CN106991145A - A kind of method and device of Monitoring Data - Google Patents
A kind of method and device of Monitoring Data Download PDFInfo
- Publication number
- CN106991145A CN106991145A CN201710178551.0A CN201710178551A CN106991145A CN 106991145 A CN106991145 A CN 106991145A CN 201710178551 A CN201710178551 A CN 201710178551A CN 106991145 A CN106991145 A CN 106991145A
- Authority
- CN
- China
- Prior art keywords
- index
- interaction data
- historical interaction
- monitoring
- average
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/30—Payment architectures, schemes or protocols characterised by the use of specific devices or networks
- G06Q20/34—Payment architectures, schemes or protocols characterised by the use of specific devices or networks using cards, e.g. integrated circuit [IC] cards or magnetic cards
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/02—Banking, e.g. interest calculation or account maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Technology Law (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Marketing (AREA)
- Fuzzy Systems (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Debugging And Monitoring (AREA)
Abstract
The present invention discloses a kind of method and device of Monitoring Data, and this method includes:Obtain the historical interaction data of each monitoring object in setting duration in trade link;The historical interaction data of each monitoring object is divided into the historical interaction data set of N number of index;For the historical interaction data set of the first index, following operation is performed, wherein, the first index is any one index in N number of index:Determine the first index historical interaction data set period of waves and period of waves in historical interaction data average, and the historical interaction data in the period of waves of the first index standard deviation;According to average and standard deviation, the monitoring baseline of the first index is determined, monitoring baseline is used for the fluctuation range for indicating the normal historical interaction data of the first index;Abnormal data in the interaction data set of first index is determined according to monitoring baseline, this method is to provide a kind of new abnormal data monitoring means to monitor the abnormality of the historical data of certain time length.
Description
Technical field
The present invention relates to data processing field, more particularly to a kind of method and device of Monitoring Data.
Background technology
At present, increasingly extensive and different field the class of business applied with cyber-net becomes increasingly abundant,
Interactive data information (such as the answer back code in process of exchange data in financial field) is analyzed and based on analysis result
Monitoring abnormal conditions become more and more important.
In existing technical scheme, when analyzing interactive data information, typically for the reality in information exchange
When data analyzed.Specifically, first gathering the magnanimity original real-time data interactive information associated with monitored object, Ran Houjin
Row data derivation operation performs anomalous discrimination operation to generate derivative achievement data, then based on derivative achievement data.Due to above-mentioned
The cycle of real time data is short, and relative change is big, thus for real time data anomalous discrimination operation be not particularly suited for it is longer when
Long historical data carries out anomalous discrimination, is primarily due to change of the existing technical scheme to full isl cycle historical data
It can not control.
The content of the invention
The embodiment of the present invention provides a kind of method and device of Monitoring Data, to provide a kind of new abnormal data monitoring
Means are to monitor the abnormality of the historical data of certain time length.
The inventive method includes a kind of method of Monitoring Data, and this method includes:Obtain the trade link in setting duration
In each monitoring object historical interaction data;
The historical interaction data of each monitoring object is divided into the historical interaction data set of N number of index;
For the historical interaction data set of first index, following operation is performed, wherein, first index is institute
State any one index in N number of index:
Determine first index historical interaction data set period of waves and the period of waves in history hand over
The average of mutual data, and the historical interaction data in the period of waves of first index standard deviation;
According to the average and the standard deviation, the monitoring baseline of first index is determined, the monitoring baseline is used for
Indicate the fluctuation range of the normal historical interaction data of first index;
The abnormal data in the interaction data set of first index is determined according to the monitoring baseline.
Based on same inventive concept, the embodiment of the present invention further provides for a kind of device of Monitoring Data, the device bag
Include:
Acquiring unit, the historical interaction data for obtaining each monitoring object in setting duration in trade link;
Division unit, the historical interaction data for the historical interaction data of each monitoring object to be divided into N number of index
Set;
Determining unit, for the historical interaction data set for first index, performs following operation, wherein, institute
It is any one index in N number of index to state the first index:Determine the historical interaction data set of first index
Period of waves and the historical interaction data in the period of waves average, and in the period of waves of first index
The standard deviation of historical interaction data;According to the average and the standard deviation, the monitoring baseline of first index is determined, it is described
Monitoring baseline is used for the fluctuation range for indicating the normal historical interaction data of first index;
Anticoincidence unit is sentenced, for determining the abnormal number in the interaction data set of first index according to the monitoring baseline
According to.
The embodiment of the present invention is by obtaining the interaction data of N number of index of the object to be monitored of certain time length, usually one
Then the data of acquisition are obtained the interaction data collection of each index by data more than week according to operational indicator classified types
Close, so that the abnormality of data in the interaction data set of the index is determined using the corresponding index baseline of the index, due to
Index baseline determines the fluctuation range of the first index of monitored object, so when the interaction data of index is higher than index baseline
The upper limit, or less than index baseline lower limit when, can trigger abnormal alarm.Wherein, index baseline is according to achievement data collection
Average and standard deviation determine, because the collection period of object to be monitored is longer with respect to real time data, and index baseline
It is the rule obtained according to history historical interaction data analysis interior for a period of time, therefore so carries out anomalous discrimination operation, it is accurate
True property is higher, reduces the probability of erroneous judgement.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, makes required in being described below to embodiment
Accompanying drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for this
For the those of ordinary skill in field, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of method flow schematic diagram of Monitoring Data provided in an embodiment of the present invention;
Fig. 2 is the schematic diagram of bank card business dealing chain index division methods provided in an embodiment of the present invention;
Fig. 3 is the schematic diagram of indicator rule collection provided in an embodiment of the present invention;
Fig. 4 is a kind of curve synoptic diagram for determining abnormal data provided in an embodiment of the present invention;
Fig. 5 is a kind of decomposing schematic representation of absolute magnitude abnormal cause provided in an embodiment of the present invention;
Fig. 6 is a kind of decomposing schematic representation of relative quantity abnormal cause provided in an embodiment of the present invention;
Fig. 7 is a kind of schematic diagram of year-on-year class data prediction abnormal data provided in an embodiment of the present invention;
Fig. 8 is a kind of schematic diagram of Multi-task Concurrency provided in an embodiment of the present invention;
Fig. 9 is a kind of schematic device of Monitoring Data provided in an embodiment of the present invention.
Embodiment
In order that the object, technical solutions and advantages of the present invention are clearer, below in conjunction with accompanying drawing the present invention is made into
One step it is described in detail, it is clear that described embodiment is only embodiment of the invention a part of, rather than whole implementation
Example.Based on the embodiment in the present invention, what those of ordinary skill in the art were obtained under the premise of creative work is not made
All other embodiment, belongs to the scope of protection of the invention.
Shown in Figure 1, the embodiment of the present invention provides a kind of method flow schematic diagram of Monitoring Data, specifically realization side
Method includes:
Step S101, obtains the historical interaction data of each monitoring object in setting duration in trade link.
Step S102, the historical interaction data of each monitoring object is divided into the historical interaction data set of N number of index.
Step S103, for the historical interaction data set of first index, performs following operation, wherein, described the
One index is any one index in N number of index:Determine the ripple of the historical interaction data set of first index
Dynamic cycle and the average of the historical interaction data in the period of waves, and the history in the period of waves of first index
The standard deviation of interaction data;
Step S104, according to the average and the standard deviation, determines the monitoring baseline of first index, the monitoring
Baseline is used for the fluctuation range for indicating the normal historical interaction data of first index;
Step S105, the abnormal data in the interaction data set of first index is determined according to the monitoring baseline.
Each monitoring object refers to the links such as bank card business dealing Lian Zhong trade companies, card issuer, bank in step S101.And
Index then refer to angularly being analyzed from finance, market, product involved by data target.On the whole, bank card is handed over
Easy link critical path index system is divided into three levels, as shown in Fig. 2 1), top layer, mainly according to the crucial ring of trade link
The class of business of monitoring, defines index system category needed for section;2), intermediate layer, is primarily referred to as the category division of mark system;3)、
Bottom, the refinement specific targets more than 6700 of index system form specific targets collection.
It should be noted that to all corresponding data of bank card business dealing chain index system, it is necessary to which advanced row data turn
Move, i.e., data are first transferred to the own numbers of operation data analysis system ODAS (Operation Data Analysis System)
According to storehouse, then the initial data in the own database is acquired and necessary filtering again, data storage is reduced after de-redundant
Pressure.In addition, the data come to transfer, it is necessary to perform following operation respectively:A1, according to index set, index derivative, rule base
And sentence different demand, the data structure of different field, level and classification is entered in terms of form, data magnitude, service priority
Row definition;A2, the data of nonstandardized technique are standardized;Dimension with some cycles, skewness is passed through into SS
Change formula to be converted;The data that are standardized after conversion are carried out Bedding storage by A3 according to index classification, field, magnitude, it is to avoid
The problem of follow-up index derives, data query carries out the influence performance such as full table scan.
So, after data acquisition is completed and preliminary processing is stored, next each achievement data will be concentrated
Data perform following handling process, step one respectively:Need advanced row data cleansing, step 2:Setting up this using rule, this refers to
Mark the index baseline of data set, step 3:The data concentrated using index baseline to the achievement data carry out sentencing different processing, step
Four:Again to there is abnormal data the reason for, carries out test analysis.
For the data cleansing in step one, it is possible to use data cleaning method present in prior art is cleaned,
It will not be repeated here.
Other step 2, which sets up index baseline, to be needed first to determine to set up the value of the parameters in the formula of index baseline, example
Such as average and standard deviation.Wherein it is determined that whether average needs the data in first agriculture products data set steady, if steadily, can
Directly to take average, that is, it is L's to take sliding window (i.e. the corresponding time cycle length of institute's observation index, day, week, the moon, season, year)
The average of sample point.If unstable, one can be formed using taking the methods such as difference, logarithm, unit root to carry out data conversion
New data sequence.So converted by difference, reach the tranquilization conversion of equal value stabilization.Then again to differentiated new
Sequence take average.Specifically.T inspections are carried out to the data in the historical interaction data set of first index, K is determined
The sequence of individual different averages;
Calculate the statistical correlation coefficient between K sequence and time attribute;
When determining that the statistical correlation coefficient is more than first threshold, difference is carried out to the K sequence, until differentiated
Statistical correlation coefficient is not more than the first threshold;
Calculate the average of differentiated sequence, and using the average of the differentiated sequence as in the period of waves
The average of historical interaction data.
That is, on the basis of for rule set generation module, being built by the method for assumed statistical inspection to monitoring baseline
Vertical, threshold value is set for processing, and touch the mark collection Auto-matching rule set, baseline, threshold value and sequence rule factor of influence algorithm,
After specific flow is as shown in figure 3, collect operational indicator collection and festivals or holidays table, it is then determined that index set and corresponding label, B1,
The bar number of agriculture products baseline is examined by T, i.e. Monday to Sunday is to set up 7 baselines respectively, still sets up working day, weekend
2 baselines.Concrete methods of realizing is, by doing T inspections two-by-two, and the equal sequence of sample average is merged into same monitoring sequence
Row, the unequal sequence of average includes supervisory sequence.
B2, by calculating Spearman (Spearman) coefficient R s of time and index, judges whether sequence has
Growth property non-stationary trend.If Rs>0.9, then difference is carried out to sequence and verify sequence newly-generated after difference and time
Whether Spearman coefficient Rs s>0.9, if it is, needing to continue difference to coefficient R s<Untill 0.9.Empirical tests,
Index system first-order difference of the present invention can meet demand.
Wherein, Spearman coefficient correlations calculation procedure is as follows, the first step:Time and index are compiled by rank order respectively
Number order is sought, draw the sequence X i and Yi (i=1,2 ... ..n) of two orders, the poor Di of each pair order row is obtained in order;Second step:
Calculate coefficient correlationWherein n is sample size, and when monitored object is without all fluctuation patterns, n is
Sliding window length L.For example when all rules, such as monitored object be Monday transaction stroke count, n be in L windows the date be Monday
Number of days;
B3, first-order difference, the sequence △ Xt using after difference calculate average and standard deviation as new supervisory sequence.
Further, gone through according to the historical interaction data of the first period in setting duration with setting second period in duration
History interaction data, determines the superposition factor of the monitoring baseline;According to the superposition factor, and the average and the standard
Difference, the monitoring baseline of first index is determined according to formula one, and the formula one is:
Wherein, λ is the superposition factor,For average of first index within period of waves, σ is the first index in fluctuation
Standard deviation in cycle, b is amplitude factor.
Wherein, the superposition factor of monitoring baseline can be with finger joint holiday factor lambda1Or moon factor lambda2, festivals or holidays λ1With the moon because
Sub- λ2Priority, defer to festivals or holidays λ1Priority be higher than moon factor lambda2, i.e., it is both for festivals or holidays and the beginning of the month when the monitoring same day
When (in, end), then baseline algorithm is adjusted toWhen it is only the beginning of the month (in, end) to monitor the same day, baseline algorithm
ForThe baseline algorithm of ordinary day isTo avoid the repeated assignment of values of Effects of Factors.
In addition, as can be seen from Fig. 3, threshold value b determination is taken and arrives the wide algorithm successively decreased step by step as strict as possible, b values since 3,Start, each iteration b=b-0.1, the abnormity point rate of monitored object is calculated, when abnormity point rate
>0%, it is determined that the b values of object are the threshold value of monitoring.
Further, according to the monitoring baseline determine abnormal data in the interaction data set of first index it
Afterwards, in addition to:
If the attribute of first index be absolute magnitude type, it is determined that the abnormal data of first index with it is described
Difference between value;
The traversal factor set of first index is analyzed in test, it is determined that being more than Second Threshold to the difference degree of making contributions
The first object factor, and generate the anomaly analysis related to the first object factor and report.
Wherein, to abnormal data judgement when unusual fluctuation detecting index exceedes index baseline bound as shown in figure 4, control water
It is flat, when being judged as abnormal, by calculating same day value with the incremental difference of history average in sliding window away from (point both positive and negative feelings
Condition), and each level value for each dimension that the increment is analyzed in reason test decomposed.Wherein, the exception to absolute magnitude is former
The decomposition of cause is as shown in figure 5, key step is as follows:
C1, travels through the determination of factor set:Default factor collection is determined before traversal by automatic algorithms, and factor set is carried out
Distribution statisticses, if the TOP1 of certain factor classification Distribution value accounting>More than k%, the factor does not enter traversal flow, and (K values are write from memory
Think 90);In addition, system provides self-defined selecting predictors window, the factor of user's voluntarily selection analysis is realized, and for expert
Analysis path automatically configures the traversal factor set of test.
C2, the determination of child node:A), automatic test analysis, calculates the increment absolute magnitude of each level value, then at 10-i
Treat that the dimension of test is traveled through;Obtain all level values under each dimensions of 10-i corresponding to the maximum level of father node contribution degree
Dimension is used as next layer of child node;B), to analysis expert path and User Defined analysis path, the determination of each level of child nodes
According to the order specified in advance.
C3, the generation of leaf:Descending arrangement selection accumulation contribution is carried out to the horizontal contribution degree under current node dimension
Degree reaches more than 90% or the leaf that is decomposed as this layer of level value of the horizontal contribution degree up to more than 10%;
C4, beta pruning:Solve how to improve the readability and validity of parsing tree, when leaf node is too thin, the tribute to root node
Degree of offering can not be summarized;Too thick when setting, it is too general to understand the reason for unusual fluctuation;Solution is:Every layer of leaf is calculated to " father saves
The contribution degree of " root node " of point ", by setting to father node contribution degree>20%, and to root node contribution degree>5% (threshold value ginseng
Numberization), then continue test, to improve the induction and conclusion of result.Because of the dimensional information that beta pruning is struck off, in displaying index
It has been shown that, and show the contribution degree of correlation as reference.
C5, the generation of parsing tree:According to 2-4 steps, test decomposition and beta pruning are being followed to all dimensions in factor set
Threshold value rule under, be all traversed and decompose, complete test analysis, and show generation parsing tree.
C6, accounting class index reason test logic is also to work as accounting class monitor control index to exceed baseline bound controlled level,
When being judged as abnormal, by calculating same day value with the incremental difference of history average in sliding window away from (point both positive and negative situation),
And each level value for each dimension for analyzing the increment in reason test is decomposed.
Further, if the attribute of first index is relative quantity type, it is determined that the abnormal number of first index
According to the ratio between the historical interaction data summation in the period of waves;
The traversal factor set of first index is analyzed in test, it is determined that the contribution degree made to the ratio is more than the 3rd threshold
Second target elements of value, and generate the anomaly analysis report related to second target elements.
Specifically, be with the main distinction of absolute figureofmerit desired value calculation it is different, the meter of accounting
Calculation mode is the ratio with the absolute magnitude of present level under current dimension and overall total amount, rather than the relative total amount with last layer level
Ratio.
Enter once, according to demands such as business, market, management level decision-makings, unusual fluctuation is being carried out to enterprise operation class KPI indexs
While monitoring, it is desirable to provide forecast function, to carry out anticipation, decision-making to business circumstance earlier.The method taken is main
It is to build year-on-year class index, using the moon as the monitoring cycle, in true sale data of the beginning of the month according to last month, monitors the same of last month
Whether the fluctuation than growth rate exceedes normal fluctuation range, when the monitored algorithm of fluctuation is determined as abnormal, automatically analyzes different
Often the reason for.Year-on-year growth rate data have time span length, totality and between part, month, different regions, different industries it
Between on year-on-year basis data difference it is larger the characteristics of.
Year-on-year class monitor control index abnormal cause test logic, by successively obtaining contribution degree traversal meter to the level value of each dimension
Ranking is calculated, using the corresponding dimension of contribution degree top ranked level value as the next layer of father node decomposed, successively test is decomposed, most
Analysis on Abnormal tree is obtained eventually, is that year-on-year data exception point is more with absolute magnitude and accounting index difference, jumping degree
It is larger, must personalisation process in data processing and the calculating of contribution degree.As shown in fig. 7, setting up dimension Indexes Abnormality baseline, that is, take
Each moon year-on-year growth rate of nearly 3 years monitored object, removes 1,2 months and year-on-year growth rate>100% abnormity point.Go after abnormity point
Carry out the calculating of following three parameters:2) E (X)=equal Data-Statistics;3) sigma (X) standard deviation statistics;4) CV (X)=sigma
(X)/E (X) coefficient of variation.Year-on-year index baseline sets up method:Year-on-year growth rate unusual fluctuation detecting module sets baseline algorithm, baseline
1 directly links up with business plan value up to standard, when less than annual plan value, triggers abnormal alarm, analyzes and flow into reason test
Journey;Baseline 2 calculates the normal range (NR) of fluctuation according to the fluctuating range of the history of monitored object, when monitored object level is higher than ripple
The dynamic upper limit, or less than monitoring lower limit when, abnormal alarm is triggered, into reason test analysis process.Need what is considered
It is, the processing of year-on-year Indexes Abnormality point:Influenceed by the Spring Festival, the jumping degree of January and the year-on-year growth rate of 2 months, calculating E (X)
The sample point of the two annual months is needed to reject during with sigma (X);Other Special section year-on-year growth rate of other part non-1,
Larger jump may also occur within 2 months, such sample point must be rejected.
In the figure 7, the calculating of contribution degree:For the dimension of unified each layer dimension horizontal contribution degree, be conducive to reducing each dimension
Level sets contribution degree calculating method as follows overall abnormal disturbance degree size:
To this layer of contribution degree:I-th dimension degree j level values obtain contribution degree=(the horizontal annual plan year-on-year growth rate * j levels of j are gone
Year same period absolute magnitude-j current absolute magnitude of level)/(i dimension plan year-on-year growth rate * i dimensions same period last year absolute magnitude-
The current absolute magnitude of i dimensions);
Contribution degree=(j level plan year-on-year growth rate * j level last years are obtained to overall contribution degree=i-th dimension degree j level values
Same period absolute magnitude-j current the absolute magnitude of level)/(monitored object plan year-on-year growth rate * monitoring same period last year absolute magnitude-
The current absolute magnitude of monitored object).
It can be seen that, found in bank card business dealing chain index system provided in an embodiment of the present invention sentences different processing example, meter
Calculate performance and there is bottleneck, abnormity point is more, test level deep, packet it is big in the case of, the actual effect of processing is difficult to meet business
Demand.To solve this problem, control is optimized in terms of two:(D1) analysis depth is controlled, and the depth capacity of parsing tree is
The number of reason test index, in instance analysis, the universal phenomenon reflected, with the increasing of depth, the contribution of leaf node
Degree is more obvious to the scattered trend of contribution degree of this layer, to overall contribution degree often<10%, anticipated to summarizing abnormal main cause
It is adopted less, be this by setting threshold value to control the depth of parsing tree:1) choosing the number n principles of leaf node is, this n leaf node
To the accumulative contribution degree of father node>90%, and each leaf node is to the contribution degree of father node>10%, when contribution degree under father node
The contribution degree of maximum leaf node<10%, choose contribution degree ranking and be shown in for TOP4 4 leaves on parsing tree.2) leaf segment
Point is to father node contribution degree>20%, the test analysis of next level of child nodes need to be carried out.3) each dimension detailed data is led under leaf node
The embodiment of displaying index is crossed, each node layer window can be supported to jump out displaying.4) new business is in dimensions such as type of transaction, channel, areas
With higher concentration degree, in the initial several layers of reason tests of index, the key dimension substantially elected all is classification number
(level value) less dimension is measured, in 1-3 layers of contribution degree are calculated, the contribution of the certain level value under multiple dimensions often occurs
Degree>90%, even more than up to 100%, so the traversal of n-i (i=0,1 ... .10) each dimension contribution degree is successively carried out, can be with
Optimize, if that is, setting is when traveling through for i-th layer, contributed if there is some level value of K dimension in new traffic module
Degree>=90%, then when calculating contribution degree for next layer, using each dimensions of K all as screening conditions, only remaining n-i-k is respectively tieed up
Degree carries out traversal and calculates contribution degree, to provide the efficiency of calculating.
Based on the abnormity point that anomalous mode block is generated is sentenced, this (is drawn by performance test with 2G, 4G with size of data filter
Critical value) packet is divided into three and applies processing pond by boundary's point, and the resource to rear end also corresponds to configuration 50G, 100G, 200G respectively
Mode configured.Thus, tasks in parallel can be initiated, as shown in figure 8, greatly improving operational efficiency, solves performance
With the bottleneck of resource.
Based on identical technical concept, the embodiment of the present invention also provides a kind of device of Monitoring Data, and the device can perform
Above method embodiment.Device provided in an embodiment of the present invention as shown in figure 9, including:Acquiring unit 401, division unit 402,
Determining unit 403, sentence anticoincidence unit 404, wherein:
Acquiring unit 401, the history for obtaining each monitoring object in setting duration in trade link interacts number
According to;
Division unit 402, the history for the historical interaction data of each monitoring object to be divided into N number of index interacts number
According to set;
Determining unit 403, for the historical interaction data set for first index, performs following operation, wherein,
First index is any one index in N number of index:Determine the historical interaction data collection of first index
In the period of waves of conjunction and the average of the historical interaction data in the period of waves, and the period of waves of first index
Historical interaction data standard deviation;According to the average and the standard deviation, the monitoring baseline of first index, institute are determined
State the fluctuation range that monitoring baseline is used to indicate the normal historical interaction data of first index;
Sentence anticoincidence unit 404, it is different in the interaction data set of first index for being determined according to the monitoring baseline
Regular data.
Further, the determining unit 403 specifically for:
T inspections are carried out to the data in the historical interaction data set of first index, K difference averages are determined
Sequence;Calculate the statistical correlation coefficient between K sequence and time attribute;Determine that the statistical correlation coefficient is more than first threshold
When, difference is carried out to the K sequence, until differentiated statistical correlation coefficient is not more than the first threshold;Calculate difference
The average of sequence afterwards, and it regard the average of the differentiated sequence as the equal of the historical interaction data in the period of waves
Value.
Further, the determining unit 403 is additionally operable to:
Number is interacted with the history of the second period in setting duration according to the historical interaction data of the first period in setting duration
According to, determine it is described monitoring baseline the superposition factor;
According to the superposition factor, and the average and the standard deviation, first index is determined according to formula one
Monitoring baseline, the formula one is:
Wherein, λ is the superposition factor,For average of first index within period of waves, σ is the first index in fluctuation
Standard deviation in cycle, b is amplitude factor.
Further, the determining unit 403 is additionally operable to:If the attribute of first index is absolute magnitude type, really
Difference between the abnormal data and the average of fixed first index;
Described device also includes:First positioning unit 405, the traversal factor set of first index is analyzed for test,
It is determined that being more than the first object factor of Second Threshold to the difference degree of making contributions, and generate and the first object factor phase
The anomaly analysis report of pass.
Further, the determining unit 403 is additionally operable to:
If the attribute of first index is relative quantity type, it is determined that the abnormal data of first index and the ripple
The ratio between historical interaction data summation in the dynamic cycle;
Described device also includes:Second positioning unit 406, the traversal factor set of first index is analyzed for test,
It is determined that the contribution degree made to the ratio is more than the second target elements of the 3rd threshold value, and generate and second target elements
Related anomaly analysis report.
In summary, the embodiment of the present invention is by obtaining the interaction data of N number of index of the object to be monitored of certain time length,
Then the data of acquisition are obtained the friendship of each index by usually one data more than week according to operational indicator classified types
Mutual data acquisition system, so as to determine the abnormal shape of data in the interaction data set of the index using the corresponding index baseline of the index
State, because index baseline determines the fluctuation range of the first index of monitored object, so when the interaction data of index is higher than finger
Mark baseline the upper limit, or less than index baseline lower limit when, can trigger abnormal alarm.Wherein, index baseline is according to finger
What the average and standard deviation of mark data set were determined, because the collection period of object to be monitored is longer with respect to real time data, and refer to
Mark baseline is also the rule obtained according to history historical interaction data analysis interior for a period of time, therefore so carries out anomalous discrimination
Operation, accuracy is higher, reduces the probability of erroneous judgement.The method of Monitoring Data provided in an embodiment of the present invention has filled up bank card
Merchandise full link index system missing;Propose the means by business, technology dual combination, solve based on Hadoop,
The problem of big data such as Hive, impala treatment technology is to single-point big data bag, multi-layer analytical performance deficiency.Realize bank
The rules self-adaptive of the index Design of each key link and Enterprise operation index, reason automation are detectd on the full link of card transaction
Survey processing system.This method is not limited to bank card business dealing field, applicable in multiple fields such as finance, manufacture, services.
The present invention is the flow with reference to method according to embodiments of the present invention, equipment (system) and computer program product
Figure and/or block diagram are described.It should be understood that can be by every first-class in computer program instructions implementation process figure and/or block diagram
Journey and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computer programs can be provided
The processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices is instructed to produce
A raw machine so that produced by the instruction of computer or the computing device of other programmable data processing devices for real
The device for the function of being specified in present one flow of flow chart or one square frame of multiple flows and/or block diagram or multiple square frames.
These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which is produced, to be included referring to
Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or
The function of being specified in multiple square frames.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that in meter
Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, thus in computer or
The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in individual square frame or multiple square frames.
, but those skilled in the art once know basic creation although preferred embodiments of the present invention have been described
Property concept, then can make other change and modification to these embodiments.So, appended claims are intended to be construed to include excellent
Select embodiment and fall into having altered and changing for the scope of the invention.
Obviously, those skilled in the art can carry out the essence of various changes and modification without departing from the present invention to the present invention
God and scope.So, if these modifications and variations of the present invention belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising including these changes and modification.
Claims (10)
1. a kind of method of Monitoring Data, it is characterised in that this method includes:
Obtain the historical interaction data of each monitoring object in setting duration in trade link;
The historical interaction data of each monitoring object is divided into the historical interaction data set of N number of index;
For the historical interaction data set of first index, following operation is performed, wherein, first index is the N
Any one index in individual index:
Determine first index historical interaction data set period of waves and the period of waves in history interaction number
According to average, and the historical interaction data in the period of waves of first index standard deviation;
According to the average and the standard deviation, the monitoring baseline of first index is determined, the monitoring baseline is used to indicate
The fluctuation range of the normal historical interaction data of first index;
The abnormal data in the interaction data set of first index is determined according to the monitoring baseline.
2. the method as described in claim 1, it is characterised in that historical interaction data in the determination period of waves
Average, including:
T inspections are carried out to the data in the historical interaction data set of first index, the sequence of K different averages is determined
Row;
Calculate the statistical correlation coefficient between K sequence and time attribute;
When determining that the statistical correlation coefficient is more than first threshold, difference is carried out to the K sequence, until differentiated statistics
Coefficient correlation is not more than the first threshold;
The average of differentiated sequence is calculated, and regard the average of the differentiated sequence as the history in the period of waves
The average of interaction data.
3. method as claimed in claim 1 or 2, it is characterised in that also include:
Historical interaction data according to the historical interaction data of the first period in setting duration with setting the second period in duration, really
The superposition factor of the fixed monitoring baseline;
It is described that the monitoring baseline of first index is determined according to the average and the standard deviation, including:
According to the superposition factor, and the average and the standard deviation, the prison of first index is determined according to formula one
Baseline is surveyed, the formula one is:
Wherein, λ is the superposition factor,For average of first index within period of waves, σ is the first index in period of waves
Interior standard deviation, b is amplitude factor.
4. the method as described in any one of claim 1 to 2, it is characterised in that described according to being determined the monitoring baseline
After abnormal data in the interaction data set of first index, in addition to:
If the attribute of first index is absolute magnitude type, it is determined that the abnormal data of first index and the average it
Between difference;
The traversal factor set of first index is analyzed in test, it is determined that being more than the of Second Threshold to the difference degree of making contributions
One target elements, and generate the anomaly analysis report related to the first object factor.
5. the method as described in claim 1, it is characterised in that described that first index is determined according to the monitoring baseline
After abnormal data in interaction data set, in addition to:
If the attribute of first index is relative quantity type, it is determined that the abnormal data of first index and the fluctuation week
The ratio between historical interaction data summation in phase;
The traversal factor set of first index is analyzed in test, it is determined that the contribution degree made to the ratio is more than the 3rd threshold value
Second target elements, and generate the anomaly analysis report related to second target elements.
6. a kind of device of Monitoring Data, it is characterised in that the device includes:
Acquiring unit, the historical interaction data for obtaining each monitoring object in setting duration in trade link;
Division unit, the historical interaction data set for the historical interaction data of each monitoring object to be divided into N number of index;
Determining unit, for the historical interaction data set for first index, performs following operation, wherein, described the
One index is any one index in N number of index:Determine the ripple of the historical interaction data set of first index
Dynamic cycle and the average of the historical interaction data in the period of waves, and the history in the period of waves of first index
The standard deviation of interaction data;According to the average and the standard deviation, the monitoring baseline of first index, the monitoring are determined
Baseline is used for the fluctuation range for indicating the normal historical interaction data of first index;
Anticoincidence unit is sentenced, for determining the abnormal data in the interaction data set of first index according to the monitoring baseline.
7. device as claimed in claim 6, it is characterised in that the determining unit specifically for:
T inspections are carried out to the data in the historical interaction data set of first index, the sequence of K different averages is determined
Row;Calculate the statistical correlation coefficient between K sequence and time attribute;Determine that the statistical correlation coefficient is more than first threshold
When, difference is carried out to the K sequence, until differentiated statistical correlation coefficient is not more than the first threshold;Calculate difference
The average of sequence afterwards, and it regard the average of the differentiated sequence as the equal of the historical interaction data in the period of waves
Value.
8. device as claimed in claims 6 or 7, it is characterised in that the determining unit is additionally operable to:
Historical interaction data according to the historical interaction data of the first period in setting duration with setting the second period in duration, really
The superposition factor of the fixed monitoring baseline;
According to the superposition factor, and the average and the standard deviation, the prison of first index is determined according to formula one
Baseline is surveyed, the formula one is:
Wherein, λ is the superposition factor,For average of first index within period of waves, σ is the first index in period of waves
Interior standard deviation, b is amplitude factor.
9. the device as described in any one of claim 6 to 7, it is characterised in that the determining unit is additionally operable to:If described first
The attribute of index is absolute magnitude type, it is determined that the difference between the abnormal data and the average of first index;
Described device also includes:First positioning unit, the traversal factor set of first index is analyzed for test, it is determined that to institute
The first object factor that difference degree of making contributions is more than Second Threshold is stated, and generates the exception related to the first object factor
Analysis report.
10. device as claimed in claim 6, it is characterised in that the determining unit is additionally operable to:
If the attribute of first index is relative quantity type, it is determined that the abnormal data of first index and the fluctuation week
The ratio between historical interaction data summation in phase;
Described device also includes:Second positioning unit, the traversal factor set of first index is analyzed for test, it is determined that to institute
State the contribution degree that ratio makes and be more than the second target elements of the 3rd threshold value, and generate related to second target elements different
Normal analysis report.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710178551.0A CN106991145B (en) | 2017-03-23 | 2017-03-23 | Data monitoring method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710178551.0A CN106991145B (en) | 2017-03-23 | 2017-03-23 | Data monitoring method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106991145A true CN106991145A (en) | 2017-07-28 |
CN106991145B CN106991145B (en) | 2021-03-23 |
Family
ID=59411781
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710178551.0A Active CN106991145B (en) | 2017-03-23 | 2017-03-23 | Data monitoring method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106991145B (en) |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108228428A (en) * | 2018-02-05 | 2018-06-29 | 百度在线网络技术(北京)有限公司 | For the method and apparatus of output information |
CN108449231A (en) * | 2018-03-15 | 2018-08-24 | 华青融天(北京)技术股份有限公司 | A kind of filter method of transaction data, device and realization device |
CN108718303A (en) * | 2018-05-09 | 2018-10-30 | 北京仁和诚信科技有限公司 | Safe operation management method and system |
CN108775914A (en) * | 2018-05-07 | 2018-11-09 | 青岛海信网络科技股份有限公司 | A kind of transit equipment detection method and detection device |
CN108829638A (en) * | 2018-06-01 | 2018-11-16 | 阿里巴巴集团控股有限公司 | A kind of business datum fluctuation processing method and processing device |
CN108923996A (en) * | 2018-05-11 | 2018-11-30 | 中国银联股份有限公司 | A kind of capacity analysis method and device |
CN109034252A (en) * | 2018-08-01 | 2018-12-18 | 中国科学院大气物理研究所 | The automatic identification method of air quality website monitoring data exception |
CN109241043A (en) * | 2018-08-13 | 2019-01-18 | 蜜小蜂智慧(北京)科技有限公司 | A kind of data quality checking method and device |
CN109635265A (en) * | 2018-11-29 | 2019-04-16 | 济南荣耀合创电力科技有限公司 | A kind of test report generation system based on image recognition |
CN109634997A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | A kind of acquisition methods, device and the electronic equipment of unusual fluctuation channel |
CN109947713A (en) * | 2017-10-31 | 2019-06-28 | 北京国双科技有限公司 | A kind of monitoring method and device of log |
WO2019218432A1 (en) * | 2018-05-14 | 2019-11-21 | 平安科技(深圳)有限公司 | Abnormal cross-border transaction determination method and apparatus, terminal and storage medium |
CN110784355A (en) * | 2019-10-30 | 2020-02-11 | 网宿科技股份有限公司 | Fault identification method and device |
CN110990242A (en) * | 2019-11-29 | 2020-04-10 | 上海观安信息技术股份有限公司 | Method and device for determining fluctuation abnormity of user operation times |
CN111047125A (en) * | 2018-10-11 | 2020-04-21 | 鸿富锦精密电子(成都)有限公司 | Product failure analysis device, method and computer readable storage medium |
CN111191881A (en) * | 2019-12-13 | 2020-05-22 | 大唐东北电力试验研究院有限公司 | Thermal power generating unit industrial equipment state monitoring method based on big data |
CN111209165A (en) * | 2020-01-05 | 2020-05-29 | 光大兴陇信托有限责任公司 | Two-stage monitoring processing method based on channel |
CN111290916A (en) * | 2020-02-18 | 2020-06-16 | 深圳前海微众银行股份有限公司 | Big data monitoring method, device and equipment and computer readable storage medium |
CN111899040A (en) * | 2019-05-05 | 2020-11-06 | 腾讯科技(深圳)有限公司 | Method, device and equipment for detecting abnormal propagation of target object and storage medium |
CN112037050A (en) * | 2020-09-03 | 2020-12-04 | 中国银行股份有限公司 | Transaction data monitoring method, device and equipment |
CN112597144A (en) * | 2020-12-29 | 2021-04-02 | 农业农村部环境保护科研监测所 | Automatic cleaning method for production area environment monitoring data |
CN112801345A (en) * | 2021-01-07 | 2021-05-14 | 山东润一智能科技有限公司 | Equipment measuring point time interval early warning method and system based on expectation and fluctuation |
CN113067747A (en) * | 2021-03-15 | 2021-07-02 | 中国工商银行股份有限公司 | Link abnormity tracing method, cluster, node and system |
CN114492529A (en) * | 2022-01-27 | 2022-05-13 | 中国汽车工程研究院股份有限公司 | Power battery system connection abnormity fault safety early warning method |
CN114978863A (en) * | 2022-05-17 | 2022-08-30 | 安天科技集团股份有限公司 | Data processing method and device, computer equipment and readable storage medium |
WO2023019560A1 (en) * | 2021-08-20 | 2023-02-23 | 京东方科技集团股份有限公司 | Data processing method and apparatus, electronic device and computer-readable storage medium |
CN117198031A (en) * | 2023-11-03 | 2023-12-08 | 浙江华东岩土勘察设计研究院有限公司 | Platform state monitoring and early warning method based on security envelope strategy |
CN117874409A (en) * | 2024-03-11 | 2024-04-12 | 榕湾科技(成都)有限公司 | Water production feedback regulation and control method, system, terminal and medium based on water quality prediction |
CN117874409B (en) * | 2024-03-11 | 2024-05-31 | 榕湾科技(成都)有限公司 | Water production feedback regulation and control method, system, terminal and medium based on water quality prediction |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103036741A (en) * | 2012-12-19 | 2013-04-10 | 北京神州绿盟信息安全科技股份有限公司 | Determination method and determination device of flow monitoring base line |
CN103236953A (en) * | 2012-10-30 | 2013-08-07 | 吉林大学 | Active monitoring method for IP (internet protocol) bearer network performance indexes based on fuzzy time series prediction model |
CN103365969A (en) * | 2013-06-24 | 2013-10-23 | 北京奇虎科技有限公司 | Abnormal data detecting and processing method and system |
CN103532940A (en) * | 2013-09-30 | 2014-01-22 | 广东电网公司电力调度控制中心 | Network security detection method and device |
CN104598361A (en) * | 2013-10-31 | 2015-05-06 | 华为技术有限公司 | Performance monitoring method and device |
CN105049291A (en) * | 2015-08-20 | 2015-11-11 | 广东睿江科技有限公司 | Method for detecting network traffic anomaly |
WO2016073379A2 (en) * | 2014-11-03 | 2016-05-12 | Vectra Networks, Inc. | A system for implementing threat detection using daily network traffic community outliers |
CN105589796A (en) * | 2014-12-31 | 2016-05-18 | 中国银联股份有限公司 | Method for monitoring information interaction data anomalies |
CN105654381A (en) * | 2015-12-28 | 2016-06-08 | 上海瀚银信息技术有限公司 | Predicting system for business transaction volume |
CN105678414A (en) * | 2015-12-31 | 2016-06-15 | 远光软件股份有限公司 | Data processing method of predicting resource consumption |
CN105743720A (en) * | 2014-12-08 | 2016-07-06 | 中国移动通信集团设计院有限公司 | Link quality assessment method and device |
CN106202389A (en) * | 2016-07-08 | 2016-12-07 | 中国银联股份有限公司 | A kind of method for monitoring abnormality based on transaction data and device |
CN106371092A (en) * | 2016-08-25 | 2017-02-01 | 中国科学院国家授时中心 | Deformation monitoring method based on GPS and strong-motion seismograph observation adaptive combination |
CN106368816A (en) * | 2016-10-27 | 2017-02-01 | 中国船舶工业系统工程研究院 | Method for online abnormity detection of low-speed diesel engine of ship based on baseline deviation |
-
2017
- 2017-03-23 CN CN201710178551.0A patent/CN106991145B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103236953A (en) * | 2012-10-30 | 2013-08-07 | 吉林大学 | Active monitoring method for IP (internet protocol) bearer network performance indexes based on fuzzy time series prediction model |
CN103036741A (en) * | 2012-12-19 | 2013-04-10 | 北京神州绿盟信息安全科技股份有限公司 | Determination method and determination device of flow monitoring base line |
CN103365969A (en) * | 2013-06-24 | 2013-10-23 | 北京奇虎科技有限公司 | Abnormal data detecting and processing method and system |
CN103532940A (en) * | 2013-09-30 | 2014-01-22 | 广东电网公司电力调度控制中心 | Network security detection method and device |
CN104598361A (en) * | 2013-10-31 | 2015-05-06 | 华为技术有限公司 | Performance monitoring method and device |
WO2016073379A2 (en) * | 2014-11-03 | 2016-05-12 | Vectra Networks, Inc. | A system for implementing threat detection using daily network traffic community outliers |
CN105743720A (en) * | 2014-12-08 | 2016-07-06 | 中国移动通信集团设计院有限公司 | Link quality assessment method and device |
CN105589796A (en) * | 2014-12-31 | 2016-05-18 | 中国银联股份有限公司 | Method for monitoring information interaction data anomalies |
CN105049291A (en) * | 2015-08-20 | 2015-11-11 | 广东睿江科技有限公司 | Method for detecting network traffic anomaly |
CN105654381A (en) * | 2015-12-28 | 2016-06-08 | 上海瀚银信息技术有限公司 | Predicting system for business transaction volume |
CN105678414A (en) * | 2015-12-31 | 2016-06-15 | 远光软件股份有限公司 | Data processing method of predicting resource consumption |
CN106202389A (en) * | 2016-07-08 | 2016-12-07 | 中国银联股份有限公司 | A kind of method for monitoring abnormality based on transaction data and device |
CN106371092A (en) * | 2016-08-25 | 2017-02-01 | 中国科学院国家授时中心 | Deformation monitoring method based on GPS and strong-motion seismograph observation adaptive combination |
CN106368816A (en) * | 2016-10-27 | 2017-02-01 | 中国船舶工业系统工程研究院 | Method for online abnormity detection of low-speed diesel engine of ship based on baseline deviation |
Cited By (44)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109947713A (en) * | 2017-10-31 | 2019-06-28 | 北京国双科技有限公司 | A kind of monitoring method and device of log |
CN109947713B (en) * | 2017-10-31 | 2021-08-10 | 北京国双科技有限公司 | Log monitoring method and device |
CN108228428A (en) * | 2018-02-05 | 2018-06-29 | 百度在线网络技术(北京)有限公司 | For the method and apparatus of output information |
CN108449231A (en) * | 2018-03-15 | 2018-08-24 | 华青融天(北京)技术股份有限公司 | A kind of filter method of transaction data, device and realization device |
CN108449231B (en) * | 2018-03-15 | 2020-07-07 | 华青融天(北京)软件股份有限公司 | Transaction data filtering method and device and implementation device |
CN108775914A (en) * | 2018-05-07 | 2018-11-09 | 青岛海信网络科技股份有限公司 | A kind of transit equipment detection method and detection device |
CN108775914B (en) * | 2018-05-07 | 2020-09-22 | 青岛海信网络科技股份有限公司 | Traffic equipment detection method and detection equipment |
CN108718303B (en) * | 2018-05-09 | 2021-03-23 | 北京仁和诚信科技有限公司 | Safe operation and maintenance management method and system |
CN108718303A (en) * | 2018-05-09 | 2018-10-30 | 北京仁和诚信科技有限公司 | Safe operation management method and system |
CN108923996B (en) * | 2018-05-11 | 2021-01-05 | 中国银联股份有限公司 | Capacity analysis method and device |
CN108923996A (en) * | 2018-05-11 | 2018-11-30 | 中国银联股份有限公司 | A kind of capacity analysis method and device |
WO2019218432A1 (en) * | 2018-05-14 | 2019-11-21 | 平安科技(深圳)有限公司 | Abnormal cross-border transaction determination method and apparatus, terminal and storage medium |
CN108829638A (en) * | 2018-06-01 | 2018-11-16 | 阿里巴巴集团控股有限公司 | A kind of business datum fluctuation processing method and processing device |
CN109034252A (en) * | 2018-08-01 | 2018-12-18 | 中国科学院大气物理研究所 | The automatic identification method of air quality website monitoring data exception |
CN109034252B (en) * | 2018-08-01 | 2020-10-30 | 中国科学院大气物理研究所 | Automatic identification method for monitoring data abnormity of air quality station |
CN109241043A (en) * | 2018-08-13 | 2019-01-18 | 蜜小蜂智慧(北京)科技有限公司 | A kind of data quality checking method and device |
CN111047125A (en) * | 2018-10-11 | 2020-04-21 | 鸿富锦精密电子(成都)有限公司 | Product failure analysis device, method and computer readable storage medium |
CN111047125B (en) * | 2018-10-11 | 2023-11-14 | 鸿富锦精密电子(成都)有限公司 | Product failure analysis apparatus, method, and computer-readable storage medium |
CN109634997A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | A kind of acquisition methods, device and the electronic equipment of unusual fluctuation channel |
CN109635265A (en) * | 2018-11-29 | 2019-04-16 | 济南荣耀合创电力科技有限公司 | A kind of test report generation system based on image recognition |
CN111899040B (en) * | 2019-05-05 | 2023-09-01 | 腾讯科技(深圳)有限公司 | Method, device, equipment and storage medium for detecting target object abnormal propagation |
CN111899040A (en) * | 2019-05-05 | 2020-11-06 | 腾讯科技(深圳)有限公司 | Method, device and equipment for detecting abnormal propagation of target object and storage medium |
CN110784355B (en) * | 2019-10-30 | 2022-03-08 | 网宿科技股份有限公司 | Fault identification method and device |
CN110784355A (en) * | 2019-10-30 | 2020-02-11 | 网宿科技股份有限公司 | Fault identification method and device |
CN110990242B (en) * | 2019-11-29 | 2023-06-20 | 上海观安信息技术股份有限公司 | Method and device for determining fluctuation abnormality of user operation times |
CN110990242A (en) * | 2019-11-29 | 2020-04-10 | 上海观安信息技术股份有限公司 | Method and device for determining fluctuation abnormity of user operation times |
CN111191881B (en) * | 2019-12-13 | 2024-05-14 | 大唐东北电力试验研究院有限公司 | Thermal power generating unit industrial equipment state monitoring method based on big data |
CN111191881A (en) * | 2019-12-13 | 2020-05-22 | 大唐东北电力试验研究院有限公司 | Thermal power generating unit industrial equipment state monitoring method based on big data |
CN111209165B (en) * | 2020-01-05 | 2021-03-16 | 光大兴陇信托有限责任公司 | Two-stage monitoring processing method based on channel |
CN111209165A (en) * | 2020-01-05 | 2020-05-29 | 光大兴陇信托有限责任公司 | Two-stage monitoring processing method based on channel |
CN111290916A (en) * | 2020-02-18 | 2020-06-16 | 深圳前海微众银行股份有限公司 | Big data monitoring method, device and equipment and computer readable storage medium |
CN112037050A (en) * | 2020-09-03 | 2020-12-04 | 中国银行股份有限公司 | Transaction data monitoring method, device and equipment |
CN112597144B (en) * | 2020-12-29 | 2022-11-08 | 农业农村部环境保护科研监测所 | Automatic cleaning method for production place environment monitoring data |
CN112597144A (en) * | 2020-12-29 | 2021-04-02 | 农业农村部环境保护科研监测所 | Automatic cleaning method for production area environment monitoring data |
CN112801345A (en) * | 2021-01-07 | 2021-05-14 | 山东润一智能科技有限公司 | Equipment measuring point time interval early warning method and system based on expectation and fluctuation |
CN113067747A (en) * | 2021-03-15 | 2021-07-02 | 中国工商银行股份有限公司 | Link abnormity tracing method, cluster, node and system |
WO2023019560A1 (en) * | 2021-08-20 | 2023-02-23 | 京东方科技集团股份有限公司 | Data processing method and apparatus, electronic device and computer-readable storage medium |
CN114492529A (en) * | 2022-01-27 | 2022-05-13 | 中国汽车工程研究院股份有限公司 | Power battery system connection abnormity fault safety early warning method |
CN114978863A (en) * | 2022-05-17 | 2022-08-30 | 安天科技集团股份有限公司 | Data processing method and device, computer equipment and readable storage medium |
CN114978863B (en) * | 2022-05-17 | 2024-03-01 | 安天科技集团股份有限公司 | Data processing method, device, computer equipment and readable storage medium |
CN117198031A (en) * | 2023-11-03 | 2023-12-08 | 浙江华东岩土勘察设计研究院有限公司 | Platform state monitoring and early warning method based on security envelope strategy |
CN117198031B (en) * | 2023-11-03 | 2024-01-26 | 浙江华东岩土勘察设计研究院有限公司 | Platform state monitoring and early warning method based on security envelope strategy |
CN117874409A (en) * | 2024-03-11 | 2024-04-12 | 榕湾科技(成都)有限公司 | Water production feedback regulation and control method, system, terminal and medium based on water quality prediction |
CN117874409B (en) * | 2024-03-11 | 2024-05-31 | 榕湾科技(成都)有限公司 | Water production feedback regulation and control method, system, terminal and medium based on water quality prediction |
Also Published As
Publication number | Publication date |
---|---|
CN106991145B (en) | 2021-03-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106991145A (en) | A kind of method and device of Monitoring Data | |
CN106951984A (en) | A kind of dynamic analyzing and predicting method of system health degree and device | |
CN108647891A (en) | Data exception classification, Reasons method and device | |
Billari et al. | Timing, sequencing, and quantum of life course events: A machine learning approach | |
US20030139957A1 (en) | Method of rule constrained statistical pattern recognition | |
WO2013043686A1 (en) | Methods and systems for assessing data quality | |
CN110069551A (en) | Medical Devices O&M information excavating analysis system and its application method based on Spark | |
CN108491991A (en) | Constraints analysis system based on the industrial big data product duration and method | |
CN108241900A (en) | Engineering project construction period prediction method, device and system | |
Oduro et al. | Do digital technologies pay off? A meta-analytic review of the digital technologies/firm performance nexus | |
CN113742118B (en) | Method and system for detecting anomalies in data pipes | |
CN103499663B (en) | A kind of system of selection based on sensor in the Longjing tea Quality Detection Grade Model of genetic algorithm | |
CN113835947B (en) | Method and system for determining abnormality cause based on abnormality recognition result | |
Wang et al. | Multi-objective parallel robotic dispensing planogram optimisation using association rule mining and evolutionary algorithms | |
CN113342939A (en) | Data quality monitoring method and device and related equipment | |
EP2365674A1 (en) | Quality-driven optimization of sensor stream processing | |
Yu et al. | Long-term trade impact of epidemic outbreaks: Is it V-shaped? | |
CN116703455A (en) | Medicine data sales prediction method and system based on time series hybrid model | |
Zhou et al. | Performance evaluation method for network monitoring based on separable temporal exponential random graph models with application to the study of autocorrelation effects | |
CN116203352A (en) | Fault early warning method, device, equipment and medium for power distribution network | |
Kliangkhlao et al. | Harnessing the power of big data digitization for market factors awareness in supply chain management | |
CN115858606A (en) | Method, device and equipment for detecting abnormity of time series data and storage medium | |
Boguhn | Forecasting power consumption of manufacturing industries using neural networks | |
Abdolbaghi Ataabadi et al. | The effectiveness of the automatic system of fuzzy logic-based technical patterns recognition: Evidence from Tehran stock exchange | |
Yu | Research and prediction of ecological footprint using machine learning: A case study of China |
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 |