CN105279386A - Method and device for determining abnormal index data - Google Patents

Abstract

The invention discloses a method and a device for determining abnormal index data. The method comprises the steps of determining a to-be-evaluated index, acquiring current index data of the to-be-evaluated index, acquiring historical data of the to-be-evaluated index according to a statistics time point which corresponds with the current index data, successively determining the average level and the fluctuation level of the to-be-evaluated index according to the historical data of the to-be-evaluated index, and determining an abnormity index of the to-be-evaluated index according to the average level of the to-be-evaluated index. Through periodically extracting the historical data of a time segment when data points of the index arrive, a periodical effect of the data can be eliminated, and an abnormity coefficient is dynamically generated through integrating the average level and the fluctuation level of the to-be-evaluated index. Personnel can determine the abnormity degree of the current to-be-evaluated index just through comparing the abnormity index of the to-be-evaluated index with a constant 1/-1.

Description

A kind of method that Indexes Abnormality data are determined and device

Technical field

The present invention relates to data analysis technique field, particularly relate to method and device that a kind of Indexes Abnormality data determine.

Background technology

Judge the abnormal data of an index, conventional method sets most high threshold and lowest threshold, and the data not within the scope of this most high threshold and lowest threshold can be judged to be abnormal data.

In prior art, the size of data of different index varies, and this just needs each target setting threshold value, makes the mode of threshold decision implement very difficult; Meanwhile, even if same index, target setting threshold value that may be corresponding at different time is different, further increases the difficulty of threshold decision mode, if set, threshold value is unreasonable also can make the judgement of abnormal data inaccurate.

Summary of the invention

A kind of method that the embodiment of the present invention provides Indexes Abnormality data to determine and device, in order to determine the intensity of anomaly of current index to be assessed.

The method that a kind of Indexes Abnormality data that the embodiment of the present invention provides are determined, comprising:

Determine index to be assessed;

Obtain the current criteria data of described index to be assessed;

In the statistics moment corresponding according to described current criteria data, obtain the historical data of described index to be assessed;

According to the historical data of described index to be assessed, determine average level and the fluctuating level of described index to be assessed successively;

According to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determine the abnormal coefficient of described index to be assessed, whether the current criteria data that described abnormal coefficient is used to indicate described index to be assessed are abnormal data.

Preferably, also comprise:

Add up the abnormal coefficient of continuous m current criteria data, m >=0;

According to the abnormal coefficient of a described continuous m current criteria data, determine the anomaly trend of described index to be assessed;

Report to the police according to described anomaly trend.

Preferably, in the described statistics moment corresponding according to described current criteria data, obtain the historical data of described index to be assessed, comprising:

Determine the preset time period centered by the statistics moment that described current criteria data are corresponding, obtain the historical data being positioned at the index described to be assessed of described preset time period; Described preset time period determines according to the cycle of described index to be assessed.

Preferably, the average level of described index to be assessed is determined according to formula (1); The fluctuating level of described index to be assessed is determined according to formula (2);

Described formula (1) is:

$\mathrm{\μ}=\frac{{\mathrm{\Σ}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n;

Described formula (2) is:

$\mathrm{\σ}=\sqrt{\frac{{\mathrm{\Σ}}_{i=0}^n-{(x_i-\mathrm{\μ})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, μ is the average level of index to be assessed, n>=0,0≤i≤n.

Preferably, the abnormal coefficient of described index to be assessed is determined according to described formula (3);

Described formula (3) is:

$m=\frac{x-\mathrm{\μ}}{3\mathrm{\σ}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed;

M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

Preferably, the anomaly trend of described index to be assessed is determined according to described formula (4);

Described formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.

Correspondingly, the embodiment of the present invention additionally provides a kind of device determining abnormal data, comprising:

First determining unit, for determining index to be assessed;

First acquiring unit, for obtaining the current criteria data of described index to be assessed;

Second acquisition unit, for the statistics moment corresponding according to described current criteria data, obtains the historical data of described index to be assessed;

Second determining unit, for the historical data according to described index to be assessed, determines average level and the fluctuating level of described index to be assessed successively;

3rd determining unit, for according to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determine the abnormal coefficient of described index to be assessed, whether the current criteria data that described abnormal coefficient is used to indicate described index to be assessed are abnormal data.

Preferably, also comprise: alarm unit;

Described alarm unit specifically for:

Add up the abnormal coefficient of a continuous m achievement data, m >=0;

According to the abnormal coefficient of a described continuous m achievement data, determine the anomaly trend of described index to be assessed;

Report to the police according to described anomaly trend.

Preferably, described second acquisition unit specifically for:

Determine the preset time period centered by the statistics moment that described current criteria data are corresponding, obtain the historical data being positioned at the index described to be assessed of described preset time period; Described preset time period determines according to the cycle of described index to be assessed.

Preferably, described second determining unit specifically for:

The average level of described index to be assessed is determined according to formula (1); The fluctuating level of described index to be assessed is determined according to formula (2);

Described formula (1) is:

$\mathrm{\&mu;}=\frac{{\mathrm{\&Sigma;}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n;

Described formula (2) is:

$\mathrm{\&sigma;}=\sqrt{\frac{{\mathrm{\&Sigma;}}_{i=0}^n-{(x_i-\mathrm{\&mu;})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, μ is the average level of index to be assessed, n>=0,0≤i≤n.

Preferably, described 3rd determining unit specifically for:

The abnormal coefficient of described index to be assessed is determined according to described formula (3);

Described formula (3) is:

$m=\frac{x-\mathrm{\&mu;}}{3\mathrm{\&sigma;}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed;

M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

Preferably, described alarm unit specifically for:

The anomaly trend of described index to be assessed is determined according to described formula (4);

Described formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.

The embodiment of the present invention shows, by determining index to be assessed, obtain the current criteria data of index to be assessed, the statistics moment corresponding according to current criteria data, obtain the historical data of index to be assessed, according to the historical data of index to be assessed, determine average level and the fluctuating level of index to be assessed successively, according to the average level of index to be assessed and the fluctuating level of index to be assessed, determine the abnormal coefficient of index to be assessed.By the historical data of same period that arrives according to the data point of periodicity extraction index as analyzing samples, the cyclic effects of data can be eliminated, average level and the fluctuating level of comprehensive index to be assessed provide abnormal coefficient dynamically, staff only needs to compare the abnormal coefficient of this index to be assessed and the size of constant 1/-1, namely can know the intensity of anomaly of current index to be assessed.For each real-time achievement data, in conjunction with the history achievement data in statistics moment corresponding to this achievement data, the unusual condition of each achievement data of real-time assessment, the accuracy that both improve assessment abnormal turn improves the ageing of assessment; Meanwhile, in conjunction with average level and the fluctuating level of historical data, the accuracy of assessment is further increased

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly introduced, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

The schematic flow sheet of the method that Fig. 1 determines for a kind of Indexes Abnormality data that the embodiment of the present invention provides;

The schematic diagram of a kind of threshold value setting that Fig. 2 provides for the embodiment of the present invention;

The schematic diagram of a kind of threshold value setting that Fig. 3 provides for the embodiment of the present invention;

The structural representation of the device that Fig. 4 determines for a kind of Indexes Abnormality data that the embodiment of the present invention provides.

Embodiment

In order to make the object of the application, technical scheme and advantage clearly, be described in further detail the application below in conjunction with accompanying drawing, obviously, described embodiment is only a part of embodiment of the application, instead of whole embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not making other embodiments all obtained under creative work prerequisite, all belong to the scope of the application's protection.

The flow process that a kind of Indexes Abnormality data that Fig. 1 shows the embodiment of the present invention to be provided are determined, the device that this flow process can be determined by Indexes Abnormality data performs.

As shown in Figure 1, the concrete steps of this flow process comprise:

Step 101, determines index to be assessed.

Step 102, obtains the current criteria data of described index to be assessed.

Step 103, in the statistics moment corresponding according to described current criteria data, obtains the historical data of described index to be assessed.

Step 104, according to the historical data of described index to be assessed, determines average level and the fluctuating level of described index to be assessed successively.

Step 105, according to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determines the abnormal coefficient of described index to be assessed.

In a step 101, need to determine index to be assessed in multiple index, this index to be assessed can be the index needing monitoring.

In a step 102, after determining index to be assessed, when the current achievement data having this index to be assessed arrives, obtain the current criteria data of this index to be assessed, and record this current achievement data corresponding statistics moment.

In step 103, first determine the preset time period centered by the statistics moment that this current achievement data is corresponding, then obtain the historical data being positioned at this index to be assessed of preset time period.Determine according to cycle of this index to be assessed during this preset time period.The cycle of this index to be assessed can rule of thumb set, and this cycle can be one day, one week or one month, empirically sets during practical application.

The cycle of this preset time period and this index to be assessed is proportional, if the cycle of index to be assessed is one day, then this preset time period can be 20 minutes, accounts for 1.4% of the cycle of this index to be assessed.If the current criteria data of index to be assessed corresponding statistics moment is 12 points, and during preset time period 20 minutes, then centered by these 12, time period between 11: 50 to 12: 10 is the time period needing the historical data obtaining index to be assessed, thus obtain the historical data of this index to be assessed in this preset time period, the historical data of this index to be assessed all about namely between 11: 50 to 12: 10.

By being positioned at the historical data of same period as analyzing samples according to periodicity extraction index to be assessed, the cyclic effects of data can be eliminated.

At step 104, when after the historical data obtaining this index to be assessed in step 103, determine the average level of this index to be assessed according to formula (1).

Above-mentioned formula (1) is:

$\mathrm{\&mu;}=\frac{{\mathrm{\&Sigma;}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n.

After the average level determining this index to be assessed, according to the average level of this index to be assessed and the historical data of this index to be assessed, and formula (2) can determine the fluctuating level of this index to be assessed.

Above-mentioned formula (2) is:

$\mathrm{\&sigma;}=\sqrt{\frac{{\mathrm{\&Sigma;}}_{i=0}^n-{(x_i-\mathrm{\&mu;})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, μ is the average level of index to be assessed, n>=0,0≤i≤n.

In step 105, according to the average level of the index to be assessed determined in step 104 and the fluctuating level of this index to be assessed, can determine the abnormal coefficient of this index to be assessed, this abnormal coefficient may be used for indicating whether the current criteria data of described index to be assessed are abnormal data.As abnormal coefficient is greater than 1, then represent that the current criteria data exception of this index to be assessed is large; Abnormal coefficient is less than-1, then represent that the current criteria data exception of this index to be assessed is little, abnormal coefficient be greater than 1 or abnormal coefficient be less than-1 and all represent that the current criteria data of this index to be assessed are abnormal data.According to abnormal coefficient, staff can judge whether the current criteria data of this index to be assessed are abnormal data, if abnormal data, can be weakened this exception by the method that afterbody is average.

The abnormal coefficient of above-mentioned index to be assessed can pass through formula (3) and determine.

Above-mentioned formula (3) is:

$m=\frac{x-\mathrm{\&mu;}}{3\mathrm{\&sigma;}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed.M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

Average level and the fluctuating level of the comprehensive index to be assessed of the embodiment of the present invention provide abnormal coefficient dynamically, and staff only needs to compare the abnormal coefficient of this index to be assessed and the size of constant 1/-1, namely can know the intensity of anomaly of current index to be assessed.

When after the abnormal coefficient determining continuous m current criteria data, add up the abnormal coefficient of this continuous m current criteria data.According to the abnormal coefficient of this continuous m current criteria data, determine the anomaly trend of this index to be assessed.Then report to the police according to this anomaly trend, after this anomaly trend exceedes alarm threshold value, namely can report to the police.The embodiment of the present invention the type of alarm such as note, Hipchat can be provided, also can be the abnormal information of real-time rendering index in webpage, as exceptional data point can by mark red.

The anomaly trend of above-mentioned index to be assessed can be determined according to formula (4).

This formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.

As can be seen from formula (4), α is more ageing also can be better, more can reflect the anomaly trend of nearest data point.

The embodiment of the present invention by the web main website made a reservation, the amount of calling of the interfaces such as rear end RPC service, is called the indexs such as duration and is monitored, and the accident detections such as docking port time-out, brush note serve good alarm function.

The abnormal data of an index is judged in prior art, conventional method sets most high threshold and lowest threshold, as shown in Figure 2, upper and lower two straight lines are most high threshold and the lowest threshold of setting, and the achievement data not within the scope of this most high threshold and lowest threshold can be judged to be abnormal data.As shown in Figure 3, the embodiment of the present invention can provide the threshold value of dynamic achievement data, thus need not set threshold value to each index.

The embodiment of the present invention shows, by determining index to be assessed, obtain the current criteria data of index to be assessed, the statistics moment corresponding according to current criteria data, obtain the historical data of index to be assessed, according to the historical data of index to be assessed, determine average level and the fluctuating level of index to be assessed successively, according to the average level of index to be assessed and the fluctuating level of index to be assessed, determine the abnormal coefficient of index to be assessed.By the historical data of same period that arrives according to the data point of periodicity extraction index as analyzing samples, the cyclic effects of data can be eliminated, according to normal distribution principle, average level and the fluctuating level of comprehensive index to be assessed provide abnormal coefficient dynamically, only need to compare the abnormal coefficient of this index to be assessed and the size of constant 1/-1, namely can know the intensity of anomaly of current index to be assessed.

Based on identical technical conceive, the structure of the device that a kind of Indexes Abnormality data that Fig. 4 shows the embodiment of the present invention to be provided are determined, this device can the flow process determined of index of performance abnormal data.

As shown in Figure 4, this device specifically comprises:

First determining unit 401, for determining index to be assessed;

First acquiring unit 402, for obtaining the current criteria data of described index to be assessed;

Second acquisition unit 403, for the statistics moment corresponding according to described current criteria data, obtains the historical data of described index to be assessed;

Second determining unit 404, for the historical data according to described index to be assessed, determines average level and the fluctuating level of described index to be assessed successively;

3rd determining unit 405, for according to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determine the abnormal coefficient of described index to be assessed, whether the current criteria data that described abnormal coefficient is used to indicate described index to be assessed are abnormal data.

Preferably, also comprise: alarm unit;

Described alarm unit specifically for:

Add up the abnormal coefficient of a continuous m achievement data, m >=0;

According to the abnormal coefficient of a described continuous m achievement data, determine the anomaly trend of described index to be assessed;

Report to the police according to described anomaly trend.

Preferably, described second acquisition unit 403 specifically for:

Determine the preset time period centered by the statistics moment that described current criteria data are corresponding, obtain the historical data being positioned at the index described to be assessed of described preset time period; Described preset time period determines according to the cycle of described index to be assessed.

Preferably, described second determining unit 404 specifically for:

The average level of described index to be assessed is determined according to formula (1); The fluctuating level of described index to be assessed is determined according to formula (2);

Described formula (1) is:

$\mathrm{\&mu;}=\frac{{\mathrm{\&Sigma;}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n;

Described formula (2) is:

$\mathrm{\&sigma;}=\sqrt{\frac{{\mathrm{\&Sigma;}}_{i=0}^n-{(x_i-\mathrm{\&mu;})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, and xi is i-th achievement data in the historical data of index to be assessed, and μ is the average level of index to be assessed, n >=0,0≤i≤n.

Preferably, described 3rd determining unit 405 specifically for:

The abnormal coefficient of described index to be assessed is determined according to described formula (3);

Described formula (3) is:

$m=\frac{x-\mathrm{\&mu;}}{3\mathrm{\&sigma;}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed;

M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

Preferably, described alarm unit specifically for:

The anomaly trend of described index to be assessed is determined according to described formula (4);

Described formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.

The application describes with reference to according to the process flow diagram of the method for the embodiment of the present application, equipment (system) and computer program and/or block scheme.Should understand can by the combination of the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or square frame.These computer program instructions can being provided to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, making the instruction performed by the processor of computing machine or other programmable data processing device produce device for realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be stored in can in the computer-readable memory that works in a specific way of vectoring computer or other programmable data processing device, the instruction making to be stored in this computer-readable memory produces the manufacture comprising command device, and this command device realizes the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

These computer program instructions also can be loaded in computing machine or other programmable data processing device, make on computing machine or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computing machine or other programmable devices is provided for the step realizing the function of specifying in process flow diagram flow process or multiple flow process and/or block scheme square frame or multiple square frame.

Although described the preferred embodiment of the application, those skilled in the art once obtain the basic creative concept of cicada, then can make other change and amendment to these embodiments.So claims are intended to be interpreted as comprising preferred embodiment and falling into all changes and the amendment of the application's scope.

Obviously, those skilled in the art can carry out various change and modification to the application and not depart from the spirit and scope of the application.Like this, if these amendments of the application and modification belong within the scope of the application's claim and equivalent technologies thereof, then the application is also intended to comprise these change and modification.

Claims (12)

1. determine a method for abnormal data, it is characterized in that, comprising:

Determine index to be assessed;

Obtain the current criteria data of described index to be assessed;

In the statistics moment corresponding according to described current criteria data, obtain the historical data of described index to be assessed;

According to the historical data of described index to be assessed, determine successively described index to be assessed average level and fluctuating level;

According to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determine the abnormal coefficient of described index to be assessed, whether the current criteria data that described abnormal coefficient is used to indicate described index to be assessed are abnormal data.

2. the method for claim 1, is characterized in that, also comprises:

Add up the abnormal coefficient of continuous m current criteria data, m >=0;

According to the abnormal coefficient of a described continuous m current criteria data, determine the anomaly trend of described index to be assessed;

Report to the police according to described anomaly trend.

3. method as claimed in claim 1 or 2, is characterized in that, in the described statistics moment corresponding according to described current criteria data, obtain the historical data of described index to be assessed, comprising:

Determine the preset time period centered by the statistics moment that described current criteria data are corresponding, obtain the historical data being positioned at the index described to be assessed of described preset time period; Described preset time period determines according to the cycle of described index to be assessed.

4. method as claimed in claim 1 or 2, is characterized in that, determine the average level of described index to be assessed according to formula (1); The fluctuating level of described index to be assessed is determined according to formula (2);

Described formula (1) is:

$\mathrm{\&mu;}=\frac{{\mathrm{\&Sigma;}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n;

Described formula (2) is:

$\mathrm{\&sigma;}=\sqrt{\frac{{\mathrm{\&Sigma;}}_{i=0}^n{(x_i-\mathrm{\&mu;})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, μ is the average level of index to be assessed, n>=0,0≤i≤n.

5. method as claimed in claim 1 or 2, is characterized in that, determine the abnormal coefficient of described index to be assessed according to described formula (3);

Described formula (3) is:

$m=\frac{x-\mathrm{\&mu;}}{3\mathrm{\&sigma;}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed;

M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

6. method as claimed in claim 2, is characterized in that, determine the anomaly trend of described index to be assessed according to described formula (4);

Described formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.

7. determine a device for abnormal data, it is characterized in that, comprising:

First determining unit, for determining index to be assessed;

First acquiring unit, for obtaining the current criteria data of described index to be assessed;

Second acquisition unit, for the statistics moment corresponding according to described current criteria data, obtains the historical data of described index to be assessed;

Second determining unit, for the historical data according to described index to be assessed, determine successively described index to be assessed average level and fluctuating level;

3rd determining unit, for according to the average level of described index to be assessed and the fluctuating level of described index to be assessed, determine the abnormal coefficient of described index to be assessed, whether the current criteria data that described abnormal coefficient is used to indicate described index to be assessed are abnormal data.

8. device as claimed in claim 7, is characterized in that, also comprise: alarm unit;

Described alarm unit specifically for:

Add up the abnormal coefficient of a continuous m achievement data, m >=0;

According to the abnormal coefficient of a described continuous m achievement data, determine the anomaly trend of described index to be assessed;

Report to the police according to described anomaly trend.

9. as claimed in claim 7 or 8 device, is characterized in that, described second acquisition unit specifically for:

Determine the preset time period centered by the statistics moment that described current criteria data are corresponding, obtain the historical data being positioned at the index described to be assessed of described preset time period; Described preset time period determines according to the cycle of described index to be assessed.

10. as claimed in claim 7 or 8 device, is characterized in that, described second determining unit specifically for:

The average level of described index to be assessed is determined according to formula (1); The fluctuating level of described index to be assessed is determined according to formula (2);

Described formula (1) is:

$\mathrm{\&mu;}=\frac{{\mathrm{\&Sigma;}}_{i=0}^n{x}_i}{n}\mathrm{...}\left(1\right)$

Wherein, μ is the average level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, n>=0,0≤i≤n;

Described formula (2) is:

$\mathrm{\&sigma;}=\sqrt{\frac{{\mathrm{\&Sigma;}}_{i=0}^n{(x_i-\mathrm{\&mu;})}^2}{n}}\mathrm{...}\left(2\right)$

Wherein, σ is the fluctuating level of index to be assessed, x _ifor i-th achievement data in the historical data of index to be assessed, μ is the average level of index to be assessed, n>=0,0≤i≤n.

11. devices as claimed in claim 7 or 8, is characterized in that, described 3rd determining unit specifically for:

The abnormal coefficient of described index to be assessed is determined according to described formula (3);

Described formula (3) is:

$m=\frac{x-\mathrm{\&mu;}}{3\mathrm{\&sigma;}}\mathrm{...}\left(3\right)$

Wherein, m is the abnormal coefficient of index to be assessed, and x is current criteria data, and σ is the fluctuating level of index to be assessed, and μ is the average level of index to be assessed;

M > 1 or m <-1 represents that the current criteria data of index to be assessed are abnormal data.

12. devices as claimed in claim 8, is characterized in that, described alarm unit specifically for:

The anomaly trend of described index to be assessed is determined according to described formula (4);

Described formula (4) is:

$t=c+{\mathrm{\&alpha;\&Sigma;}}_{j=0}^m{(1-\mathrm{\&alpha;})}^{m-j}{x}_j\mathrm{...}\left(4\right)$

Wherein, t is the anomaly trend of index to be assessed, and c is constant, 0 < α < 1, x _jfor a jth current criteria data, m>=0,0≤j≤m.