CN106844152A - 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
- CN106844152A CN106844152A CN201710030688.1A CN201710030688A CN106844152A CN 106844152 A CN106844152 A CN 106844152A CN 201710030688 A CN201710030688 A CN 201710030688A CN 106844152 A CN106844152 A CN 106844152A
- Authority
- CN
- China
- Prior art keywords
- correlation analysis
- batch
- bank
- data set
- temporal correlation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Physics (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：The transaction system information of banking system is gathered, 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；Data set according to batch temporal correlation analysis model is run is set up and runs batch temporal correlation analysis model, to obtain correlation analysis result.The method can set up race batch temporal correlation analysis model, and the correlation criticized between time and numerous system state amounts is run so as to deduce bank's background task, improve the degree of accuracy and the efficiency of analysis, simple easily realization.
Description
Technical field
The present invention relates to computer application and bank technology field, more particularly to a kind of bank's background task runs batch time
Correlation analysis and device.
Background technology
At present, the security and high efficiency of banking system are just particularly important, and wherein security is even more the lifeblood of banking system,
But even so, the large-scale failure of bank's aspect still happens occasionally.And large-scale failure is frequently not by foreground
What work mistake was caused, because the thorough transaction step in bank foreground can almost prevent the generation of human error, even and if losing
Mistake is also the small-scale mistake of one or two 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 be set about from background system.But bank
Background system is often sufficiently complex, even more varied, Ke Nengyou the reason for cause failure：Linked network between bank, be
The mismatch of system quantity of state and system mode, the server etc. for running transaction program produces failure.And one of those
Failure often causes a series of chain reaction, such as, when database is paralysed, all of transaction request will start
Pile up, so as to cause the inadequate resource of server；If conversely, the internal memory of server produces leakage, then system money gradually
Source can be fewer and feweri, so as to inadequate resource needed for the operation for causing database, final paralysis.As can be seen here, the system phase of rear end
Closing property is considerably complicated, it is desirable to hardly possible by the rule and method Direct Analysis Producing reason that is out of order.It is secondary that failure is produced
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 is often more prone to monitoring than failure, we can be by monitoring point in real time
The parameter of analysis system, so as to predict when failure will occur, this is also an important field of research in artificial intelligence.
One accurate failure predication can give people to make warning in advance before the failure occurs, such that it is able to use example
Such as malfunction elimination, data backup and hardware and software equipment are restarted appropriate mode and are tackled.Evaluate the steady of 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 that situation very high, i.e., few can break down for banking system, therefore is difficult from the angle of reliability
Performance to system makes a lifting；And after availability refers to failure, system recover required for time length, this individual character
Energy index is also highly important during actually used.Correspondence can be taken to arrange with look-ahead by failure prediction method
Apply, so as under conditions of certain reliability, acceleration system resume speed, the availability of lifting system improves systematic function.
On the other hand, since it is understood that some systematic parameters related to failure, then we just can be by these parameters
Artificial limitation and adjustment are carried out so that in the advance generation for avoiding failure, system volume reliability is improved in certain degree.
Due to the privacy of banking system, therefore it is difficult to find the related text of the failure predication for being directed to bank transaction system
Offer.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, constantly becomes complicated with system, and the method for failure predication is also growing with each passing hour
Development.
The content of the invention
It is contemplated that at least solving one of technical problem in correlation technique to a certain extent.
Therefore, it is an object of the present invention to propose that a kind of bank's background task runs the correlation analysis side of batch time
Method, the method can improve the degree of accuracy and the efficiency of analysis, simple easily to realize.
It is another object of the present invention to propose a kind of correlation analysis device of bank's background task race batch time.
To reach above-mentioned purpose, one aspect of the present invention embodiment proposes the correlation that a kind of bank's background task runs batch time
Property analysis method, comprises the following steps：The transaction system information of banking system is gathered, wherein, the transaction system information includes
System state amount information and bank's periodic task run batch time；Obtained according to the transaction system information and current trading situation
Run the data set of batch temporal correlation analysis model；Set up 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, by the transaction of banking system
System information is set up and runs batch temporal correlation analysis model, and batch time and numerous system shapes are run so as to deduce bank's background task
Correlation between state amount, improves the degree of accuracy and the 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 calculation procedure for running batch temporal correlation analysis model
Including：The data set for running batch temporal correlation analysis model is pre-processed, transaction system information vector is obtained；Obtain
Correlation coefficient and descending arrangement in the transaction system information vector between each information content, to obtain correlation analysis knot
Really.
Further, in one embodiment of the invention, it is described to the number for running batch temporal correlation analysis model
Pretreatment is carried out according to collection to further include：According to bank data form is removed using canonical formula and critical data characteristic matching
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, also include：If the current trading situation is less than the first threshold
Value, then be estimated using absolute error to batch temporal correlation analysis model of running；If the current trading situation is high
In Second Threshold, then batch temporal correlation analysis model of running is estimated using relative error, wherein, second threshold
Value is more than the first threshold.
Further, in one embodiment of the invention, it is described that temporal correlation analysis model is criticized according to described race
Data set is set up and runs batch temporal correlation analysis model, further includes：Obtain the race batch time in the data set；Respectively to institute
State the race batch time carries out independent correlation analysis with other data in the data set, during obtaining different performance data with race batch
Between correlation analysis model code.
To reach above-mentioned purpose, another aspect of the present invention embodiment proposes the phase that a kind of bank's background task runs batch time
Closing property analytical equipment, including：Acquisition module, the transaction system information for gathering banking system, wherein, the transaction system letter
Breath includes that system state amount information and bank's periodic task run batch time；Acquisition module, for being believed 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 race
The data set of temporal correlation analysis model is set up and 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, by the transaction of banking system
System information is set up and runs batch temporal correlation analysis model, and batch time and numerous system shapes are run so as to deduce bank's background task
Correlation between state amount, improves the degree of accuracy and the 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 calculation procedure for running batch temporal correlation analysis model
Including：The data set for running batch temporal correlation analysis model is pre-processed, transaction system information vector is obtained；Obtain
Correlation coefficient and descending arrangement in the transaction system information vector between each information content, to obtain correlation analysis knot
Really.
Further, in one embodiment of the invention, it is described to the number for running batch temporal correlation analysis model
Pretreatment is carried out according to collection to further include：According to bank data form is removed using canonical formula and critical data characteristic matching
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, also include：Evaluation module, in the current transaction feelings
When condition is less than first threshold, batch temporal correlation analysis model of running is estimated using absolute error, and described
Current trading situation is then estimated using relative error when being higher than Second Threshold to batch temporal correlation analysis model of running,
Wherein, the Second Threshold is more than the first threshold.
Further, in one embodiment of the invention, the analysis module is additionally operable to obtain race in the data set
Batch time, and to described run batch time and carry out independent correlation analysis with other data in the data set respectively, to obtain not
With the model code that performance data is analyzed with race batch temporal correlation.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partly become from the following description
Obtain substantially, or recognized by practice of the invention.
Brief description of the drawings
The above-mentioned and/or additional aspect of the present invention and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and be readily appreciated that, wherein：
Fig. 1 is the flow of the correlation analysis that batch time is run according to bank's background task of one embodiment of the invention
Figure；
Fig. 2 is the structure of the correlation analysis device that batch time is run according to bank's background task of one embodiment of the invention
Schematic diagram.
Specific embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from start to finish
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
It is exemplary to scheme the embodiment of description, it is intended to for explaining the present invention, and be not considered as limiting the invention.
The bank's background task for describing proposition according to embodiments of the present invention with reference to the accompanying drawings runs the correlation point of batch time
Analysis method and device, the bank's background task for describing proposition according to embodiments of the present invention with reference to the accompanying drawings first runs the phase of batch time
Closing property analysis method.
Fig. 1 is the flow chart of the correlation analysis of bank's background task race batch time of one embodiment of the invention.
As shown in figure 1, the correlation analysis that bank's background task runs batch time are comprised the following steps：
In step S101, the transaction system information of banking system is gathered, wherein, transaction system information includes system mode
Amount information and bank's periodic task run 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 (running the information such as batch time including system state amount information, bank's periodic task) of collection, to set up silver
Row background task run correlation analysis model batch between time and system state amount, run towards bank's background task batch time with
The conspicuousness that the feature reduction model of system state amount, reduction system state amount and bank's background task are run between batch time quantum is examined
Test analysis model.
In step s 102, obtained running batch temporal correlation analysis model according to transaction system information and current trading situation
Data set.
In step s 103, set up according to the data set for running batch temporal correlation analysis model and run 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 the factor of batch execution time, and quantify each factor and batch processing is performed
The influence degree size of time, and then the race batch duration of each job stream and entirety is predicted according to system environments characteristic variable, in advance
Survey granularity is in units of job stream, key operation.
Wherein, in one embodiment of the invention, the calculation procedure for running batch temporal correlation analysis model includes：To running
The data set for criticizing temporal correlation analysis model is pre-processed, and obtains transaction system information vector；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, first can be for the transaction in bank's backstage transaction system
System information is cleaned and denoising, efficiently to extract effective information.For example, the data of the embodiment of the present invention are mainly greatly
Type business bank backstage All Activity and its time of origin, are included by the information for extracting useful：Current trading situation, backstage are appointed
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, can be used for carrying out the analysis of next step.Comprising a lot " noise letters useless to this research work in original back-end data
Breath ".In data prediction by filling in the value of missing, smooth noise data, recognize or delete outlier and solve inconsistent
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 form, remove irrelevant information.Afterwards
Data are carried out with reduction to carry out data Feature Dimension Reduction, the amount of calculation of subsequent process is reduced.
Further, in one embodiment of the invention, the data set for running batch temporal correlation analysis model is carried out
Pretreatment is further included：Used according to bank data form 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 reliable and effective data dependence analysis method in current data analysis come complete cost model
Calculate, correlation calculations are demonstrated by taking Pearson's coefficient as an example, the calculating process for running batch temporal correlation analysis model can be concluded
It is following steps：
Step S1, for the original sample being input into, obtains running the vector of the effective informations such as batch time arrow after pretreatment
Change and represent.
Step S2, the correlation coefficient and descending calculated using Pearson correlation coefficient computational methods between each information content is arranged
Row.
Step S3, the relation for running batch time and other background job information is drawn according to result of calculation.
Further, in one embodiment of the invention, the method for the embodiment of the present invention also includes：If current transaction
Situation is less than first threshold, then be estimated to running batch temporal correlation analysis model using absolute error；If current transaction
Situation is higher than Second Threshold, then be estimated to running batch temporal correlation analysis model using relative error, wherein, Second Threshold
More than first threshold.
That is, for the assessment of algorithm effect, we are according to the characteristics of commercial banks data：It is every during 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
The mode combined with relative error is evaluated.Specifically when trading volume be less than certain threshold value when, we using absolute error come
Judge：
Δ=X-L,
Wherein X is predicted value, and L is actual issued transaction amount per second, when trading volume is higher than certain threshold value, it is possible to use
Relative error：
Specifically, correlation is described in detail below, it is specific as follows：
(1) return
Return, one stochastic variable Y is to another (X) or the dependence relation of one group of (X1, X2 ..., Xk) variable for research
Analysis method.Commonly referred to as Y is dependent variable, and Xk is independent variable.Regression analysis is a class Mathematical Modeling.Regression analysis it is main in
Appearance is the quantitative relation formula for determining between some 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 multi-door subject such as convextiry analysis.Machine Learning Theory be mainly analysis and design some allow computer can " study " automatically calculation
Method.Machine learning algorithm is that a class is automatically analyzed from data and obtains rule, and assimilated equations are predicted to unknown data
Algorithm.Because devising substantial amounts of statistical theory in learning algorithm, machine learning is particularly close with system of statistical inference student's federation,
Referred to as Statistical Learning Theory.
(3) correlation analysis
Correlation analysis (correlation analysis), correlation analysis is with the presence or absence of certain dependence between research object
Relation, and phenomenon to specifically there is dependence inquires into its related direction and its degree of correlation, is between research stochastic variable
A kind of statistical method of dependency relation.
Dependency relation is a kind of relation of uncertainty, for example, remembering the height and body weight of people respectively with X and Y, or is divided
Not Ji per hectare dose and per hectare wheat yield, then X and Y obviously have relation, and can be by therein one definitely arriving
It is individual to go accurately to determine another degree, here it is dependency relation.
Correlation analysis is generally referred to as Linear correlative analysis.
(4) positive correlation
If X with Y change directions are consistent, such as height and the relation of body weight, r>0；Usually, | r |>0.95, exist significantly
Property it is related；|r|>=0.8, height correlation；0.5<=| r |<0.8, moderate is related；0.3<=| r |<0.5, lower correlation；|r|<
0.3, relation is extremely weak, it is believed that uncorrelated.
(5) it is negatively correlated
If X's and Y is in opposite direction, such as smoking and the relation of PFT, r<0.
(6) Pearson correlation coefficient
In statistics, Pearson correlation coefficient (Pearson product-moment correlation
Coefficient) it is used to measure two correlations between variable X and Y, is worth between -1 and 1.
Pearson correlation coefficient between two variables is defined as the business of the covariance and standard deviation between two variables：
Above formula defines population correlation coefficient, and conventional lowercase Greek alpha p is used as representing symbol.Estimate the covariance of sample
And standard deviation, sample correlation coefficient is can obtain, commonly use English lower case r and represent：
R also can obtain the expression formula of equal value with above formula by the criterion score Estimation of Mean of (Xi, Yi) sample point：
The excursion 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 is described, and all of data point all falls point-blank well, and Y increases with the increase of X.The value of coefficient
Mean that all of data point all falls 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 to weigh two nonparametric indexs of the dependence of variable.It
Two correlations of statistical variable are 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, coefficient correlation p and are：
Initial data is assigned a corresponding grade according to its average descending position in conceptual data.This skin
Germania is related to be alternatively referred to as " rank is related "；That is, " grade " that is observed data is replaced by " rank ".Continuous
In distribution, the rank of data is observed, generally always less than the half of grade.However, in this case, rank and grade phase
Relation number is consistent.More generally, " rank " of data and the ratio of the population sample estimated are observed 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 (dependence variable).If when X increases,
Y is intended to increase, and Spearman's correlation coefficient is then for just.If when X increases, Y is intended to reduce, Spearman's correlation coefficient
It is then negative.Spearman's correlation coefficient is zero to show that Y does not have any taxis when X increases.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 monotones are related, Spearman phase
The absolute value of relation number is 1.Complete monotonic increase relation means any two pairs of data Xi, Yi and Xj, Yj, have Xi-Xj and
Yi-Yj always jack per lines.Complete monotone decreasing relation means any two pairs of data Xi, Yi and Xj, and Yj has 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.First, when the relation of X and Y
Described by any monotonic function, then they are related complete Pearson cames.Corresponding with this, Pearson correlation coefficient can only
Provide the correlation of the X and Y described by linear equation.Secondly, Spearman do not need priori (that is, it is known that its
Parameter) just can accurately obtain X and Y sampled probability distribution.
(8) Kendall's tau coefficient
Kendall's correlations coefficient is one for measuring two statistical values of stochastic variable correlation.One Ken Deer inspection
It is a printenv hypothesis testing, it goes to check two statistics dependences of stochastic variable using calculated coefficient correlation.
The span of Kendall's correlations coefficient, when τ is 1, represents that two stochastic variables possess consistent grade phase between -1 to 1
Guan Xing；When τ is -1, represent that two stochastic variables possess antipodal rank correlation；When τ is 0, two are represented at random
Variable is separate.
Assuming that two stochastic variables are respectively X, Y (can also regard two set as), their element number is N, two
It is individual to become i-th (1 for measuring immediately<=i<=N) individual value represents with Xi, Yi respectively.Corresponding element in X and Y constitutes an element
To set XY, the element that it is included is (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 is situation 1 or 2；Situation 1：Xi>Xj and Yi>Yj, situation 2：Xi<Xj and Yi<
Yj), the two elements are regarded as consistent.(the situation 3 when there are situation 3 or 4：Xi>Xj and Yi<Yj, situation 4：Xi<
Xj and Yi>Yj), the two elements are considered as inconsistent.(the situation 5 when there are situation 5 or 6：Xi=Xj, situation 6：Yi
=Yj), the two elements are neither consistent nor inconsistent：
Further, in one embodiment of the invention, built according to the data set for running batch temporal correlation analysis model
It is vertical to run batch temporal correlation analysis model, further include：Obtain the race batch time in data set；Respectively to running batch time and data
Concentrating other data carries out independent correlation analysis, to obtain different performance data and run the model generation that batch temporal correlation is analyzed
Code.
That is, the main purpose of the method for the embodiment of the present invention be given large scale business bank transaction system information with
Bank's background task runs the correlation analysis model of batch time, i.e., on the basis of the initial data that bank provides, obtain first
The daily race batch time, secondly various system datas are individually analyzed with batch temporal correlation is run, finally developed for not
With the model code that performance data is analyzed with race batch temporal correlation.
To sum up, after the completion of correlation models training, the association rules being calculated are formed the race batch task
Temporal correlation model, for the race batch task tested the need for new, need to only be input into is carried out into the correlation models set up
Background task runs batch temporal correlation analysis, can obtain its corresponding correlation analysis result.Set up according to the data set
Correlation models can intuitively visualize the relation obtained between each data.
Bank's background task according to embodiments of the present invention runs the correlation analysis of batch time, by banking system
Transaction system information is set up and runs batch temporal correlation analysis model, and batch time and numerous systems are run so as to deduce bank's background task
Correlation between system quantity of state, improves the degree of accuracy and the efficiency of analysis, simple easily to realize.
The bank's background task proposed according to embodiments of the present invention referring next to Description of Drawings runs the correlation point of batch time
Analysis apparatus.
Fig. 2 is the structural representation of the correlation analysis device of bank's background task race batch time of one embodiment of the invention
Figure.
As shown in Fig. 2 the correlation analysis device 10 that bank's background task runs batch time includes：Acquisition module 100, obtain
Modulus block 200 and analysis module 300.
Wherein, acquisition module 100 is used to gather the transaction system information of banking system, wherein, transaction system information includes
System state amount information and bank's periodic task run batch time.Acquisition module 200 is 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 is set up and runs batch temporal correlation analysis model, to obtain correlation analysis result.The embodiment of the present invention
Device 10 can set up race batch temporal correlation analysis model, run batch time and numerous systems so as to deduce bank's background task
Correlation between system quantity of state, improves the degree of accuracy and the efficiency of analysis, simple easily to realize.
Further, in one embodiment of the invention, the calculation procedure for running batch temporal correlation analysis model includes：
Data set to running batch temporal correlation analysis model is pre-processed, and obtains transaction system information vector；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 is further included：Used according to bank data form 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 also includes：Evaluation module.
Wherein, evaluation module is used for when current trading situation is less than first threshold, using absolute error to running batch temporal correlation point
Analysis model is estimated, and when current trading situation is higher than Second Threshold then using relative error to running batch temporal correlation
Analysis model is estimated, wherein, Second Threshold is more than first threshold.
Further, in one embodiment of the invention, analysis module 300 be additionally operable to obtain data set in run batch when
Between, and to running batch time carry out independent correlation analysis with other data in data set respectively, with obtain different performance data with
Run batch model code of temporal correlation analysis.
It should be noted that the foregoing correlation analysis embodiment that batch time is run to bank 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 here is omitted.
For example, the large scale business bank background task of the embodiment of the present invention runs batch temporal correlation analytical equipment, 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 set up on plinth and runs batch temporal correlation model, induction and conclusion can be carried out 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 function mould such as these algorithms and graphic user interface
Block is realized under Windows with language developments such as C++, java.
In addition, being based on above-mentioned development platform, the deployment operation that whole background task runs batch temporal correlation analysis system is needed
Want the support of following several level running environment.First in operating system layer, forecasting system need Windows XP or itself and
Run on the operating system platform of appearance；Also need to program run time infrastructure, that is, java run time infrastructure simultaneously.Only
Have and possessed above-mentioned back-up environment, background task is run batch temporal correlation analysis system and could normally be run.And system makes
User only needs to the correlation analysis result after local runtime system can just see prediction.
Bank's background task according to embodiments of the present invention runs the correlation analysis device of batch time, by banking system
Transaction system information is set up and runs batch temporal correlation analysis model, and batch time and numerous systems are run so as to deduce bank's background task
Correlation between system quantity of state, improves the degree of accuracy and the efficiency of analysis, simple easily to realize.
In the description of the invention, it is to be understood that term " " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
" thickness ", " on ", D score, "front", "rear", "left", "right", " vertical ", " level ", " top ", " bottom " " interior ", " outward ", " up time
The orientation or position relationship of the instruction such as pin ", " counterclockwise ", " axial direction ", " radial direction ", " circumference " be based on orientation shown in the drawings or
Position relationship, is for only for ease of the description present invention and simplifies description, must rather than the device or element for indicating or imply meaning
With specific orientation, with specific azimuth configuration and operation, therefore must be not considered as limiting the invention.
Additionally, term " first ", " second " are only used for describing purpose, and it is not intended that indicating or implying relative importance
Or the implicit quantity for indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can express or
Implicitly include at least one this feature.In the description of the invention, " multiple " is meant that at least two, such as two, three
It is individual etc., unless otherwise expressly limited specifically.
In the present invention, unless otherwise clearly defined and limited, term " installation ", " connected ", " connection ", " fixation " etc.
Term should be interpreted broadly, for example, it may be fixedly connected, or be detachably connected, or integrally；Can be that machinery connects
Connect, or electrically connect；Can be joined directly together, it is also possible to be indirectly connected to by intermediary, can be in two elements
The connection in portion or two interaction relationships of element, unless otherwise clearly restriction.For one of ordinary skill in the art
For, can as the case may be understand above-mentioned term concrete meaning in the present invention.
In the present invention, unless otherwise clearly defined and limited, fisrt feature second feature " on " or D score can be with
It is the first and second feature directly contacts, or the first and second features are by intermediary mediate contact.And, fisrt feature exists
Second feature " on ", " top " and " above " but fisrt feature are directly over second feature or oblique upper, or be merely representative of
Fisrt feature level height is higher than second feature.Fisrt feature second feature " under ", " lower section " and " below " can be
One feature is immediately below second feature or obliquely downward, or is merely representative of fisrt feature level 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 to combine specific features, structure, material or spy that the embodiment or example are described
Point is contained at least one embodiment of the invention or example.In this manual, to the schematic representation of above-mentioned term not
Identical embodiment or example must be directed to.And, the specific features of description, structure, material or feature can be with office
Combined in an appropriate manner in one or more embodiments or example.Additionally, in the case of not conflicting, the skill of this area
Art personnel can be tied the feature of the different embodiments or example described in this specification and different embodiments or example
Close and combine.
Although embodiments of the invention have been shown and described above, it is to be understood that above-described embodiment is example
Property, it is impossible to limitation of the present invention is interpreted as, one of ordinary skill in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, changes, replacing and modification.
Claims (10)
1. a kind of bank's background task runs the correlation analysis of batch time, it is characterised in that comprise the following steps：
The transaction system information of banking system is gathered, 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
Set up according to the data set for running batch temporal correlation analysis model and run batch temporal correlation analysis model, to obtain phase
Closing property analysis result.
2. bank's background task according to claim 1 runs the correlation analysis of batch time, it is characterised in that described
The calculation procedure for running 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；
Correlation coefficient and the descending arrangement between each information content in the transaction system information vector are obtained, to obtain correlation
Analysis result.
3. bank's background task according to claim 1 runs the correlation analysis of batch time, it is characterised in that described
Pretreatment is carried out to the data set for running batch temporal correlation analysis model to further include：
Irrelevant information in canonical formula and the critical data characteristic matching removal data set is used according to bank data form；
Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
4. bank's background task according to claim 1 runs the correlation analysis of batch time, it is characterised in that also wrap
Include：
If the current trading situation is less than first threshold, batch temporal correlation analysis mould is run to described using absolute error
Type is estimated；
If the current trading situation is higher than Second Threshold, batch temporal correlation analysis mould is run to described using relative error
Type is estimated, wherein, the Second Threshold is more than the first threshold.
5. bank's background task according to claim 1 runs the correlation analysis of batch time, it is characterised in that described
Set up according to the data set for running batch temporal correlation analysis model and run batch temporal correlation analysis model, further included：
Obtain the race batch time in the data set；
Independent correlation analysis are carried out with other data in the data set to the race batch time respectively, to obtain different performance
Data criticize the model code of temporal correlation analysis with running.
6. a kind of bank's background task runs the correlation analysis device of batch time, it is characterised in that including：
Acquisition module, the transaction system information for gathering banking system, wherein, the transaction system information includes system mode
Amount information and bank's periodic task run batch time；
Acquisition module, for being obtained running batch temporal correlation analysis model according to the transaction system information and current trading situation
Data set；And
Analysis module, batch temporal correlation analysis is run for being set up according to the data set for running batch temporal correlation analysis model
Model, to obtain correlation analysis result.
7. bank's background task according to claim 6 runs the correlation analysis device of batch time, it is characterised in that described
The calculation procedure for running 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；
Correlation coefficient and the descending arrangement between each information content in the transaction system information vector are obtained, to obtain correlation
Analysis result.
8. bank's background task according to claim 6 runs the correlation analysis device of batch time, it is characterised in that described
Pretreatment is carried out to the data set for running batch temporal correlation analysis model to further include：
Irrelevant information in canonical formula and the critical data characteristic matching removal data set is used according to bank data form；
Reduction is carried out to pretreated data set, to carry out Feature Dimension Reduction.
9. bank's background task according to claim 6 runs the correlation analysis device of batch time, it is characterised in that also wrap
Include：
Evaluation module, for when the current trading situation is less than first threshold, batch time being run to described using absolute error
Correlation analysis model is estimated, and when the current trading situation is higher than Second Threshold then using relative error to institute
Race batch temporal correlation analysis model is stated to be estimated, wherein, the Second Threshold is more than the first threshold.
10. bank's background task according to claim 6 runs the correlation analysis device of batch time, it is characterised in that institute
State during analysis module is additionally operable to obtain the data set and run batch time, and to described run in batch time and the data set it respectively
Its data carries out independent correlation analysis, to obtain different performance data and run the model code that batch temporal correlation is analyzed.
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 true CN106844152A (en) | 2017-06-13 |
| CN106844152B 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) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509344A (en) * | 2018-04-04 | 2018-09-07 | 深圳前海微众银行股份有限公司 | Cut race batch test method, equipment and readable storage medium storing program for executing day |
| CN109800887A (en) * | 2018-12-28 | 2019-05-24 | 东软集团股份有限公司 | Predict generation method, device, storage medium and the electronic equipment of procedural model |
| CN113298510A (en) * | 2018-07-10 | 2021-08-24 | 马上消费金融股份有限公司 | Deduction instruction initiating method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2470995A2 (en) * | 2009-09-14 | 2012-07-04 | Sony Computer Entertainment Europe Limited | A method of determining the state of a tile based deferred rendering processor and apparatus thereof |
| JP2014038476A (en) * | 2012-08-16 | 2014-02-27 | Bank Of Tokyo-Mitsubishi Ufj Ltd | Information processing apparatus |
| CN104123592A (en) * | 2014-07-15 | 2014-10-29 | 清华大学 | Method and system for predicting transaction per second (TPS) transaction events of bank background |
| 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
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2470995A2 (en) * | 2009-09-14 | 2012-07-04 | Sony Computer Entertainment Europe Limited | A method of determining the state of a tile based deferred rendering processor and apparatus thereof |
| JP2014038476A (en) * | 2012-08-16 | 2014-02-27 | Bank Of Tokyo-Mitsubishi Ufj Ltd | Information processing apparatus |
| CN104123592A (en) * | 2014-07-15 | 2014-10-29 | 清华大学 | Method and system for predicting transaction per second (TPS) transaction events of bank background |
| 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 |
Non-Patent Citations (1)
| Title |
|---|
| 周磊等: "系统批量运行时间同交易量关联性分析", 《中国金融电脑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509344A (en) * | 2018-04-04 | 2018-09-07 | 深圳前海微众银行股份有限公司 | Cut race batch test method, equipment and readable storage medium storing program for executing day |
| CN113298510A (en) * | 2018-07-10 | 2021-08-24 | 马上消费金融股份有限公司 | Deduction instruction initiating method and device |
| 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 |
| CN109800887A (en) * | 2018-12-28 | 2019-05-24 | 东软集团股份有限公司 | Predict generation method, device, storage medium and the electronic equipment of procedural model |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106844152B (en) | 2019-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109255506B (en) | Internet financial user loan overdue prediction method based on big data | |
| Sahu et al. | Predicting software bugs of newly and large datasets through a unified neuro-fuzzy approach: Reliability perspective | |
| US20240078475A1 (en) | Attributing reasons to predictive model scores with local mutual information | |
| Bøttcher et al. | deal: A package for learning Bayesian networks | |
| CN104123592B (en) | Bank's backstage TPS transaction events trend forecasting method and system | |
| Hassan et al. | A HMM-based adaptive fuzzy inference system for stock market forecasting | |
| Li et al. | Attention-based deep survival model for time series data | |
| Ringsquandl et al. | Semantic-guided feature selection for industrial automation systems | |
| CN106844152B (en) | Bank's background task runs the correlation analysis and device of batch time | |
| JP7481902B2 (en) | Management computer, management program, and management method | |
| CN113889262A (en) | Model-based data prediction method and device, computer equipment and storage medium | |
| CN113807951A (en) | Transaction data trend prediction method and system based on deep learning | |
| CN116028315A (en) | Operation early warning method, device, medium and electronic equipment | |
| CN117076993A (en) | Multi-agent game decision-making system and method based on cloud protogenesis | |
| Zhao et al. | A new domain adaption residual separable convolutional neural network model for cross-domain remaining useful life prediction | |
| Varuna et al. | Trend prediction of GitHub using time series analysis | |
| CN116048956A (en) | Software defect occurrence prediction method, device, equipment and storage medium | |
| CN115344386A (en) | Method, device and equipment for predicting cloud simulation computing resources based on sequencing learning | |
| US20220188647A1 (en) | Model learning apparatus, data analysis apparatus, model learning method and program | |
| CN106919755A (en) | A kind of cloud manufacture system uncertainty quantitative analysis method and device based on data | |
| Zhang et al. | Arise: Aperiodic semi-parametric process for efficient markets without periodogram and gaussianity assumptions | |
| US11822564B1 (en) | Graphical user interface enabling interactive visualizations using a meta-database constructed from autonomously scanned disparate and heterogeneous sources | |
| US20230419104A1 (en) | High dimensional dense tensor representation for log data | |
| US11928128B2 (en) | Construction of a meta-database from autonomously scanned disparate and heterogeneous sources | |
| CN112801372B (en) | Data processing method, device, electronic equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |