CN110472827A - System strategy method for early warning, device, server and readable storage medium storing program for executing - Google Patents
System strategy method for early warning, device, server and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN110472827A CN110472827A CN201910617727.7A CN201910617727A CN110472827A CN 110472827 A CN110472827 A CN 110472827A CN 201910617727 A CN201910617727 A CN 201910617727A CN 110472827 A CN110472827 A CN 110472827A
- Authority
- CN
- China
- Prior art keywords
- grouping
- index value
- sample
- target signature
- sample size
- 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
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 238000013139 quantization Methods 0.000 claims abstract description 148
- 238000012545 processing Methods 0.000 claims abstract description 57
- 238000011002 quantification Methods 0.000 claims abstract description 45
- 230000002159 abnormal effect Effects 0.000 claims abstract description 31
- 230000008859 change Effects 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 230000008901 benefit Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 description 13
- 230000000694 effects Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 6
- 230000007423 decrease Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 208000001613 Gambling Diseases 0.000 description 4
- 230000002354 daily effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
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
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/04—Payment circuits
- G06Q20/06—Private payment circuits, e.g. involving electronic currency used among participants of a common payment scheme
- G06Q20/065—Private payment circuits, e.g. involving electronic currency used among participants of a common payment scheme using e-cash
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention discloses a kind of system strategy method for early warning, device, server and readable storage medium storing program for executing, the method is applied to goal systems, the goal systems is configured with more than one system strategy, the system strategy is configured based on corresponding target signature, it include: that the multiple accounts for including construct sample set based on the goal systems, it include the adventure account as black sample, and the credible account as white sample in the sample set；Based on the corresponding target signature of the system strategy, characteristic quantification processing is carried out to the sample set, obtains the quantization index value of the target signature；It determines whether the quantization index value meets abnormal quantization indicator conditions, if the quantization index value meets the abnormal quantization indicator conditions, warning information is generated and push, to carry out early warning to the corresponding system strategy of the target signature.It in above scheme, is monitored by the quantization index value to target signature, effectively increases the efficiency of system strategy early warning.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of system strategy method for early warning, device, servers and can
Read storage medium.
Background technique
With the continuous development of science and technology, many business functions can be realized automatically by system, for example, passing through
Transaction system realizes online transaction commodity, is transferred accounts function etc. by fund transfer system realization.It is to be treated for system
Data volume be it is huge, it is also considerably complicated for handling the process of data.Therefore, in order to realizing the complex process to data
And reply system is possible to the risk occurred, can configure corresponding system strategy.
Due to system data and system business be not it is unalterable, can change with the difference of demand,
Original system strategy is possible to the system data after can not being suitable for changing and business.In the prior art, by manually carrying out
Periodic system strategy updates, but since the update cycle is longer, not can guarantee strategy of the system strategy in non-reproducting periods
Effect.
Summary of the invention
This specification embodiment provides and a kind of system strategy method for early warning, device, server and readable storage medium storing program for executing.
In a first aspect, this specification embodiment provides a kind of system strategy method for early warning, it is applied to goal systems, the mesh
Mark system is configured with more than one system strategy, and the system strategy is configured based on corresponding target signature, the method packet
It includes:
The multiple accounts for including based on the goal systems construct sample set, include as black in the sample set
The adventure account of sample, and the credible account as white sample；
Based on the corresponding target signature of the system strategy, characteristic quantification processing is carried out to the sample set, obtains institute
State the quantization index value of target signature；
Determine whether the quantization index value meets abnormal quantization indicator conditions, if the quantization index value meet it is described different
Constant indicator conditions then generate and push warning information, to carry out early warning to the corresponding system strategy of the target signature.
Second aspect, this specification embodiment provide a kind of system strategy prior-warning device, are applied to goal systems, the mesh
Mark system is configured with more than one system strategy, and the system strategy is configured based on corresponding target signature, described device packet
It includes:
Sample constructs module, and multiple accounts for including based on the goal systems construct sample set, the sample
It include the adventure account as black sample in set, and as the credible account of white sample；
Quantification treatment module carries out the sample set special for being based on the corresponding target signature of the system strategy
Quantification treatment is levied, the quantization index value of the target signature is obtained；
Warning module, for determining whether the quantization index value meets abnormal quantization indicator conditions, if the quantization refers to
Scale value meets the abnormal quantization indicator conditions, then warning information is generated and push, to the corresponding system of the target signature
Strategy carries out early warning.
The third aspect, this specification embodiment provide a kind of server, including memory, processor and are stored in memory
The step of computer program that is upper and can running on a processor, the processor executes any of the above-described the method.
Fourth aspect, this specification embodiment provide a kind of computer readable storage medium, are stored thereon with computer journey
Sequence, when which is executed by processor the step of realization any of the above-described the method.
This specification embodiment has the beneficial effect that:
In the system strategy method for early warning that this specification embodiment provides, multiple account structures for including based on goal systems
Build sample set；Based on the corresponding target signature of system strategy, characteristic quantification processing is carried out to sample set, obtains target signature
Quantization index value；If quantization index value meets abnormal quantization indicator conditions, warning information is generated and pushes, to target spy
It levies corresponding system strategy and carries out early warning.In above scheme, quantified by the target signature to system strategy, and further
Quantify indicator conditions by abnormal the quantization index value of target signature be monitored, monitoring abnormal quantizating index
When value, show to include that the effect of the system strategy of the target signature is possible to will appear decline, at this point, generating warning information simultaneously
Warning information is pushed, system for prompting administrator can be effectively improved in time there are the system strategy of effect decline risk
The efficiency of system strategy early warning.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 is a kind of flow chart for system strategy method for early warning that this specification embodiment first aspect provides；
Fig. 2 is the schematic diagram for the system strategy prior-warning device that this specification embodiment second aspect provides；
Fig. 3 is the schematic diagram for the server that this specification embodiment third aspect provides.
Specific embodiment
In order to better understand the above technical scheme, below by attached drawing and specific embodiment to this specification embodiment
Technical solution be described in detail, it should be understood that the specific features in this specification embodiment and embodiment are to this explanation
The detailed description of book embodiment technical solution, rather than the restriction to this specification technical solution, in the absence of conflict,
Technical characteristic in this specification embodiment and embodiment can be combined with each other.
In a first aspect, this specification embodiment provides a kind of system strategy method for early warning, this method is applied to goal systems
In, goal systems is configured with more than one system strategy.As shown in Figure 1, the system strategy provided for this specification embodiment
The flow chart of method for early warning, method includes the following steps:
Step S11: the multiple accounts for including based on the goal systems construct sample set, include in the sample set
There are the adventure account as black sample, and the credible account as white sample；
Step S12: being based on the corresponding target signature of the system strategy, carries out at characteristic quantification to the sample set
Reason, obtains the quantization index value of the target signature；
Step S13: determining whether the quantization index value meets abnormal quantization indicator conditions, if the quantization index value is full
The foot abnormal quantization indicator conditions, then generate and push warning information, with to the corresponding system strategy of the target signature into
Row early warning.
Scheme in this specification embodiment can be applied in the system of multiple fields, for example, goal systems can be
Online trading system, electronic fare payment system of Third-party payment platform of electric business platform etc..For goal systems, Ke Yipei
It is equipped with more than one system strategy, such as risk identification strategy, system access strategy, migration efficiency etc..
It should be noted that system strategy is made of feature, i.e., system strategy is configured based on feature, citing
For, risk identification strategy depends on system risk data characteristics, and access strategy is dependent on system access data characteristics etc..Work as number
When being widely varied according to feature, system strategy possibly can not adapt to the data characteristics after variation, so as to cause system plan
Slightly fail.In this specification embodiment, the quantization index value of the target signature of system strategy is influenced by calculating, and is being detected
When quantization index value exception, warning information is generated, and warning information is pushed to system manager, to inform system administration in time
Member's target signature occurs abnormal, it is possible to cause the effect of system strategy relevant to target signature to fail, so that system
Administrator can adjust system strategy in time, and system strategy failure bring is avoided to influence.
In this specification embodiment, firstly, executing step S11: the multiple accounts building for including based on the goal systems
Sample set includes the adventure account as black sample, and the credible account as white sample in the sample set.
In the specific implementation process, it when user completes the registration operation in goal systems, is just created in goal systems
The account of the user.Account can wrap containing corresponding account information, such as registration information, account risk information etc..With target
System is for electronic fare payment system, account information may include essential information, credit information, Transaction Information of user etc..
According to account information, the account of goal systems can be divided into black sample and white sample, for example, according to for use
The calling information at family and/or the historical transactional information of user, it can be determined that the account of the user whether there is risk.In a reality
It applies in example, it, will be this when the complaint quantity for user, which is greater than, includes risk trade in a threshold value and/or historical trading
Account is determined as the black sample that account information is risk information.Do not have when user was not connected in complaint and/or historical trading
When crossing risk trade, this account is determined as the white sample that account information is reliable information.It is, of course, also possible to pass through other accounts
Family information determines black sample and white sample, such as the qualification information of user, authentication information, here without limitation.
When constructing sample set, can be constructed based on the account for including in goal systems in preset time period, for example,
Based on the account of same day goal systems, adventure account and credible account to same day goal systems sample respectively, building
Sample set.Sample set can also be constructed based on the whole accounts for including in goal systems, be also based on other modes
Construct sample set, here without limitation.
Next, executing step S12: being based on the corresponding target signature of the system strategy, carried out to the sample set
Characteristic quantification processing, obtains the quantization index value of the target signature.
In this specification embodiment, goal systems is configured with more than one system strategy, and a system strategy is come
It says, target signature is to influence the one or more features of the system strategy.Feature due to influencing each system strategy may phase
Together may also be different, therefore, target signature generally includes multiple features.For example, target signature may include account
Attributive character (such as age characteristics, sex character etc.), transaction feature (such as transaction amount feature, trading frequency feature etc.)
Deng.
It should be noted that the quantizating index of target signature can be set according to actual needs, quantizating index can
One or more is thought, for example, quantizating index may include: evidence weight (Weight of Evidence, WOE), information valence
It is worth (Information Value, IV), the order of magnitude.Certainly, quantizating index can also be other indexs, here without limitation.
For every kind of quantizating index, corresponding preset characteristic quantification method can be provided with, according to preset characteristic quantification method
Sample set is handled, available corresponding quantization index value.
Based on target signature, characteristic quantification processing is carried out to sample set, can be and sample set is integrally handled,
It is also possible to be grouped sample set processing, here without limitation.For example, it may be determined that each sample in sample set
The characteristic value of target signature, then according to preset characteristic quantification method to the characteristic value of the target signature of each sample at
Reason, obtains quantization index value.Processing first can also be grouped to sample set according to target signature, then according to preset spy
Quantization method is levied to the sample set after packet transaction, obtains quantization index value.It is, of course, also possible to select it according to actual needs
His mode, is just not listed one by one here.
Finally, executing step S13: determining whether the quantization index value meets abnormal quantization indicator conditions, if the amount
Change index value and meet the abnormal quantization indicator conditions, then warning information is generated and push, with corresponding to the target signature
System strategy carries out early warning.
In the specific implementation process, abnormal quantization indicator conditions can be set according to actual needs, such as abnormal amount
Changing indicator conditions can be preset quantization index value range, can also be quantization index value carrying out further processing processing
Afterwards, determine processing result whether Yi Chang preset condition, here without limitation.When quantization index value meets abnormal quantizating index item
When part, show that the corresponding target signature of quantization index value is changed, at this point, system strategy relevant to the target signature has
It is possible that the risk of effect decline, therefore, the scheme in this specification embodiment can meet abnormal amount in quantization index value
When changing indicator conditions, warning information is generated, and warning information is pushed to system manager, to notify that system manager's target is special
Sign changes, if needs to be adjusted relevant system strategy, has reached the effect for carrying out early warning to system strategy in time
Fruit.
In above scheme, quantified by the target signature to system strategy, and quantifies indicator conditions by abnormal
The quantization index value of target signature is monitored, when monitoring abnormal quantization index value, shows to include the target
The effect of the system strategy of feature is possible to will appear decline, at this point, generating warning information and pushing warning information, energy
Enough timely system for prompting administrators effectively increase the efficiency of system strategy early warning there are the system strategy of effect decline risk.
In addition, the scheme in this specification embodiment can quickly orient abnormal target signature, without manually to system strategy into
Row off-line analysis shortens the time of Developing Tactics to find problem characteristic.
In the specific implementation process, step S12 can be accomplished by the following way: the target signature is based on, to described
Sample set is grouped processing, obtains multiple groupings；Characteristic quantity is carried out to the sample in each grouping of the multiple grouping
Change processing, obtains the quantization index value of the target signature.
In this embodiment, the quantization index value of target signature is obtained by packet transaction.It is special for each target
Sign, processing is grouped to the sample set built, obtains multiple groupings, that is to say, that when the quantity of target signature is N
When, it needs to carry out n times packet transaction respectively to sample set, obtains N number of group result.It can be with to the packet transaction of sample set
It is accomplished in several ways, such as by branch mailbox processing, customized grouping etc., here without limitation.
For example, when the target signature of a system strategy includes age characteristics and day trade transaction amount of money feature,
Processing can be grouped to sample set according to the two features respectively.It, can be by sample for age characteristics
Set the age characteristics based on each sample be grouped, 20 years old sample below is such as divided into one group, by the age 20 years old~
40 years old samples are divided into one group, and the sample by the age at 40 years old~60 years old is divided into one group, the sample by the age at 60 years old~80 years old
It is divided into one group, in this way, the sample in sample set is divided into four groups based on the age.For day trade transaction amount of money feature
For, sample set can be grouped based on the day trade transaction amount of money of each sample, such as the day trade transaction amount of money is less than
5000 yuan of sample is divided into one group, and the sample by the day trade transaction amount of money more than or equal to 5000 yuan is divided into one group, passes through this side
Sample in sample set is based on the day trade transaction amount of money and is divided into two groups by formula.
It should be noted that, for each target signature, there is respective grouping when the quantity of target signature is multiple
It as a result, in the quantization index value for determining the target signature, is handled for the group result of the target signature.Still edge
With above-mentioned example, when target signature is above-mentioned age characteristics, corresponding group result is that sample set is divided into four
Grouping when calculating the quantization index value of age characteristics is handled the sample in this four groupings.Likewise, in mesh
When mark feature is above-mentioned day trade transaction amount of money feature, corresponding group result is that sample set is divided into two groupings, is being counted
It when calculating the quantization index value of day trade transaction amount of money feature, is handled the sample in the two groupings.
In the specific implementation process, by taking branch mailbox is handled as an example, it is described be based on the target signature, to the sample set into
Row packet transaction obtains multiple groupings, comprising: according to the sample characteristics that sample each in the sample set is included, determines
It out include the target signature in sample characteristics to branch mailbox sample；Described branch mailbox processing will be carried out to branch mailbox sample, obtained
Multiple branch mailbox, wherein the corresponding grouping of each branch mailbox.
In this specification embodiment, some target signatures can not cover all samples, i.e., some samples do not have certain
A little target signatures, for example, some users did not trade on the same day, then just this is special without the day trade transaction amount of money by these users
Sign.Therefore, in packet transaction, the sample with target signature can be first determined in sample set, will not have target signature
Sample filter out, processing further then is grouped to the sample with target signature.
Specifically, whether when being handled using branch mailbox to be grouped to sample set can first pass through has target special
Sign is to handle sample set, that is, using include target signature sample as to branch mailbox sample, then treat branch mailbox sample
Carry out branch mailbox processing.In this specification embodiment, branch mailbox processing can using wide branch mailbox, etc. the modes such as frequency divisions case, here not
It limits.
By taking equal frequency divisions case as an example, waiting frequency divisions case is the boundary value by selecting each branch mailbox section, so that each branch mailbox area
Interior includes roughly equal sample size.Such as to the quantity of branch mailbox sample be 100, can will to branch mailbox sample according to
Respective object feature value size is ranked up, and then carries out branch mailbox to sorted sample, such as will be to branch mailbox sample point
For 10 casees (corresponding 10 groupings), i.e., interval number is 10, then each section probably includes 10 samples.It had been embodied
Cheng Zhong can determine the boundary value in each section according to the distribution situation to the object feature value of each sample in branch mailbox sample
And interval number avoids the occurrence of empty van so as to both fall within the roughly the same sample of quantity in each section.
Further, after sample set is divided into multiple groupings, the amount of target signature can be obtained in the following manner
Change index value: each grouping is directed to, according to the black sample size and white sample size in the grouping, in the grouping
Sample carries out characteristic quantification processing, obtains the group quantization index value of the grouping；Referred to according to the group quantization of each grouping
Scale value and the sample size of each grouping, obtain the quantization index value of the target signature.
In the specific implementation process, after having obtained group result, for each grouping, the black sample of the grouping is counted
Then quantity and white sample size carry out characteristic quantification processing to the black sample size of each grouping and white sample size,
The group quantization index of each grouping is obtained, for ease of description, in this specification embodiment, using quantizating index as weight evidence
For weight, the order of magnitude, information value, to be illustrated.Further, according to the group quantization index value of each grouping and
The sample size of each grouping can carry out according to actual needs come the method for determining the quantization index value of final target signature
Selection, such as weighted sum, weighted average etc., the quantization in following example, by taking weighted sum as an example, to target signature
The calculation of index value is illustrated.
Specifically, when quantization index value is evidence weight index value, it is described to be directed to each grouping, according to the grouping
In black sample size and white sample size, in the grouping sample carry out characteristic quantification processing, obtain the grouping point
Group quantization index value, comprising: be directed to each grouping, it is black in the sample set to determine that the black sample size in the grouping accounts for
First ratio of sample size, and the white sample size that determines in the grouping account for the of white sample size in the sample set
Two ratios；Based on first ratio and second ratio, the grouping evidence weight index value of the grouping is determined；It is described
According to the group quantization index value of each grouping and the sample size of each grouping, the target signature is obtained
Quantization index value, comprising: the grouping evidence weight index value based on each grouping, to the sample number of each grouping
Amount is weighted processing, obtains the evidence weight index value of the target signature.
In the specific implementation process, the calculation method of evidence weight index value can refer to following formula:
Wherein, WOEiThe grouping evidence weight index value being grouped for i-th, pyiFor the black sample size in i-th of grouping
Account for the first ratio of all black sample sizes in sample set, pniSample set is accounted for for the white sample size in i-th of grouping
In all white sample sizes the second ratio, #yiFor the black sample size in i-th of grouping, #yTTo own in sample set
The quantity of black sample, #niFor the white sample size in i-th of grouping, #nTFor the quantity of white samples all in sample set.
Grouping evidence weight index value can symbolize target signature in the grouping, the differentiation of dialogue sample and black sample
Degree, grouping evidence weight index value is higher, and discrimination is more obvious, and shows that the target signature is more effective.Further, will each divide
The grouping evidence weight index value of group is weighted summation to the sample number for falling into each grouping, obtains final as weight
The evidence weight index value of target signature.
For example, when target signature is the day trade transaction amount of money, according to the numerical value of the day trade transaction amount of money of each sample,
Processing is grouped to sample, it is assumed that 3 groupings are obtained, first group of sample size is 10, and second group of sample size is 11,
The sample size of third group is 10.According to the formula of evidence weight index value calculated above, it is assumed that obtain first group of grouping card
It is WOE according to weight index value1, second group of grouping evidence weight index value is WOE2, the grouping evidence weight index value of third group
For WOE3, using each grouping evidence weight index value as the weight of each group sample size, it is weighted summation, is obtained final
Target signature, i.e. the evidence weight index value WOE of the day trade transaction amount of money, specifically: WOE=10 × WOE1+11×WOE2+10×
WOE3。
It is described to be directed to each grouping when quantization index value is order of magnitude index value, according to the black sample in the grouping
This quantity and white sample size carry out characteristic quantification processing to the sample in the grouping, and the group quantization for obtaining the grouping refers to
Scale value, comprising: be directed to each grouping, determine in the black sample size in the grouping and the grouping between white sample size
Third ratio, the packet count magnitude index value as the grouping；The group quantization index value according to each grouping, with
And the sample size of each grouping, obtain the quantization index value of the target signature, comprising: based on each grouping
Packet count magnitude index value is weighted processing to the sample size of each grouping, obtains the quantity of the target signature
Grade index value.
It should be noted that for various reasons, meeting the account magnitude meeting of the target signature for target signature
It mutates, and then performance of the target signature finally in corresponding strategies is caused to generate huge variation.For example, with electronics
For payment system, during New Year, user's abruptly increase of red packet is provided, causes the account order of magnitude of red packet transaction feature to occur prominent
Become.Therefore, can be by calculating the corresponding order of magnitude quantizating index of target signature in this specification embodiment, and monitor the number
Magnitude quantizating index to carry out early warning to system strategy in time.
Specifically, the calculation method of order of magnitude index value can refer to following formula:
Ci=#yi/#ni
Wherein, CiThe packet count magnitude index being grouped for i-th, #yiFor the black sample size in i-th of grouping, #niFor
White sample size in i-th of grouping.
Further, after the packet count magnitude index of each grouping has been determined, the packet count magnitude index of each group is made
For the weight of each group sample number, it is weighted summation, obtains the order of magnitude index value of final target signature.
The example that object above feature is the day trade transaction amount of money is still continued to use, after obtaining 3 groupings, according to calculated above
The formula of order of magnitude index value, it is assumed that the packet count magnitude index value for obtaining first group is C1, second group of packet count magnitude refers to
Scale value is C2, the packet count magnitude index value of third group is C3, handled by weighted sum, obtain final order of magnitude index value
C, specifically: C=10 × C1+11×C2+10×C3。
It is described to be directed to each grouping when quantization index value is information value index value, according to black in the grouping
Sample size and white sample size carry out characteristic quantification processing to the sample in the grouping, obtain the group quantization of the grouping
Index value, comprising: be directed to each grouping, determine that the black sample size in the grouping accounts for the black sample in the sample set
4th ratio of quantity, and determine that the white sample size in the grouping accounts for the 5th of the white sample size in the sample set
Ratio；Based on the difference of the 4th ratio, the 5th ratio and the 4th ratio and the 5th ratio, determine
The grouping information value index value of the grouping；The group quantization index value according to each grouping and described each
The sample size of grouping obtains the quantization index value of the target signature, comprising: the grouping information valence based on each grouping
It is worth index value, processing is weighted to the sample size of each grouping, obtains the information value index of the target signature
Value.
Value information is capable of the predictive ability of evaluation goal feature, when the order of magnitude or discrimination of feature are without too big
When changing, but combining larger to system strategy influential effect, value information can be used and calculate its synthesis differentiation effect.
In the specific implementation process, the calculation method of information value index value can be with reference to formula once:
Wherein, IViThe grouping value information index value being grouped for i-th, pyiIt is accounted for for the black sample size in i-th of grouping
The third ratio of all black sample sizes in sample set, pniIt is accounted in sample set for the white sample size in i-th of grouping
All white sample sizes the 4th ratio, #yiFor the black sample size in i-th of grouping, #yTIt is all black in sample set
The quantity of sample, #niFor the white sample size in i-th of grouping, #nTFor the quantity of white samples all in sample set.
Information value index value joined the black sample number accounting of grouping compared with evidence weight index value, in calculation formula
With the difference of white sample book accounting, the quantity for having embodied sample in target signature current group accounts for the ratio of entire sample size,
Influence to feature predictive ability.Further, using the grouping information value index value of each group as the weight of each group sample, into
Row weighted sum obtains the information value index value of final target signature.
The example that object above feature is the day trade transaction amount of money is still continued to use, after obtaining 3 groupings, according to calculated above
The formula of information value index value, it is assumed that the grouping information value index value for obtaining first group is IV1, second group of grouping information
Value index value is IV2, the grouping information value index value of third group is IV3, handled by weighted sum, obtain final number
Magnitude index value IV, specifically: IV=10 × IV1+11×IV2+10×IV3。
Certainly, in addition to above-mentioned three kinds of quantizating index, other quantizating index, such as black sample can also be selected according to actual needs
With the sample number ratio etc. of target signature in this, in the specific implementation process, can choose one of above-mentioned quantizating index or
More persons carry out the quantification treatment of target signature, here without limitation.
Further, after quantization index value has been determined, judge whether quantization index value meets abnormal quantizating index item
Part can specifically be accomplished by the following way: the history quantization index value based on the target signature determines the quantizating index
Fluctuating range relative to the history quantization index value；If the fluctuating range generates simultaneously in preset amplitude range
Push the warning information.
In this specification embodiment, history quantization index value can be whole quantization index values of history calculating, either
The average value for whole quantization index values that history calculates.In one embodiment, a target will be calculated at predetermined time intervals
The quantization index value of feature, for example, by taking target signature is day trade transaction number, quantizating index is evidence weight index as an example, base
Sample set, and the day trade transaction number to daily sample set based on each sample are determined in daily goal systems account
It is grouped, to calculate evidence weight index value.In this way, the evidence weight index value calculated daily is carried out in graph form
It indicates, obtains evidence weight index value versus time curve.It can be equal by each evidence weight index value on the curve
As a history quantization index value, the fluctuating range of the quantization index value on the same day is then calculated.Certainly, this specification embodiment
In, obtained target signature curve can also be visualized, enable system manager's master goal feature in real time
Variation.
The calculation of fluctuating range can be set according to actual needs, first calculate the same day as fluctuating range can be
Quantization index value and whole history quantization index value average value between difference, then calculating difference and whole history quantify to refer to
The ratio of the average value of scale value, using ratio as fluctuating range.Or the calculation of fluctuating range can refer to for same day quantization
Ratio between scale value and the maximum value of history quantization index value, or the ratio between the minimum value of history quantization index value
Deng here without limitation.
Further, after fluctuating range has been determined, fluctuating range is compared with preset amplitude range, example
Such as, amplitude range is that fluctuating range is greater than 30%, when obtained fluctuating range is greater than 30%, then it is assumed that the target signature occurs
It is widely varied, system strategy effect may fail, and generate warning information at this time and warning information is pushed to system pipes
Reason person.
Warning information may include changed target signature, can also include system plan corresponding with target signature
Slightly.All include since a target signature can correspond to multiple system strategies, i.e., in multiple system strategies target signature this
Feature, therefore, warning information may include multiple system strategies corresponding with target signature, these information are pushed to system pipes
Reason person, so that system manager performs corresponding processing.
After the quantization index value meets the abnormal quantization indicator conditions, the method also includes: according to described
Quantization index value is adjusted the corresponding system strategy of the target signature.
In this specification embodiment, system strategy is made of various features, and the application scenarios of system strategy are also
It is diversified, therefore, different strategies can be used when target signature changes based on different types of system strategy
Adjustment mode, such as the weight etc. of target signature in the target signature that replacement system strategy includes, or adjustment system strategy.
In the specific implementation process, still using goal systems as electronic fare payment system, system strategy is to judge gambling
For system strategy, the target signature in the system strategy may include red packet transaction feature, i.e., using red packet transaction feature as
Judge whether it is a feature of gambling.During New Year, user's abruptly increase of red packet is provided, red packet transaction feature is caused
Quantization index value mutates, then cannot be distinguished whether the user with red packet transaction feature is gambling user, therefore, this is special
Sign is invalid in the system strategy frame, at this point, the adjustment mode of the system strategy can be to remove red packet transaction spy
Sign changes the target signature of more effectively judgement gambling.
For another example the target in the system strategy is special when system strategy is the system strategy that guarantee red packet is normally provided
Sign may include red packet transaction feature, i.e., using red packet transaction feature as a feature for judging whether normally to provide red packet.
During New Year, user's abruptly increase of red packet is provided, the quantization index value of red packet transaction feature is caused to mutate, at this point, in order to
Guarantee the normal granting of red packet, system can smoothly execute the transmission of a large amount of red packets, not generating system Caton, at this point, system plan
Adjustment mode slightly can be weight of the adjustment red packet transaction feature in system strategy, such as increase weight, guarantee system strategy
Normal operation.
It should be understood that the adjustment mode of system strategy can be and pre-set, i.e., all for every kind of system strategy
A kind of Developing Tactics mode is pre-set, it, can be according to preset tune when the feature of system strategy is widely varied
Perfect square formula carries out Developing Tactics automatically.
In order to better understand the system strategy method for early warning of this specification embodiment offer, below to calculate target daily
For the quantization index value of feature, to be illustrated to the system strategy method for early warning in this specification embodiment.In the implementation
In example, system strategy is risk recognition system strategy, and target signature is day trade transaction stroke count feature, and quantization index value is weight evidence
Weight index value, order of magnitude index value and information value index value.
Firstly, being sampled respectively to the black sample (adventure account) on the goal systems same day and white sample (credible account), constitute
Sample set.Day trade transaction stroke count feature is put into sample set, by sample set according to the day trade transaction stroke count of each sample
Equal frequency divisions case is carried out, multiple branch mailbox are obtained.The black sample size and white sample number for including based on each branch mailbox in multiple branch mailbox
Amount, calculates evidence weight index value, order of magnitude index value and information value index value according to above formula respectively.By this
Three indexs are compared with respective history quantization index value respectively, when the fluctuating range of a certain item feature is greater than preset threshold
When, the administrator of downstream strategy can be reminded, so that administrator carries out Developing Tactics (such as replacement target strategy).Separately
Outside, for capableing of for the Policy model of Automatic Optimal, strategy can be carried out for target signature according to preset adjustment mode
Tuning (such as the weight of adjustment target signature in system strategy).
As it can be seen that the quantification monitoring of target signature can be carried out according to actual needs in this specification embodiment, it is such as every
Day understands feature used in system strategy in time whether there is decay, and be directly targeted to problem characteristic, be not necessarily to people
Work carries out off-line analysis to entire strategy.Administrator is reminded to carry out Developing Tactics when in addition, failing by feature abnormalities in time, or
Person passes through tactful automated tuning, and can greatly shorten system strategy optimizes and revises the time, improves efficiency.
Second aspect, based on the same inventive concept, this specification embodiment provide a kind of system strategy prior-warning device, application
In goal systems, the goal systems is configured with more than one system strategy, and it is special that the system strategy is based on corresponding target
Sign configuration, referring to FIG. 2, the device includes:
Sample constructs module 21, and multiple accounts for including based on the goal systems construct sample set, the sample
Include the adventure account as black sample, and the credible account as white sample in this set；
Quantification treatment module 22 carries out the sample set for being based on the corresponding target signature of the system strategy
Characteristic quantification processing, obtains the quantization index value of the target signature；
Warning module 23, for determining whether the quantization index value meets abnormal quantization indicator conditions, if the quantization
Index value meets the abnormal quantization indicator conditions, then warning information is generated and push, to the corresponding system of the target signature
System strategy carries out early warning.
In a kind of optional implementation, quantification treatment module 22 is used for:
Based on the target signature, processing is grouped to the sample set, obtains multiple groupings；
Characteristic quantification processing is carried out to the sample in each grouping of the multiple grouping respectively, obtains the target signature
Quantization index value.
In a kind of optional implementation, quantification treatment module 22 is used for:
According to the sample characteristics that sample each in the sample set is included, determine to include described in sample characteristics
Target signature to branch mailbox sample；
Described branch mailbox processing will be carried out to branch mailbox sample, and obtain multiple branch mailbox.
In a kind of optional implementation, quantification treatment module 22 is used for:
For each grouping, according to the black sample size and white sample size in the grouping, in the grouping
Sample carries out characteristic quantification processing, obtains the group quantization index value of the grouping；
According to the group quantization index value of each grouping and the sample size of each grouping, obtain described
The quantization index value of target signature.
In a kind of optional implementation, when the quantization index value is evidence weight index value, the quantification treatment mould
Block 22, is used for:
For each grouping, determine that the black sample size in the grouping accounts for black sample size in the sample set
First ratio, and determine that the white sample size in the grouping accounts for the second ratio of white sample size in the sample set；Base
In first ratio and second ratio, the grouping evidence weight index value of the grouping is determined；
Grouping evidence weight index value based on each grouping, is weighted the sample size of each grouping
Processing, obtains the evidence weight index value of the target signature.
In a kind of optional implementation, when the quantization index value is order of magnitude index value, quantification treatment module 22 is used
In:
For each grouping, in the black sample size in the grouping and the grouping between white sample size is determined
Three ratios, the packet count magnitude index value as the grouping；
Packet count magnitude index value based on each grouping is weighted place to the sample size of each grouping
Reason, obtains the order of magnitude index value of the target signature.
In a kind of optional implementation, when the quantization index value is information value index value, quantification treatment module 22,
For:
For each grouping, determine that the black sample size in the grouping accounts for the black sample size in the sample set
The 4th ratio, and determine the grouping in white sample size account for the white sample size in the sample set the 5th ratio
Example；Based on the difference of the 4th ratio, the 5th ratio and the 4th ratio and the 5th ratio, determining should
The grouping information value index value of grouping；
Grouping information value index value based on each grouping, is weighted the sample size of each grouping
Processing, obtains the information value index value of the target signature.
In a kind of optional implementation, warning module 23 is used for:
History quantization index value based on the target signature determines that the quantizating index refers to relative to history quantization
The fluctuating range of scale value；
If the fluctuating range in preset amplitude range, generates and pushes the warning information.
In a kind of optional implementation, described device further include:
Module is adjusted, for being adjusted to the corresponding system strategy of the target signature according to the quantization index value.
About above-mentioned apparatus, wherein the concrete function of modules is pre- in system strategy provided in an embodiment of the present invention
It is described in detail in the embodiment of alarm method, no detailed explanation will be given here.
The third aspect, is based on inventive concept same as system strategy method for early warning in previous embodiment, and the present invention also mentions
For a kind of server, as shown in figure 3, including memory 404, processor 402 and being stored on memory 404 and can be in processor
The computer program run on 402, the processor 402 realize system strategy method for early warning described previously when executing described program
Either method the step of.
Wherein, in Fig. 3, bus architecture (is represented) with bus 400, and bus 400 may include any number of interconnection
Bus and bridge, bus 400 will include the one or more processors represented by processor 402 and what memory 404 represented deposits
The various circuits of reservoir link together.Bus 400 can also will peripheral equipment, voltage-stablizer and management circuit etc. it
Various other circuits of class link together, and these are all it is known in the art, therefore, no longer carry out further to it herein
Description.Bus interface 406 provides interface between bus 400 and receiver 401 and transmitter 403.Receiver 401 and transmitter
403 can be the same element, i.e. transceiver, provide the unit for communicating over a transmission medium with various other devices.Place
It manages device 402 and is responsible for management bus 400 and common processing, and memory 404 can be used for storage processor 402 and execute behaviour
Used data when making.
Fourth aspect, based on the inventive concept based on system strategy method for early warning in previous embodiment, the present invention also mentions
For a kind of computer readable storage medium, it is stored thereon with computer program, institute above is realized when which is executed by processor
The step of stating based on either system strategy method for early warning method.
This specification is referring to the method, equipment (system) and computer program product according to this specification embodiment
Flowchart and/or the block diagram describes.It should be understood that can be realized by computer program instructions every in flowchart and/or the block diagram
The combination of process and/or box in one process and/or box and flowchart and/or the block diagram.It can provide these computers
Processor of the program instruction to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
To generate a machine, so that generating use by the instruction that computer or the processor of other programmable data processing devices execute
In setting for the function that realization is specified in one or more flows of the flowchart and/or one or more blocks of the block diagram
It is standby.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of equipment, the commander equipment realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the scope of the invention.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (20)
1. a kind of system strategy method for early warning, is applied to goal systems, the goal systems is configured with more than one system plan
Slightly, the system strategy is configured based on corresponding target signature, which comprises
The multiple accounts for including based on the goal systems construct sample set, include as black sample in the sample set
Adventure account, and the credible account as white sample；
Based on the corresponding target signature of the system strategy, characteristic quantification processing is carried out to the sample set, obtains the mesh
Mark the quantization index value of feature；
Determine whether the quantization index value meets abnormal quantization indicator conditions, if the quantization index value meets the abnormal amount
Change indicator conditions, then generate and push warning information, to carry out early warning to the corresponding system strategy of the target signature.
2. according to the method described in claim 1, described be based on the corresponding target signature of the system strategy, to the sample set
It closes and carries out characteristic quantification processing, obtain the quantization index value of the target signature, comprising:
Based on the target signature, processing is grouped to the sample set, obtains multiple groupings；
Characteristic quantification processing is carried out to the sample in each grouping of the multiple grouping respectively, obtains the amount of the target signature
Change index value.
3. being grouped place to the sample set according to the method described in claim 2, described be based on the target signature
Reason, obtains multiple groupings, comprising:
According to the sample characteristics that sample each in the sample set is included, determine to include the target in sample characteristics
Feature to branch mailbox sample；
Described branch mailbox processing will be carried out to branch mailbox sample, and obtain multiple branch mailbox.
4. according to the method in claim 2 or 3, the sample in each grouping to the multiple grouping carries out respectively
Characteristic quantification processing, obtains the quantization index value of the target signature, comprising:
For each grouping, according to the black sample size and white sample size in the grouping, to the sample in the grouping
Characteristic quantification processing is carried out, the group quantization index value of the grouping is obtained；
According to the group quantization index value of each grouping and the sample size of each grouping, the target is obtained
The quantization index value of feature.
5. described for described every according to the method described in claim 4, when the quantization index value is evidence weight index value
A grouping carries out at characteristic quantification the sample in the grouping according to the black sample size and white sample size in the grouping
Reason, obtains the group quantization index value of the grouping, comprising:
For each grouping, determine that the black sample size in the grouping accounts for first of black sample size in the sample set
Ratio, and determine that the white sample size in the grouping accounts for the second ratio of white sample size in the sample set；Based on institute
The first ratio and second ratio are stated, determines the grouping evidence weight index value of the grouping；
It is described according to the group quantization index value of each grouping and the sample size of each grouping, obtain described
The quantization index value of target signature, comprising:
Grouping evidence weight index value based on each grouping is weighted place to the sample size of each grouping
Reason, obtains the evidence weight index value of the target signature.
6. described for described each according to the method described in claim 4, when the quantization index value is order of magnitude index value
Grouping carries out characteristic quantification processing to the sample in the grouping according to the black sample size and white sample size in the grouping,
Obtain the group quantization index value of the grouping, comprising:
For each grouping, the third ratio in the black sample size in the grouping and the grouping between white sample size is determined
Example, the packet count magnitude index value as the grouping；
It is described according to the group quantization index value of each grouping and the sample size of each grouping, obtain described
The quantization index value of target signature, comprising:
Packet count magnitude index value based on each grouping is weighted processing to the sample size of each grouping,
Obtain the order of magnitude index value of the target signature.
7. described for described every according to the method described in claim 4, when the quantization index value is information value index value
A grouping carries out at characteristic quantification the sample in the grouping according to the black sample size and white sample size in the grouping
Reason, obtains the group quantization index value of the grouping, comprising:
For each grouping, the of the black sample size that the black sample size in the grouping accounts in the sample set is determined
Four ratios, and determine that the white sample size in the grouping accounts for the 5th ratio of the white sample size in the sample set；Base
In the difference of the 4th ratio, the 5th ratio and the 4th ratio and the 5th ratio, the grouping is determined
Grouping information value index value；
It is described according to the group quantization index value of each grouping and the sample size of each grouping, obtain described
The quantization index value of target signature, comprising:
Grouping information value index value based on each grouping is weighted place to the sample size of each grouping
Reason, obtains the information value index value of the target signature.
8. according to the method described in claim 1, whether the determination quantization index value meets abnormal quantization indicator conditions,
If the quantization index value meets the abnormal quantization indicator conditions, generates and pushes warning information, comprising:
History quantization index value based on the target signature determines the quantizating index relative to the history quantization index value
Fluctuating range；
If the fluctuating range in preset amplitude range, generates and pushes the warning information.
9. according to the method described in claim 1, after the quantization index value meets the abnormal quantization indicator conditions, institute
State method further include:
According to the quantization index value, the corresponding system strategy of the target signature is adjusted.
10. a kind of system strategy prior-warning device, is applied to goal systems, the goal systems is configured with more than one system plan
Slightly, the system strategy is configured based on corresponding target signature, and described device includes:
Sample constructs module, and multiple accounts for including based on the goal systems construct sample set, the sample set
In include adventure account as black sample, and the credible account as white sample；
Quantification treatment module carries out characteristic quantity to the sample set for being based on the corresponding target signature of the system strategy
Change processing, obtains the quantization index value of the target signature；
Warning module, for determining whether the quantization index value meets abnormal quantization indicator conditions, if the quantization index value
Meet the abnormal quantization indicator conditions, then warning information is generated and push, to the corresponding system strategy of the target signature
Carry out early warning.
11. device according to claim 10, the quantification treatment module, are used for:
Based on the target signature, processing is grouped to the sample set, obtains multiple groupings；
Characteristic quantification processing is carried out to the sample in each grouping of the multiple grouping respectively, obtains the amount of the target signature
Change index value.
12. device according to claim 11, the quantification treatment module, are used for:
According to the sample characteristics that sample each in the sample set is included, determine to include the target in sample characteristics
Feature to branch mailbox sample；
Described branch mailbox processing will be carried out to branch mailbox sample, and obtain multiple branch mailbox.
13. device according to claim 11 or 12, the quantification treatment module, are used for:
For each grouping, according to the black sample size and white sample size in the grouping, to the sample in the grouping
Characteristic quantification processing is carried out, the group quantization index value of the grouping is obtained；
According to the group quantization index value of each grouping and the sample size of each grouping, the target is obtained
The quantization index value of feature.
14. device according to claim 13, when the quantization index value is evidence weight index value, the quantification treatment
Module is used for:
For each grouping, determine that the black sample size in the grouping accounts for first of black sample size in the sample set
Ratio, and determine that the white sample size in the grouping accounts for the second ratio of white sample size in the sample set；Based on institute
The first ratio and second ratio are stated, determines the grouping evidence weight index value of the grouping；
Grouping evidence weight index value based on each grouping is weighted place to the sample size of each grouping
Reason, obtains the evidence weight index value of the target signature.
15. device according to claim 13, when the quantization index value is order of magnitude index value, the quantification treatment mould
Block is used for:
For each grouping, the third ratio in the black sample size in the grouping and the grouping between white sample size is determined
Example, the packet count magnitude index value as the grouping；
Packet count magnitude index value based on each grouping is weighted processing to the sample size of each grouping,
Obtain the order of magnitude index value of the target signature.
16. device according to claim 13, when the quantization index value is information value index value, the quantification treatment
Module is used for:
For each grouping, the of the black sample size that the black sample size in the grouping accounts in the sample set is determined
Four ratios, and determine that the white sample size in the grouping accounts for the 5th ratio of the white sample size in the sample set；Base
In the difference of the 4th ratio, the 5th ratio and the 4th ratio and the 5th ratio, the grouping is determined
Grouping information value index value；
Grouping information value index value based on each grouping is weighted place to the sample size of each grouping
Reason, obtains the information value index value of the target signature.
17. device according to claim 10, the warning module, are used for:
History quantization index value based on the target signature determines the quantizating index relative to the history quantization index value
Fluctuating range；
If the fluctuating range in preset amplitude range, generates and pushes the warning information.
18. device according to claim 10, described device further include:
Module is adjusted, for being adjusted to the corresponding system strategy of the target signature according to the quantization index value.
19. a kind of server including memory, processor and stores the computer that can be run on a memory and on a processor
The step of program, the processor realizes any one of claim 1-9 the method when executing described program.
20. a kind of computer readable storage medium, is stored thereon with computer program, power is realized when which is executed by processor
Benefit requires the step of any one of 1-9 the method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910617727.7A CN110472827B (en) | 2019-07-10 | 2019-07-10 | System policy early warning method, device, server and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910617727.7A CN110472827B (en) | 2019-07-10 | 2019-07-10 | System policy early warning method, device, server and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110472827A true CN110472827A (en) | 2019-11-19 |
| CN110472827B CN110472827B (en) | 2023-12-19 |
Family
ID=68507503
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910617727.7A Active CN110472827B (en) | 2019-07-10 | 2019-07-10 | System policy early warning method, device, server and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110472827B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399384A (en) * | 2022-03-25 | 2022-04-26 | 鲁担(山东)数据科技有限公司 | Risk strategy generation method, system and device based on privacy calculation |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106897880A (en) * | 2015-12-18 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of account methods of risk assessment and equipment |
| US20170289091A1 (en) * | 2015-05-20 | 2017-10-05 | Tencent Technology (Shenzhen) Company Limited | Warning method and apparatus, and processing server |
| CN107833137A (en) * | 2017-11-03 | 2018-03-23 | 上海宽全智能科技有限公司 | Quantization trading strategies generation method and device, equipment and storage medium based on multiple-objection optimization |
| CN109615503A (en) * | 2018-10-26 | 2019-04-12 | 深圳市元征科技股份有限公司 | A kind of prediction technique and relevant device of credit risk |
-
2019
- 2019-07-10 CN CN201910617727.7A patent/CN110472827B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170289091A1 (en) * | 2015-05-20 | 2017-10-05 | Tencent Technology (Shenzhen) Company Limited | Warning method and apparatus, and processing server |
| CN106897880A (en) * | 2015-12-18 | 2017-06-27 | 阿里巴巴集团控股有限公司 | A kind of account methods of risk assessment and equipment |
| CN107833137A (en) * | 2017-11-03 | 2018-03-23 | 上海宽全智能科技有限公司 | Quantization trading strategies generation method and device, equipment and storage medium based on multiple-objection optimization |
| CN109615503A (en) * | 2018-10-26 | 2019-04-12 | 深圳市元征科技股份有限公司 | A kind of prediction technique and relevant device of credit risk |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114399384A (en) * | 2022-03-25 | 2022-04-26 | 鲁担(山东)数据科技有限公司 | Risk strategy generation method, system and device based on privacy calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110472827B (en) | 2023-12-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104737132B (en) | For the message queue in on-demand service environment based on the resource-sharing bidded | |
| CN108985553A (en) | A kind of recognition methods and equipment of abnormal user | |
| CN109447354A (en) | A kind of intelligent bank note distribution method and device based on GA-BP neural network | |
| CN108665366A (en) | Determine method, terminal device and the computer readable storage medium of consumer's risk grade | |
| CN109034244A (en) | Line loss abnormality diagnostic method and device based on electric quantity curve characteristic model | |
| CN106817424B (en) | For controlling the method and system of flowing of access | |
| CN109191307A (en) | Asset Allocation reasonability judgment method, system, computer equipment and storage medium | |
| CN109034823A (en) | risk identification method, device and server | |
| CN106776651A (en) | The processing method and processing device of business dispute | |
| CN110471821A (en) | Abnormal alteration detection method, server and computer readable storage medium | |
| CN110348981A (en) | Loan air control method and device, electronic equipment and computer readable storage medium | |
| CN109255517A (en) | Generation method, device, server and the readable storage medium storing program for executing of air control strategy | |
| CN115689752A (en) | Method, device and equipment for adjusting wind control rule and storage medium | |
| CN110472827A (en) | System strategy method for early warning, device, server and readable storage medium storing program for executing | |
| CN106803815A (en) | A kind of flow control methods and device | |
| CN112702410A (en) | Evaluation system and method based on block chain network and related equipment | |
| CN116228312A (en) | Processing method and device for large-amount point exchange behavior | |
| CN110706122A (en) | Method, device and readable medium for improving social security agent efficiency based on big data analysis | |
| CN104090834B (en) | P2P platform information monitoring module, system and method | |
| CN115577983A (en) | Enterprise task matching method based on block chain, server and storage medium | |
| CN105868217A (en) | Automatic conversion method and system for numerical value | |
| CN113610627A (en) | Data processing method and device for risk early warning | |
| CN113538157B (en) | Annuity cost management method and annuity cost management device | |
| CN109447127A (en) | Data processing method and device | |
| CN117217901A (en) | Risk test method and device, storage medium and electronic device |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200925 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200925 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |