample training is greater than or equal to the error threshold, the weight vectors are carried out more
Newly, and to the audit model the tn+1 times sample training is carried out.
The error threshold can be configured according to the actual situation, for example, can be set to 5,10,20 or its
Its value.
The renewal process of the weight vectors is shown below:
WeightVectn+1 T=WeightVectn T-λ×SpMatrixT×Errortn T
Wherein, λ is preset update coefficient, can be configured according to the actual situation, for example, can be set to
0.1,0.01,0.001 or other values,
If the error amount of the tn times sample training is less than the error threshold, it is believed that the audit model has been trained
Finish, terminates the training to the audit model.
After the completion of the audit model training, carried out using merchant information vector of the audit model to the trade company
It calculates, the auditing result of the trade company can be obtained.
The auditing result then can be determined that trade company audit passes through, if the auditing result if more than preset result threshold value
Less than or equal to the result threshold value, then it can be determined that trade company audit does not pass through.The result threshold value can be according to the actual situation
Be configured, for example, can be set to 0.4,0.5,0.6 or other values be preferably arranged in the present embodiment
It is 0.5.
In conclusion the embodiment of the present invention receive trade company terminal device send trade company audit application after, from institute
The merchant identification that the trade company is extracted in trade company's audit application is stated, according to the merchant identification respectively from each preset data source
The middle merchant information for obtaining the trade company in each preset audit dimension, and according to the merchant information in each audit dimension
The merchant information vector of the trade company is constructed, is finally carried out using merchant information vector of the preset audit model to the trade company
It calculates, obtains the auditing result of the trade company.Since the audit model have passed through a large amount of sample training in advance, pass through machine
Study has obtained general audit mode from history trade company audit logging, in the case where intervening without auditor, both may be used
Automatically trade company is audited, substantially increases review efficiency, and the personal experience without relying on auditor, ensure that audit
As a result just objective.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Corresponding to a kind of trade company checking method described in foregoing embodiments, Fig. 3 has gone out one kind provided in an embodiment of the present invention
One embodiment structure chart of trade company's audit device.
In the present embodiment, a kind of trade company audit device may include:
Trade company's audit application receiving module 301, trade company's audit application that the terminal device for receiving trade company is sent, and from
The merchant identification of the trade company is extracted in trade company's audit application；
Merchant information obtains module 302, for being obtained from each preset data source respectively according to the merchant identification
Merchant information of the trade company in each preset audit dimension, wherein each data source corresponds at least one audit dimension
Merchant information on degree；
Merchant information vector constructing module 303, for constructing the trade company according to the merchant information in each audit dimension
Merchant information vector；
Audit model computation module 304, for use preset audit model to the merchant information vector of the trade company into
Row calculates, and obtains the auditing result of the trade company, and the audit model is the machine learning by the training of preset sample set
Model, include in the sample set auditing result be by positive sample and auditing result be unacceptable negative sample, often
A sample standard deviation corresponds to a history trade company audit logging in preset database.
Further, trade company's audit device can also include:
Sample set constructing module, for obtaining each history trade company audit logging from the database, and according to institute
State history trade company audit logging construction sample set, wherein any sample standard deviation in the sample set include merchant information to
Amount and auditing result；
Model construction module is audited, for constructing audit model as follows:
Wherein, WtEmdnThe weight coefficient of the merchant information in dimension is audited for the dn, 1≤dn≤DN, DN are audit dimension
The sum of degree, InfoEmdnThe merchant information audited in dimension for the dn, Result are the auditing result of model output；
Sample training module, for being trained using the sample set to the audit model, until error amount is small
Until preset error threshold.
Further, the sample training module may include:
Sample training unit, for carrying out the tn times sample training to the audit model according to the following formula:
Wherein, sn is the sample serial number in the sample set, and 1≤sn≤SN, SN are the sample in the sample set
Sum, SampleVecsnFor the merchant information vector of the sn sample, and SampleVecsn=(SpEmsn,1,SpEmsn,2,...,
SpEmsn,dn,...,SpEmsn,DN), SpEmsn,dnTaking in dimension is audited at the dn for the merchant information vector of the sn sample
Value, tn are the number serial number of sample training, WeightVectnFor the weight vectors of the tn times sample training, and WeightVectn
=(WtEmtn,1,WtEmtn,2,...,WtEmtn,dn,...,WtEmtn,DN), WtEmtn,dnFor WeightVectnIt is audited at the dn
Value in dimension, T are transposition symbol, EsResultVectnFor the result vector of the tn times sample training, and
EsResultVectn=(EsValtn,1,EsValtn,2,...,EsValtn,sn,...,EsValtn,SN),
Error vector computing unit, for calculating the error vector of the tn times sample training according to the following formula:
Errortn=EsResultVectn-SpResultVec
Wherein, SpResultVec is objective result vector, and SpResultVec=(SpVal1,SpVal2,...,
SpValsn,...,SpValSN), SpValsnFor the auditing result of the sn sample, ErrortnFor the error of the tn times sample training
Vector, and Errortn=(ErrEmtn,1,ErrEmtn,2,...,ErrEmtn,sn,...,ErrEmtn,SN), ErrEmtn,sn=
SpValsn-EsValtn,sn；
Error amount computing unit, for calculating the error amount of the tn times sample training according to the following formula:
Wherein, ErrValtnThe error amount of as the tn times sample training；
Weight vectors updating unit, if the error amount for the tn times sample training is greater than or equal to the error threshold,
Then the weight vectors are updated；
Terminate training unit, if the error amount for the tn times sample training is less than the error threshold, terminates to institute
State the training of audit model.
Further, the weight vectors updating unit is specifically used for according to the following formula being updated the weight vectors:
WeightVectn+1 T=WeightVectn T-λ×SpMatrixT×Errortn T
Wherein, λ is preset update coefficient,
Further, the merchant information acquisition module may include:
Data grant request transmitting unit, for sending data grant request to the terminal device of the trade company；
Authorization message receiving unit, the authorization message for the trade company that the terminal device for receiving the trade company is fed back；
Data server selection unit, for choosing clothes corresponding with s-th of data source from preset server list
Device be engaged in as data server, the server list has recorded the corresponding relationship between each data source and each server, and 1
≤ s≤SourceNum, SourceNum are the sum of data source；
Request of data transmission unit is used to send request of data to the data server, includes in the request of data
The merchant identification and authorization message of the trade company；
Merchant information receiving unit, for receiving the merchant information for the trade company that the data server is sent.
It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description,
The specific work process of module and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
The schematic block diagram that Fig. 4 shows a kind of server provided in an embodiment of the present invention illustrates only for ease of description
Part related to the embodiment of the present invention.
In the present embodiment, the server 4 may include: processor 40, memory 41 and be stored in the storage
In device 41 and the computer-readable instruction 42 that can run on the processor 40, such as execute above-mentioned trade company's checking method
Computer-readable instruction.The processor 40 realizes above-mentioned each trade company's checking method when executing the computer-readable instruction 42
Step in embodiment, such as step S101 to S104 shown in FIG. 1.Alternatively, the processor 40 execute the computer can
The function of each module/unit in above-mentioned each Installation practice, such as the function of module 301 to 304 shown in Fig. 3 are realized when reading instruction 42
Energy.
Illustratively, the computer-readable instruction 42 can be divided into one or more module/units, one
Or multiple module/units are stored in the memory 41, and are executed by the processor 40, to complete the present invention.Institute
Stating one or more module/units can be the series of computation machine readable instruction section that can complete specific function, the instruction segment
For describing implementation procedure of the computer-readable instruction 42 in the server 4.
The processor 40 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 41 can be the internal storage unit of the server 4, such as the hard disk or memory of server 4.
The memory 41 is also possible to the External memory equipment of the server 4, such as the plug-in type being equipped on the server 4 is hard
Disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card, flash card
(Flash Card) etc..Further, the memory 41 can also both include the internal storage unit of the server 4 or wrap
Include External memory equipment.The memory 41 is for storing needed for the computer-readable instruction and the server 4 it
Its instruction and data.The memory 41 can be also used for temporarily storing the data that has exported or will export.
The functional units in various embodiments of the present invention may be integrated into one processing unit, is also possible to each
Unit physically exists alone, and can also be integrated in one unit with two or more units.Above-mentioned integrated unit both may be used
To use formal implementation of hardware, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Person says that all or part of the part that contributes to existing technology or the technical solution can body in the form of software products
Reveal and, which is stored in a storage medium, including several computer-readable instructions are used so that one
Platform computer equipment (can be personal computer, server or the network equipment etc.) executes described in each embodiment of the present invention
The all or part of the steps of method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-
Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can be with
Store the medium of computer-readable instruction.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations；Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of trade company's checking method characterized by comprising
Trade company's audit application that the terminal device of trade company is sent is received, and extracts the trade company from trade company audit application
Merchant identification；
The trade company is obtained from each preset data source respectively according to the merchant identification in each preset audit dimension
On merchant information, wherein each data source correspond at least one audit dimension on merchant information；
The merchant information vector of the trade company is constructed according to the merchant information in each audit dimension；
The merchant information vector of the trade company is calculated using preset audit model, obtains the audit knot of the trade company
Fruit, the audit model are the machine learning model trained by preset sample set, include audit in the sample set
As a result for by positive sample and auditing result be unacceptable negative sample, each sample standard deviation corresponds in preset database
A history trade company audit logging.
2. trade company's checking method according to claim 1, which is characterized in that it is described audit model building process include:
Each history trade company audit logging is obtained from the database, and sample is constructed according to history trade company audit logging
Set, wherein any sample standard deviation in the sample set includes merchant information vector sum auditing result；
Construct audit model as follows:
Wherein, WtEmdnThe weight coefficient of the merchant information in dimension is audited for the dn, 1≤dn≤DN, DN are audit dimension
Sum, InfoEmdnThe merchant information audited in dimension for the dn, Result are the auditing result of model output；
The audit model is trained using the sample set, until error amount is less than preset error threshold.
3. trade company's checking method according to claim 2, which is characterized in that described to be examined using the sample set described
Nuclear model, which is trained, includes:
The tn times sample training is carried out to the audit model according to the following formula:
Wherein, sn is the sample serial number in the sample set, and 1≤sn≤SN, SN are the total sample number in the sample set,
SampleVecsnFor the merchant information vector of the sn sample, and SampleVecsn=(SpEmsn,1,SpEmsn,2,...,
SpEmsn,dn,...,SpEmsn,DN), SpEmsn,dnTaking in dimension is audited at the dn for the merchant information vector of the sn sample
Value, tn are the number serial number of sample training, WeightVectnFor the weight vectors of the tn times sample training, and WeightVectn
=(WtEmtn,1,WtEmtn,2,...,WtEmtn,dn,...,WtEmtn,DN), WtEmtn,dnFor WeightVectnIt is audited at the dn
Value in dimension, T are transposition symbol, EsResultVectnFor the result vector of the tn times sample training, and
EsResultVectn=(EsValtn,1,EsValtn,2,...,EsValtn,sn,...,EsValtn,SN),
The error vector of the tn times sample training is calculated according to the following formula:
Errortn=EsResultVectn-SpResultVec
Wherein, SpResultVec is objective result vector, and SpResultVec=(SpVal1,SpVal2,...,
SpValsn,...,SpValSN), SpValsnFor the auditing result of the sn sample, ErrortnFor the error of the tn times sample training
Vector, and Errortn=(ErrEmtn,1,ErrEmtn,2,...,ErrEmtn,sn,...,ErrEmtn,SN), ErrEmtn,sn=
SpValsn-EsValtn,sn；
The error amount of the tn times sample training is calculated according to the following formula:
Wherein, ErrValtnThe error amount of as the tn times sample training；
If the error amount of the tn times sample training is greater than or equal to the error threshold, the weight vectors are updated,
And the tn+1 times sample training is carried out to the audit model；
If the error amount of the tn times sample training is less than the error threshold, terminate the training to the audit model.
4. trade company's checking method according to claim 3, which is characterized in that described to be updated packet to the weight vectors
It includes:
The weight vectors are updated according to the following formula:
WeightVectn+1 T=WeightVectn T-λ×SpMatrixT×Errortn T
Wherein, λ is preset update coefficient,
5. trade company's checking method according to any one of claim 1 to 4, which is characterized in that described according to the trade company
Mark respectively from obtained in each preset data source the trade company it is each it is preset audit dimension on merchant information include:
Data grant request is sent to the terminal device of the trade company；
Receive the authorization message of the trade company of the terminal device feedback of the trade company；
Server corresponding with s-th of data source is chosen from preset server list as data server, the service
Device the list records corresponding relationship between each data source and each server, 1≤s≤SourceNum, SourceNum are
The sum of data source；
Request of data is sent to the data server, includes the merchant identification and authorization letter of the trade company in the request of data
Breath；
Receive the merchant information for the trade company that the data server is sent.
6. a kind of trade company audits device characterized by comprising
Trade company's audit application receiving module, trade company's audit application that the terminal device for receiving trade company is sent, and from the quotient
The merchant identification of the trade company is extracted in the audit application of family；
Merchant information obtains module, for obtaining the trade company from each preset data source respectively according to the merchant identification
Merchant information in each preset audit dimension, wherein each data source corresponds to the quotient at least one audit dimension
Family information；
Merchant information vector constructing module, the trade company for constructing the trade company according to the merchant information in each audit dimension believe
Cease vector；
Model computation module is audited, for being calculated using preset audit model the merchant information vector of the trade company,
The auditing result of the trade company is obtained, the audit model is by the machine learning model of preset sample set training, institute
State in sample set include auditing result be by positive sample and auditing result be unacceptable negative sample, each sample standard deviation
Corresponding to a history trade company audit logging in preset database.
7. trade company according to claim 6 audits device, which is characterized in that further include:
Sample set constructing module for obtaining each history trade company audit logging from the database, and is gone through according to described
History trade company audit logging constructs sample set, wherein any sample standard deviation in the sample set includes merchant information vector sum
Auditing result；
Model construction module is audited, for constructing audit model as follows:
Wherein, WtEmdnThe weight coefficient of the merchant information in dimension is audited for the dn, 1≤dn≤DN, DN are audit dimension
Sum, InfoEmdnThe merchant information audited in dimension for the dn, Result are the auditing result of model output；
Sample training module, for being trained using the sample set to the audit model, until error amount is less than in advance
If error threshold until.
8. trade company according to claim 7 audits device, which is characterized in that the sample training module includes:
Sample training unit, for carrying out the tn times sample training to the audit model according to the following formula:
Wherein, sn is the sample serial number in the sample set, and 1≤sn≤SN, SN are the total sample number in the sample set,
SampleVecsnFor the merchant information vector of the sn sample, and SampleVecsn=(SpEmsn,1,SpEmsn,2,...,
SpEmsn,dn,...,SpEmsn,DN), SpEmsn,dnTaking in dimension is audited at the dn for the merchant information vector of the sn sample
Value, tn are the number serial number of sample training, WeightVectnFor the weight vectors of the tn times sample training, and WeightVectn
=(WtEmtn,1,WtEmtn,2..., WtEmtn,dn,...,WtEmtn,DN), WtEmtn,dnFor WeightVectnIt is audited at the dn
Value in dimension, T are transposition symbol, EsResultVectnFor the result vector of the tn times sample training, and
EsResultVectn=(EsValtn,1,EsValtn,2,...,EsValtn,sn,...,EsValtn,SN),
Error vector computing unit, for calculating the error vector of the tn times sample training according to the following formula:
Errortn=EsResultVectn-SpResultVec
Wherein, SpResultVec is objective result vector, and SpResultVec=(SpVal1,SpVal2,...,
SpValsn,...,SpValSN), SpValsnFor the auditing result of the sn sample, ErrortnFor the error of the tn times sample training
Vector, and Errortn=(ErrEmtn,1,ErrEmtn,2,...,ErrEmtn,sn,...,ErrEmtn,SN), ErrEmtn,sn=
SpValsn-EsValtn,sn；
Error amount computing unit, for calculating the error amount of the tn times sample training according to the following formula:
Wherein, ErrValtnThe error amount of as the tn times sample training；
Weight vectors updating unit, it is right if the error amount for the tn times sample training is greater than or equal to the error threshold
The weight vectors are updated；
Terminate training unit, if the error amount for the tn times sample training is less than the error threshold, terminates to examine described
The training of nuclear model.
9. a kind of computer readable storage medium, the computer-readable recording medium storage has computer-readable instruction, special
Sign is, realizes that the trade company as described in any one of claims 1 to 5 examines when the computer-readable instruction is executed by processor
The step of kernel method.
10. a kind of server, including memory, processor and storage can transport in the memory and on the processor
Capable computer-readable instruction, which is characterized in that realized when the processor executes the computer-readable instruction as right is wanted
Described in asking any one of 1 to 5 the step of trade company's checking method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910041444.2A CN109858927A (en) | 2019-01-16 | 2019-01-16 | A kind of trade company's checking method, device, computer readable storage medium and server |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910041444.2A CN109858927A (en) | 2019-01-16 | 2019-01-16 | A kind of trade company's checking method, device, computer readable storage medium and server |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109858927A true CN109858927A (en) | 2019-06-07 |
Family
ID=66894929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910041444.2A Pending CN109858927A (en) | 2019-01-16 | 2019-01-16 | A kind of trade company's checking method, device, computer readable storage medium and server |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109858927A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766340A (en) * | 2019-10-31 | 2020-02-07 | 支付宝(杭州)信息技术有限公司 | Business auditing method, device and equipment |
| CN111178203A (en) * | 2019-12-20 | 2020-05-19 | 江苏常熟农村商业银行股份有限公司 | Signature verification method and device, computer equipment and storage medium |
| CN111586448A (en) * | 2020-01-20 | 2020-08-25 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform anchor pre-screening method and device and storage medium |
| CN113313279A (en) * | 2020-02-27 | 2021-08-27 | 北京沃东天骏信息技术有限公司 | Document auditing method and device |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104077700A (en) * | 2014-06-09 | 2014-10-01 | 中国建设银行股份有限公司 | Task processing method, device and system used for e-business platform |
| CN104077701A (en) * | 2014-06-09 | 2014-10-01 | 中国建设银行股份有限公司 | Task processing method and device used for e-business platform |
| US20150161620A1 (en) * | 2013-12-06 | 2015-06-11 | Cube, Co. | System and method for risk and fraud mitigation for merchant on-boarding |
| WO2017028630A1 (en) * | 2015-08-20 | 2017-02-23 | 中国银联股份有限公司 | Verification method and server |
| CN107123010A (en) * | 2016-08-26 | 2017-09-01 | 北京小度信息科技有限公司 | Trade company's screening technique, the method and relevant apparatus for setting up trade company's evaluation model |
| CN107622432A (en) * | 2017-07-28 | 2018-01-23 | 阿里巴巴集团控股有限公司 | Trade company's evaluation method and system |
| CN108090817A (en) * | 2018-01-02 | 2018-05-29 | 吉林师范大学 | Sincere platform and its application method under background of travelling |
| CN108509458A (en) * | 2017-02-28 | 2018-09-07 | 阿里巴巴集团控股有限公司 | A kind of business object recognition methods and device |
| CN108564386A (en) * | 2018-04-28 | 2018-09-21 | 腾讯科技（深圳）有限公司 | Trade company's recognition methods and device, computer equipment and storage medium |
| CN108647977A (en) * | 2018-04-25 | 2018-10-12 | 合肥上城信息技术有限公司 | A kind of trade company audit evaluation and test platform and its method |
| CN108665364A (en) * | 2018-05-11 | 2018-10-16 | 深圳市买买提信息科技有限公司 | Sell bill of lading checking method and terminal device |
| CN108960782A (en) * | 2018-07-10 | 2018-12-07 | 北京木瓜移动科技股份有限公司 | content auditing method and device |
-
2019
- 2019-01-16 CN CN201910041444.2A patent/CN109858927A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150161620A1 (en) * | 2013-12-06 | 2015-06-11 | Cube, Co. | System and method for risk and fraud mitigation for merchant on-boarding |
| CN104077700A (en) * | 2014-06-09 | 2014-10-01 | 中国建设银行股份有限公司 | Task processing method, device and system used for e-business platform |
| CN104077701A (en) * | 2014-06-09 | 2014-10-01 | 中国建设银行股份有限公司 | Task processing method and device used for e-business platform |
| WO2017028630A1 (en) * | 2015-08-20 | 2017-02-23 | 中国银联股份有限公司 | Verification method and server |
| CN107123010A (en) * | 2016-08-26 | 2017-09-01 | 北京小度信息科技有限公司 | Trade company's screening technique, the method and relevant apparatus for setting up trade company's evaluation model |
| CN108509458A (en) * | 2017-02-28 | 2018-09-07 | 阿里巴巴集团控股有限公司 | A kind of business object recognition methods and device |
| CN107622432A (en) * | 2017-07-28 | 2018-01-23 | 阿里巴巴集团控股有限公司 | Trade company's evaluation method and system |
| CN108090817A (en) * | 2018-01-02 | 2018-05-29 | 吉林师范大学 | Sincere platform and its application method under background of travelling |
| CN108647977A (en) * | 2018-04-25 | 2018-10-12 | 合肥上城信息技术有限公司 | A kind of trade company audit evaluation and test platform and its method |
| CN108564386A (en) * | 2018-04-28 | 2018-09-21 | 腾讯科技（深圳）有限公司 | Trade company's recognition methods and device, computer equipment and storage medium |
| CN108665364A (en) * | 2018-05-11 | 2018-10-16 | 深圳市买买提信息科技有限公司 | Sell bill of lading checking method and terminal device |
| CN108960782A (en) * | 2018-07-10 | 2018-12-07 | 北京木瓜移动科技股份有限公司 | content auditing method and device |
Non-Patent Citations (1)
| Title |
|---|
| 孙权;赵金涛;: "基于数据挖掘的商户风险评分方法和系统", 软件产业与工程, no. 01 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766340A (en) * | 2019-10-31 | 2020-02-07 | 支付宝(杭州)信息技术有限公司 | Business auditing method, device and equipment |
| CN111178203A (en) * | 2019-12-20 | 2020-05-19 | 江苏常熟农村商业银行股份有限公司 | Signature verification method and device, computer equipment and storage medium |
| CN111586448A (en) * | 2020-01-20 | 2020-08-25 | 全息空间(深圳)智能科技有限公司 | Live broadcast platform anchor pre-screening method and device and storage medium |
| CN113313279A (en) * | 2020-02-27 | 2021-08-27 | 北京沃东天骏信息技术有限公司 | Document auditing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109858927A (en) | A kind of trade company's checking method, device, computer readable storage medium and server | |
| CN108665159A (en) | A kind of methods of risk assessment, device, terminal device and storage medium | |
| CN108763277B (en) | Data analysis method, computer readable storage medium and terminal device | |
| CN111181757B (en) | Information security risk prediction method and device, computing equipment and storage medium | |
| CN112712429A (en) | Remittance service auditing method, remittance service auditing device, computer equipment and storage medium | |
| CN110610431A (en) | Intelligent claim settlement method and intelligent claim settlement system based on big data | |
| CN111428217A (en) | Method and device for identifying cheat group, electronic equipment and computer readable storage medium | |
| CN112702410B (en) | Evaluation system, method and related equipment based on blockchain network | |
| CN110796539A (en) | Credit investigation evaluation method and device | |
| CN110852809A (en) | Data processing method, device, equipment and medium | |
| CN111209403A (en) | Data processing method, device, medium and electronic equipment | |
| CN116596277A (en) | Task planning method, device, equipment and storage medium | |
| CN112446777A (en) | Credit evaluation method, device, equipment and storage medium | |
| CN115578027A (en) | Data quality evaluation method and device, electronic equipment and storage medium | |
| CN113935738B (en) | Transaction data processing method, device, storage medium and equipment | |
| CN109379334B (en) | Network security risk assessment index weight self-adaptive construction method and device | |
| CN109726882A (en) | The method and apparatus that a kind of pair of object is evaluated | |
| CN115689567A (en) | Risk control method, apparatus, device, medium, and program product | |
| CN114298806A (en) | Risk identification method, apparatus, storage medium, and program product | |
| CN117408772A (en) | Financial product recommendation method, apparatus, computer device and storage medium | |
| CN113052692A (en) | Data processing method and device, electronic equipment and computer readable storage medium | |
| CN115409636A (en) | Product risk prediction method, device, equipment and medium | |
| CN116861398A (en) | Authentication processing method, authentication processing device, computer equipment and storage medium | |
| CN117875606A (en) | Method, device, equipment, storage medium and program product for processing resource bill | |
| CN117541193A (en) | Business auditing method, device, computer equipment and storage medium |
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 |