CN106844152B - Bank's background task runs the correlation analysis and device of batch time - Google Patents
Bank's background task runs the correlation analysis and device of batch time Download PDFInfo
- Publication number
- CN106844152B CN106844152B CN201710030688.1A CN201710030688A CN106844152B CN 106844152 B CN106844152 B CN 106844152B CN 201710030688 A CN201710030688 A CN 201710030688A CN 106844152 B CN106844152 B CN 106844152B
- Authority
- CN
- China
- Prior art keywords
- batch
- correlation analysis
- bank
- runs
- data set
- 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.)
- Active
Links
- 238000010219 correlation analysis Methods 0.000 title claims abstract description 121
- 230000002123 temporal effect Effects 0.000 claims abstract description 74
- 238000004458 analytical method Methods 0.000 claims abstract description 33
- 230000000737 periodic effect Effects 0.000 claims abstract description 8
- 230000009467 reduction Effects 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 17
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 5
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 5
- 230000002596 correlated effect Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 206010033799 Paralysis Diseases 0.000 description 2
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- YBMRDBCBODYGJE-UHFFFAOYSA-N germanium dioxide Chemical compound O=[Ge]=O YBMRDBCBODYGJE-UHFFFAOYSA-N 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004883 computer application Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000008140 language development Effects 0.000 description 1
- 230000004199 lung function Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000029305 taxis Effects 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/3006—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3051—Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3065—Monitoring arrangements determined by the means or processing involved in reporting the monitored data
- G06F11/3072—Monitoring arrangements determined by the means or processing involved in reporting the monitored data where the reporting involves data filtering, e.g. pattern matching, time or event triggered, adaptive or policy-based reporting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses correlation analysis and device that a kind of bank's background task runs batch time, wherein, method includes: to acquire the transaction system information of banking system, wherein transaction system information includes that system state amount information and bank's periodic task run batch time；Obtain running the data set of batch temporal correlation analysis model according to transaction system information and current trading situation；It is established according to the data set for running batch temporal correlation analysis model and runs batch temporal correlation analysis model, to obtain correlation analysis result.This method can establish race batch temporal correlation analysis model, so that deducing bank's background task runs the correlation criticized between time and numerous system state amounts, improve the accuracy and efficiency of analysis, simple easily to realize.
Description
Technical field
The present invention relates to computer application and bank technology field, in particular to a kind of bank's background task runs batch time
Correlation analysis and device.
Background technique
Currently, the safety of banking system and high efficiency, with regard to particularly important, wherein safety is even more the lifeblood of banking system,
But even so, large-scale failure still happens occasionally in terms of bank.And large-scale failure is frequently not by foreground
Caused by work mistake, because the thorough transaction step in bank foreground can almost prevent the generation of human error, and even if losing
Accidentally generation is also the small-scale mistake of one liang of transaction.Large-scale failure is caused by the failure of the system on backstage
's.Therefore, it is desirable to the significantly more efficient generation for avoiding bank's failure, we should focus on to set about from background system.But bank
The reason of background system is often sufficiently complex, causes failure is even more varied, Ke Nengyou: the linked network between bank is
The mismatch of system quantity of state and system mode, server for running transaction program etc. generate failure.And one of those
Failure often will cause a series of chain reaction, for example, all transaction requests will start when database is paralysed
Accumulation, so as to cause the inadequate resource of server；On the contrary, system gradually provides if the memory of server generates leakage
Source can be fewer and fewer, inadequate resource needed for the operation so as to cause database, final to paralyse.It can be seen that the system phase of rear end
Closing property is considerably complicated, it is desirable to it is almost impossible that the Producing reason that is out of order directly is analyzed by rule and method.Time that failure generates
Although number is rare, be not it is irregular follow, according to the experience in terms of bank, often system can be produced before the failure occurs
Raw some abnormal states, and the state of system often is more easier to monitor than failure, we can be by monitoring point in real time
The parameter of analysis system, to predict when failure will occur, this is also an important field of research in artificial intelligence.
One accurate failure predication can make warning to people in advance before the failure occurs, so as to use example
Such as malfunction elimination, data backup and hardware and software equipment are restarted appropriate mode and are coped with.Evaluate the steady of a system
It is qualitative to be evaluated from reliabilty and availability two indices.Here reliability refers to the probability of system jam, for
Reliability is often very high for banking system, i.e., few situation can break down, therefore is difficult from the angle of reliability
One promotion is made to the performance of system；And after availability refers to failure, the length of time required for system is restored, this individual character
Energy index is also highly important during actual use.Correspondence can be taken to arrange with look-ahead by failure prediction method
It applies, so that acceleration system resume speed, the availability of lifting system improves system performance under conditions of certain reliability.
On the other hand, since it is understood that some system parameters relevant with failure, then we can be by these parameters
Artificial limitation and adjustment are carried out to improve system volume reliability in certain degree in the preparatory generation for avoiding failure.
Due to the privacy of banking system, it is difficult to find the related text of the failure predication for being directed to bank transaction system
It offers.But failure predication this problem is always a general orientation of artificial intelligence field.Prediction of the people for the system failure
Technique study history has been over 30 years, and as system constantly becomes complicated, the method for failure predication is also growing with each passing hour
Development.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.
For this purpose, an object of the present invention is to provide the correlation analysis sides that a kind of bank's background task runs batch time
The accuracy and efficiency of analysis can be improved in method, this method, simple easily to realize.
It is another object of the present invention to the correlation analysis devices for proposing a kind of bank's background task race batch time.
In order to achieve the above objectives, one aspect of the present invention embodiment proposes the correlation of bank's background task race batch time a kind of
Property analysis method, comprising the following steps: acquire the transaction system information of banking system, wherein the transaction system information includes
System state amount information and bank's periodic task run batch time；It is obtained according to the transaction system information and current trading situation
Run the data set of batch temporal correlation analysis model；It establishes to run according to the data set for running batch temporal correlation analysis model and criticize
Temporal correlation analysis model, to obtain correlation analysis result.
Bank's background task of the embodiment of the present invention runs the correlation analysis of batch time, passes through the transaction of banking system
System information, which is established, runs batch temporal correlation analysis model, so that deducing bank's background task runs batch time and numerous system shapes
Correlation between state amount improves the accuracy and efficiency of analysis, simple easily to realize.
In addition, the correlation analysis that bank's background task according to the above embodiment of the present invention runs batch time can be with
With following additional technical characteristic:
Further, in one embodiment of the invention, the calculating step for running batch temporal correlation analysis model
Include: to be pre-processed to the data set for running batch temporal correlation analysis model, obtains transaction system information vector；It obtains
Correlation coefficient and descending arrangement in the transaction system information vector between each information content, to obtain correlation analysis knot
Fruit.
Further, in one embodiment of the invention, described to the number for running batch temporal correlation analysis model
Carrying out pretreatment according to collection further comprises: being used described in canonical formula and the removal of critical data characteristic matching according to bank data format
Irrelevant information in data set；Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
Further, in one embodiment of the invention, further includes: if the current trading situation is lower than the first threshold
Value then assesses batch temporal correlation analysis model of running using absolute error；If the current trading situation is high
In second threshold, then batch temporal correlation analysis model of running is assessed using relative error, wherein second threshold
Value is greater than the first threshold.
Further, in one embodiment of the invention, described that temporal correlation analysis model is criticized according to described run
Data set, which is established, runs batch temporal correlation analysis model, further comprises: obtaining and run batch time in the data set；Respectively to institute
It states race and criticizes the independent correlation analysis of other data progress in time and the data set, when obtaining different performance data and run to criticize
Between correlation analysis model code.
In order to achieve the above objectives, another aspect of the present invention embodiment proposes the phase of bank's background task race batch time a kind of
Closing property analytical equipment, comprising: acquisition module, for acquiring the transaction system information of banking system, wherein the transaction system letter
Breath includes that system state amount information and bank's periodic task run batch time；Module is obtained, for believing according to the transaction system
Breath and current trading situation obtain running the data set of batch temporal correlation analysis model；Analysis module, for being criticized according to described run
The data set of temporal correlation analysis model, which is established, runs batch temporal correlation analysis model, to obtain correlation analysis result.
Bank's background task of the embodiment of the present invention runs the correlation analysis device of batch time, passes through the transaction of banking system
System information, which is established, runs batch temporal correlation analysis model, so that deducing bank's background task runs batch time and numerous system shapes
Correlation between state amount improves the accuracy and efficiency of analysis, simple easily to realize.
In addition, the correlation analysis device that bank's background task according to the above embodiment of the present invention runs batch time can be with
With following additional technical characteristic:
Further, in one embodiment of the invention, the calculating step for running batch temporal correlation analysis model
Include: to be pre-processed to the data set for running batch temporal correlation analysis model, obtains transaction system information vector；It obtains
Correlation coefficient and descending arrangement in the transaction system information vector between each information content, to obtain correlation analysis knot
Fruit.
Further, in one embodiment of the invention, described to the number for running batch temporal correlation analysis model
Carrying out pretreatment according to collection further comprises: being used described in canonical formula and the removal of critical data characteristic matching according to bank data format
Irrelevant information in data set；Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
Further, in one embodiment of the invention, further includes: evaluation module, in the current transaction feelings
When condition is lower than first threshold, batch temporal correlation analysis model of running is assessed using absolute error, and described
Current trading situation then uses relative error to assess batch temporal correlation analysis model of running when being higher than second threshold,
Wherein, the second threshold is greater than the first threshold.
Further, in one embodiment of the invention, the analysis module is also used to obtain runs in the data set
The time is criticized, and runs the independent correlation analysis of other data progress in batch time and the data set to described respectively, to obtain not
With performance data and run batch model code of temporal correlation analysis.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is the process that the correlation analysis of batch time is run according to bank's background task of one embodiment of the invention
Figure；
Fig. 2 is the structure that the correlation analysis device of batch time is run according to bank's background task of one embodiment of the invention
Schematic diagram.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
The correlation point that the bank's background task proposed according to embodiments of the present invention runs batch time is described with reference to the accompanying drawings
Method and device is analysed, describes the phase that the bank's background task proposed according to embodiments of the present invention runs batch time with reference to the accompanying drawings first
Closing property analysis method.
Fig. 1 is the flow chart that bank's background task of one embodiment of the invention runs the correlation analysis of batch time.
As shown in Figure 1, bank's background task run the correlation analysis of batch time the following steps are included:
In step s101, the transaction system information of banking system is acquired, wherein transaction system information includes system mode
It measures information and bank's periodic task runs batch time.
It is understood that the method for the embodiment of the present invention can be according to bank transaction system backstage O&M monitoring tools institute
The transaction system information (including system state amount information, bank's periodic task run the information such as batch time) of acquisition, to establish silver
Row background task runs correlation analysis model batch between time and system state amount, batch time is run towards bank's background task with
Feature reduction model, reduction system state amount and the bank's background task of system state amount are run the conspicuousness between batch time quantum and are examined
Test analysis model.
In step s 102, it is obtained running batch temporal correlation analysis model according to transaction system information and current trading situation
Data set.
In step s 103, it is established according to the data set for running batch temporal correlation analysis model and runs batch temporal correlation analysis
Model, to obtain correlation analysis result.
It is understood that running correlation analysis, data regularization and the statistics of batch time performance by quantity of state and task
Significance test excavates and meets the transaction system monitoring state amount information of conditional independence with analysis, can according to historical data,
Influence system is found out from system environments characteristic variable and runs batch factor of execution time, and quantifies each factor and batch processing is executed
The influence degree size of time, and then predict that duration is criticized in the race of each job stream and entirety according to system environments characteristic variable, in advance
Granularity is surveyed as unit of job stream, key operation.
Wherein, in one embodiment of the invention, the calculating step for running batch temporal correlation analysis model includes: to race
The data set for criticizing temporal correlation analysis model is pre-processed, and transaction system information vector is obtained；Obtain transaction system information
Correlation coefficient and descending arrangement in vector between each information content, to obtain correlation analysis result.
It is understood that in an embodiment of the present invention, it first can be for the transaction in bank's backstage transaction system
System information is cleaned and is denoised, efficiently to extract effective information.For example, the data of the embodiment of the present invention are mainly big
Type business bank backstage All Activity and its time of origin include: current trading situation, backstage times by extracting useful information
Batch time, system state amount are run in business.The number that background task runs batch temporal correlation analysis model can be formed by the step
According to collection, it can be used to carry out the analysis of next step.Include " noise letter much useless to this research work in original back-end data
Breath ".By filling in the value of missing, smooth noise data, identification or deleting outlier and solve inconsistent in data prediction
Property come " cleaning " data.The sons such as the removing of standard data format, abnormal data removing, error correcting and repeated data are completed to appoint
Business.In this project, canonical formula and critical data characteristic matching are used according to bank data format, remove irrelevant information.Later
It carries out reduction to carry out Feature Dimension Reduction to data data, reduces the calculation amount of subsequent process.
Further, in one embodiment of the invention, the data set for running batch temporal correlation analysis model is carried out
Pretreatment further comprises: being used according to bank data format unrelated in canonical formula and critical data characteristic matching removal data set
Information；Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
For example, using data dependence analysis method reliable and effective in current data analysis come complete cost model
It calculates, correlation calculations is demonstrated by taking Pearson's coefficient as an example, the calculating process for running batch temporal correlation analysis model can be concluded
For following steps:
Step S1 obtains in the vector for running the effective informations such as batch time arrow the original sample of input after pretreatment
Changing indicates.
Step S2 calculates the correlation coefficient between each information content using Pearson correlation coefficient calculation method and descending is arranged
Column.
Step S3 obtains the relationship for running batch time and other background job information according to calculated result.
Further, in one embodiment of the invention, the method for the embodiment of the present invention further include: if current transaction
Situation is lower than first threshold, then is assessed using absolute error batch temporal correlation analysis model is run；If current transaction
Situation is higher than second threshold, then is assessed using relative error batch temporal correlation analysis model is run, wherein second threshold
Greater than first threshold.
That is, the assessment for algorithm effect, the characteristics of we are according to commercial banks data: every when rush periods
Second trading volume may have thousands of pens, and may there was only two or three transaction in 5 minutes in the time-division in morning.Absolute error is used
Mode in conjunction with relative error is evaluated.Specifically when trading volume be lower than some threshold value when, we using absolute error come
It judges:
Δ=X-L,
Wherein X is predicted value, and L is actual issued transaction amount per second, when trading volume is higher than some threshold value, be can be used
Relative error:
Specifically, correlation is described in detail below, specific as follows:
(1) it returns
It returns, studies a stochastic variable Y to the dependence relation of another (X) or one group of (X1, X2 ..., Xk) variable
Analysis method.Commonly referred to as Y is dependent variable, and Xk is independent variable.Regression analysis is a kind of mathematical model.Regression analysis it is main in
Appearance is the quantitative relation formula determined between certain variables from one group of data, i.e., founding mathematical models and estimates therein unknown
Parameter.A certain production process is predicted or controlled using required model.
(2) machine learning
Machine learning be nearly more than 20 years rise a multi-field cross discipline, be related to probability theory, statistics, Approximation Theory,
The multiple subjects such as convextiry analysis.Machine Learning Theory be mainly analysis and design it is some allow computer can automatic " study " calculation
Method.Machine learning algorithm is that a kind of automatically analyze from data obtains rule, and assimilated equations predict unknown data
Algorithm.Because devising a large amount of statistical theory in learning algorithm, machine learning and system, statistical inference student's federation are especially close,
Referred to as Statistical Learning Theory.
(3) correlation analysis
Correlation analysis (correlation analysis), correlation analysis are between research object with the presence or absence of certain dependence
Relationship, and to specifically thering is the phenomenon that dependence to inquire into its related direction and its degree of correlation, it is between research stochastic variable
A kind of statistical method of correlativity.
Correlativity is a kind of relationship of uncertainty, for example, remember the height and weight of a people respectively with X and Y, or point
Not Ji per hectare dose and per hectare wheat yield, then X and Y obviously have relationship, and not arriving definitely can be by therein one
A degree for going accurately to determine another, here it is correlativities.
Correlation analysis is generally referred to as Linear correlative analysis.
(4) it is positively correlated
If X with Y change direction is consistent, such as the relationship of height and weight, r > 0；Generally, | r | > 0.95, exist significant
Property it is related；| r | >=0.8, it is highly relevant；0.5≤| r | < 0.8, moderate is related；0.3≤| r | < 0.5, lower correlation；|r|<
0.3, relationship is extremely weak, it is believed that uncorrelated.
(5) negatively correlated
If X's and Y is contrary, such as relationship of smoking and lung function, r < 0.
(6) Pearson correlation coefficient
In statistics, Pearson correlation coefficient (Pearson product-moment correlation
Coefficient) for measuring two correlations between variable X and Y, it is worth between -1 and 1.
Pearson correlation coefficient between two variables is defined as the quotient of covariance and standard deviation between two variables:
Above formula defines population correlation coefficient, and common lowercase Greek alpha p, which is used as, represents symbol.Estimate the covariance of sample
And standard deviation, sample correlation coefficient can be obtained, commonly use English lower case r and represent:
R also can obtain the expression formula with above formula equivalence by the criterion score Estimation of Mean of (Xi, Yi) sample point:
The variation range of Pearson correlation coefficient is -1 to 1.The value of coefficient means that X and Y can be very good by straight line for 1
Equation describes, and all data points are all fallen point-blank well, and Y increases with the increase of X.The value of coefficient
Mean that all data points are all fallen on straight line for -1, and Y is reduced with the increase of X.The value of coefficient means two for 0
There is no linear relationship between variable.
(7) Spearman rank correlation coefficient
In statistics, Spearman rank correlation coefficient is the nonparametric index for measuring the dependence of two variables.It
The correlation of two statistical variables is evaluated using dull equation.If there is no repetition values in data, and when two variables are complete
When being monotonically correlated, Spearman's correlation coefficient is then+1 or -1.
Spearman's correlation coefficient is defined as the Pearson correlation coefficient between grade variables.It is n for sample size
Sample, n initial data Xi, Yi are converted into level data xi, yi, related coefficient p are as follows:
The initial data descending position average in conceptual data according to it, is assigned a corresponding grade.This skin
Germania correlation is alternatively referred to as " rank is related "；That is, " grade " that is observed data is replaced by " rank ".Continuous
In distribution, it is observed the rank of data, is usually always less than the half of grade.However, in this case, rank and grade phase
Relationship number is consistent.More generally, the ratio for being observed " rank " of data and the population sample of estimation is less than specified value,
It is observed the half of value.That is, one kind that it is corresponding equivalent coefficient possible solution.Although being of little use, "
Rank correlation " still still has and is used.
Spearman's correlation coefficient shows the related direction of X (independent variable) and Y (relying on variable).If when X increases,
Y is intended to increase, and Spearman's correlation coefficient is then positive.If Y is intended to reduce, Spearman's correlation coefficient when X increases
Then it is negative.Spearman's correlation coefficient be zero show when X increase when Y there is no any taxis.When X and Y become closer to completely
When being monotonically correlated, Spearman's correlation coefficient can increase on absolute value.When X to Y completely monotone is related, Spearman phase
The absolute value of relationship number is 1.Complete monotonic increase relationship means any two pairs of data Xi, Yi and Xj, Yj, have Xi-Xj and
Yi-Yj always jack per line.Complete monotone decreasing relationship means that any two pairs of data Xi, Yi and Xj, Yj have Xi-Xj and Yi-Yj
Always contrary sign.
Spearman's correlation coefficient is frequently referred to as " nonparametric ".Here there is two layers of meaning.Firstly, working as the relationship of X and Y
It is to be described by any monotonic function, then they are that complete Pearson came is relevant.With this corresponding, Pearson correlation coefficient can only
Provide the correlation of the X and Y by linear equation description.Secondly, Spearman do not need priori knowledge (that is, it is known that its
Parameter) can accurately obtain X and Y sampled probability distribution.
(8) Kendall's tau coefficient
Kendall's correlations coefficient is the statistical value for being used to measure two stochastic variable correlations.One Ken Deer is examined
It is a printenv hypothesis testing, it goes to examine the statistics dependence of two stochastic variables using calculated related coefficient.
The value range of Kendall's correlations coefficient, when τ is 1, indicates that two stochastic variables possess consistent grade phase between -1 to 1
Guan Xing；When τ is -1, indicate that two stochastic variables possess antipodal rank correlation；When τ is 0, expression two random
Variable is independent from each other.
Assuming that two stochastic variables are respectively X, Y (be also considered as two set), their element number is N, two
I-th (1≤i≤N) a value that a variable immediately takes is indicated with Xi, Yi respectively.Corresponding element in X and Y forms an element
To set XY, it includes element be (Xi, Yi) (1≤i≤N).When in set XY any two element (Xi, Yi) with (Xj,
Yj when seniority among brothers and sisters) is identical (that is when there are situation 1 or 2；Situation 1:Xi>Xj and Yi>Yj, situation 2:Xi<Xj and Yi<
Yj), the two elements are regarded as being consistent.When there are situation 3 or 4 (situation 3:Xi>Xj and Yi<Yj, situation 4:Xi<
Xj and Yi > Yj), the two elements are considered inconsistent.(situation 5:Xi=Xj, situation 6:Yi when there are situation 5 or 6
=Yj), the two elements are neither consistent nor inconsistent:
Further, in one embodiment of the invention, it is built according to the data set for running batch temporal correlation analysis model
It is vertical to run batch temporal correlation analysis model, further comprise: obtaining and run batch time in data set；Respectively to race batch time and data
Other data are concentrated to carry out independent correlation analysis, to obtain different performance data and run batch model generation of temporal correlation analysis
Code.
That is, the main purpose of the method for the embodiment of the present invention be provide large scale business bank transaction system information with
Bank's background task is run the correlation analysis model of batch time and is obtained first that is, on the basis of the initial data that bank provides
Secondly the daily race batch time carries out independent analysis to various system datas and race batch temporal correlation, finally develops for not
With performance data and run batch model code of temporal correlation analysis.
To sum up, after the completion of correlation models training, the association rules being calculated are formed the race batch task
Temporal correlation model, it is to be tested for new need to run batch task, it need to only be input in established correlation models and carry out
Background task runs batch temporal correlation analysis, can obtain its corresponding correlation analysis result.It is established according to the data set
Correlation models can intuitively visualize to obtain the relationship between each data.
Bank's background task according to an embodiment of the present invention runs the correlation analysis of batch time, passes through banking system
Transaction system information, which is established, runs batch temporal correlation analysis model, so that deducing bank's background task runs batch time and numerous systems
Correlation between system quantity of state, improves the accuracy and efficiency of analysis, simple easily to realize.
Referring next to the correlation point for bank's background task race batch time that attached drawing description proposes according to embodiments of the present invention
Analysis apparatus.
Fig. 2 is the structural representation that bank's background task of one embodiment of the invention runs the correlation analysis device of batch time
Figure.
As shown in Fig. 2, the correlation analysis device 10 that bank's background task runs batch time includes: acquisition module 100, obtains
Modulus block 200 and analysis module 300.
Wherein, acquisition module 100 is used to acquire the transaction system information of banking system, wherein transaction system information includes
System state amount information and bank's periodic task run batch time.Module 200 is obtained for according to transaction system information and currently
Trading situation obtains running the data set of batch temporal correlation analysis model.Analysis module 300 is used for according to race batch temporal correlation
The data set of analysis model, which is established, runs batch temporal correlation analysis model, to obtain correlation analysis result.The embodiment of the present invention
Device 10 can establish race batch temporal correlation analysis model, so that deducing bank's background task runs batch time and numerous systems
Correlation between system quantity of state, improves the accuracy and efficiency of analysis, simple easily to realize.
Further, in one embodiment of the invention, the calculating step of race batch temporal correlation analysis model includes:
The data set for running batch temporal correlation analysis model is pre-processed, transaction system information vector is obtained；Obtain transaction system
Correlation coefficient and descending arrangement in information vector between each information content, to obtain correlation analysis result.
Further, in one embodiment of the invention, the data set for running batch temporal correlation analysis model is carried out
Pretreatment further comprises: being used according to bank data format unrelated in canonical formula and critical data characteristic matching removal data set
Information；Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
Further, in one embodiment of the invention, the device 10 of the embodiment of the present invention further include: evaluation module.
Wherein, evaluation module is used for when current trading situation is lower than first threshold, is divided using absolute error batch temporal correlation is run
Analysis model is assessed, and then uses relative error to race batch temporal correlation when current trading situation is higher than second threshold
Analysis model is assessed, wherein second threshold is greater than first threshold.
Further, in one embodiment of the invention, when analysis module 300 is also used to obtain race batch in data set
Between, and carry out independent correlation analysis to running other data in batch time and data set respectively, with obtain different performance data with
Run batch model code of temporal correlation analysis.
It should be noted that the aforementioned correlation analysis embodiment for running batch time to bank's background task is explained
The bright bank's background task for being also applied for the embodiment runs the correlation analysis device of batch time, and details are not described herein again.
For example, the large scale business bank background task of the embodiment of the present invention runs batch temporal correlation analytical equipment, it can be right
Bank's backstage All Activity and its transaction time of origin data are analyzed, and the correlation of back-end data is extracted in analysis.In this base
Background task is established on plinth and runs batch temporal correlation model, can carry out induction and conclusion to emerging data.
It should be noted that the embodiment of the present invention can use data prediction and cleaning technique, Pearson correlation coefficient
The core technologies such as calculating, Kendall's tau coefficient computing technique, wherein the functions mould such as these algorithms and graphic user interface
Block is realized at Windows with language developments such as C++, java.
In addition, being based on above-mentioned development platform, the deployment that entire background task runs batch temporal correlation analysis system runs and needs
Want the support of following several level running environment.First in operating system layer, forecasting system is needed in Windows XP or its is simultaneous
It is run on the operating system platform of appearance；Program run time infrastructure, that is, java run time infrastructure are also needed simultaneously.Only
Have and had above-mentioned back-up environment, background task is run batch temporal correlation analysis system and just be can operate normally.And system makes
User only needs to see by local runtime system the correlation analysis result after prediction.
Bank's background task according to an embodiment of the present invention runs the correlation analysis device of batch time, passes through banking system
Transaction system information, which is established, runs batch temporal correlation analysis model, so that deducing bank's background task runs batch time and numerous systems
Correlation between system quantity of state, improves the accuracy and efficiency of analysis, simple easily to realize.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
" thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom" "inner", "outside", " up time
The orientation or positional relationship of the instructions such as needle ", " counterclockwise ", " axial direction ", " radial direction ", " circumferential direction " be orientation based on the figure or
Positional relationship is merely for convenience of description of the present invention and simplification of the description, rather than the device or element of indication or suggestion meaning must
There must be specific orientation, be constructed and operated in a specific orientation, therefore be not considered as limiting the invention.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present invention, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected；It can be directly connected, can also can be in two elements indirectly connected through an intermediary
The interaction relationship of the connection in portion or two elements, unless otherwise restricted clearly.For those of ordinary skill in the art
For, the specific meanings of the above terms in the present invention can be understood according to specific conditions.
In the present invention unless specifically defined or limited otherwise, fisrt feature in the second feature " on " or " down " can be with
It is that the first and second features directly contact or the first and second features pass through intermediary mediate contact.Moreover, fisrt feature exists
Second feature " on ", " top " and " above " but fisrt feature be directly above or diagonally above the second feature, or be merely representative of
First feature horizontal height is higher than second feature.Fisrt feature can be under the second feature " below ", " below " and " below "
One feature is directly under or diagonally below the second feature, or is merely representative of first feature horizontal height less than second feature.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (6)
1. the correlation analysis that a kind of bank's background task runs batch time, which comprises the following steps:
Acquire the transaction system information of banking system, wherein the transaction system information includes system state amount information and bank
Periodic task runs batch time；
Obtain running the data set of batch temporal correlation analysis model according to the transaction system information and current trading situation；And
It is established according to the data set for running batch temporal correlation analysis model and runs batch temporal correlation analysis model, to obtain phase
Closing property analysis result, wherein the calculating step for running batch temporal correlation analysis model includes: to run batch time correlation to described
The data set of property analysis model is pre-processed, and obtains transaction system information vector, wherein the pretreatment further comprises:
Irrelevant information in the data set is removed using canonical formula and critical data characteristic matching according to bank data format, to pretreatment
Data set afterwards carries out reduction, to carry out Feature Dimension Reduction；Obtain the phase in the transaction system information vector between each information content
Pass degree coefficient and descending arrangement, to obtain correlation analysis result.
2. the correlation analysis that bank's background task according to claim 1 runs batch time, which is characterized in that also wrap
It includes:
If the current trading situation is lower than first threshold, mould is analyzed to batch temporal correlation that runs using absolute error
Type is assessed；
If the current trading situation is higher than second threshold, mould is analyzed to batch temporal correlation that runs using relative error
Type is assessed, wherein the second threshold is greater than the first threshold.
3. the correlation analysis that bank's background task according to claim 1 runs batch time, which is characterized in that described
It is established according to the data set for running batch temporal correlation analysis model and runs batch temporal correlation analysis model, further comprise:
It obtains and runs batch time in the data set；
The independent correlation analysis of other data progress in batch time and the data set is run to described respectively, to obtain different performance
The model code of data and race batch temporal correlation analysis.
4. the correlation analysis device that a kind of bank's background task runs batch time characterized by comprising
Acquisition module, for acquiring the transaction system information of banking system, wherein the transaction system information includes system mode
It measures information and bank's periodic task runs batch time；
Module is obtained, runs batch temporal correlation analysis model for obtaining according to the transaction system information and current trading situation
Data set；And
Analysis module runs batch temporal correlation analysis for establishing according to the data set for running batch temporal correlation analysis model
Model, to obtain correlation analysis result, wherein the calculating step for running batch temporal correlation analysis model includes: to institute
The data set for stating race batch temporal correlation analysis model is pre-processed, and obtains transaction system information vector, wherein the pre- place
Reason further comprises: being used according to bank data format unrelated in canonical formula and the critical data characteristic matching removal data set
Information carries out reduction to pretreated data set, to carry out Feature Dimension Reduction；It obtains and respectively believes in the transaction system information vector
Correlation coefficient and descending arrangement between breath amount, to obtain correlation analysis result.
5. the correlation analysis device that bank's background task according to claim 4 runs batch time, which is characterized in that also wrap
It includes:
Evaluation module, for running batch time to described using absolute error when the current trading situation is lower than first threshold
Correlation analysis model is assessed, and then uses relative error to institute when the current trading situation is higher than second threshold
It states race batch temporal correlation analysis model to be assessed, wherein the second threshold is greater than the first threshold.
6. the correlation analysis device that bank's background task according to claim 4 runs batch time, which is characterized in that described
Analysis module, which is also used to obtain, runs batch time in the data set, and respectively to it is described run batch time with it is other in the data set
Data carry out independent correlation analysis, to obtain different performance data and run batch model code of temporal correlation analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710030688.1A CN106844152B (en) | 2017-01-17 | 2017-01-17 | Bank's background task runs the correlation analysis and device of batch time |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710030688.1A CN106844152B (en) | 2017-01-17 | 2017-01-17 | Bank's background task runs the correlation analysis and device of batch time |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106844152A CN106844152A (en) | 2017-06-13 |
| CN106844152B true CN106844152B (en) | 2019-10-29 |
Family
ID=59124927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710030688.1A Active CN106844152B (en) | 2017-01-17 | 2017-01-17 | Bank's background task runs the correlation analysis and device of batch time |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106844152B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509344B (en) * | 2018-04-04 | 2022-01-28 | 深圳前海微众银行股份有限公司 | Daily cutting batch test method, equipment and readable storage medium |
| CN113298510B (en) * | 2018-07-10 | 2022-06-17 | 马上消费金融股份有限公司 | Deduction instruction initiating method and device |
| CN109800887B (en) * | 2018-12-28 | 2021-01-22 | 东软集团股份有限公司 | Generation method and device of prediction process model, storage medium and electronic equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2473513B (en) * | 2009-09-14 | 2012-02-01 | Sony Comp Entertainment Europe | A method and apparatus for determining processor performance |
| JP5822799B2 (en) * | 2012-08-16 | 2015-11-24 | 株式会社三菱東京Ｕｆｊ銀行 | Information processing device |
| CN104123592B (en) * | 2014-07-15 | 2017-10-24 | 清华大学 | Bank's backstage TPS transaction events trend forecasting method and system |
| CN104156562A (en) * | 2014-07-15 | 2014-11-19 | 清华大学 | Failure predication system and failure predication method for background operation and maintenance system of bank |
-
2017
- 2017-01-17 CN CN201710030688.1A patent/CN106844152B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106844152A (en) | 2017-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11875232B2 (en) | Attributing reasons to predictive model scores | |
| EP3623964A1 (en) | Framework for the automated determination of classes and anomaly detection methods for time series | |
| CN111555921B (en) | Method and device for positioning alarm root cause, computer equipment and storage medium | |
| CN110929752B (en) | Grouping method based on knowledge driving and data driving and related equipment | |
| CN111314173B (en) | Monitoring information abnormity positioning method and device, computer equipment and storage medium | |
| CN106844152B (en) | Bank's background task runs the correlation analysis and device of batch time | |
| EP3916667A1 (en) | Real-time time series prediction for anomaly detection | |
| CN112289442A (en) | Method and device for predicting disease endpoint event and electronic equipment | |
| US20210365813A1 (en) | Management computer, management program, and management method | |
| US20240085274A1 (en) | Hybrid bearing fault prognosis with fault detection and multiple model fusion | |
| CN112421770A (en) | Automatic auxiliary inspection method and system for power distribution automation system | |
| CN113779882A (en) | Method, device, equipment and storage medium for predicting residual service life of equipment | |
| Wang et al. | Multilayer perceptron for sparse functional data | |
| CN113889262A (en) | Model-based data prediction method and device, computer equipment and storage medium | |
| CN116340796A (en) | Time sequence data analysis method, device, equipment and storage medium | |
| JP7120914B2 (en) | Production performance data analyzer | |
| CN110740063B (en) | Network flow characteristic index prediction method based on signal decomposition and periodic characteristics | |
| CN116048956A (en) | Software defect occurrence prediction method, device, equipment and storage medium | |
| CN114098764A (en) | Data processing method and device, electronic equipment and storage medium | |
| Wani et al. | Data Drift Monitoring for Log Anomaly Detection Pipelines | |
| Wang et al. | Health indicator forecasting for improving remaining useful life estimation | |
| Chan et al. | An unobserved components model of total factor productivity and the relative price of investment | |
| Niknafs et al. | Two-phase interarrival time prediction for runtime resource management | |
| Szabó et al. | In-silico simulation based evaluation of insulin prediction method for personalized medical treatment | |
| US20230368920A1 (en) | Learning apparatus, mental state sequence prediction apparatus, learning method, mental state sequence prediction method and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |