CN105447323A - Data abnormal fluctuations detecting method and apparatus - Google Patents
Data abnormal fluctuations detecting method and apparatus Download PDFInfo
- Publication number
- CN105447323A CN105447323A CN201510920803.3A CN201510920803A CN105447323A CN 105447323 A CN105447323 A CN 105447323A CN 201510920803 A CN201510920803 A CN 201510920803A CN 105447323 A CN105447323 A CN 105447323A
- Authority
- CN
- China
- Prior art keywords
- index data
- fluctuation
- dimension
- data
- abnormal fluctuation
- 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.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 292
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000004458 analytical method Methods 0.000 claims abstract description 143
- 238000001514 detection method Methods 0.000 claims abstract description 44
- 238000007405 data analysis Methods 0.000 claims description 5
- 125000004122 cyclic group Chemical group 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 8
- 238000003066 decision tree Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000013178 mathematical model Methods 0.000 description 3
- 241001347978 Major minor Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000014510 cooky Nutrition 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention discloses a data abnormal fluctuation detecting method and apparatus. The method comprises: obtaining to-be-detected data, wherein the to-be-detected data comprises at least one piece of index data; according to a data abnormal fluctuation detecting method, determining at least one piece of abnormal fluctuation index data in the at least one piece of index data; and carrying out analysis on the abnormal fluctuation index data respectively according to corresponding preset dimension information, and generating a corresponding abnormal fluctuation analysis result, the preset dimension information comprises dimensions of the abnormal fluctuation index data, and at least one dimension index data corresponding to each dimension. According to the technical scheme provided by embodiments of the present invention, technical problems in the prior art that abnormal fluctuation detection and analysis time is long, a cost is high and an accuracy rate is low due to a data analyst determining whether each index data in the to-be-detected data occurs abnormal fluctuation and analyzing causes of abnormal fluctuation are solved, and efficiency of abnormal fluctuation detection accuracy and abnormal fluctuation cause analysis is improved.
Description
Technical Field
The embodiment of the invention relates to the technical field of data detection, in particular to a method and a device for detecting abnormal fluctuation of data.
Background
With the development of information technology, various industries generate a large amount of data every day in the operation process, for example, some report data, which is a direction with great research value for detecting whether the data are abnormally fluctuated and analyzing the reason of the abnormal fluctuation.
At present, data analysts usually rely on experience to determine whether each index in data has abnormal fluctuation and analyze the reason of the abnormal fluctuation, so that the time for detecting the data abnormal fluctuation is long, the accuracy is low, and under the condition of new data abnormal fluctuation, the data analysts are difficult to quickly locate the reason of the data abnormal fluctuation by experience, and the efficiency of locating the reason of the data abnormal fluctuation is also reduced.
Disclosure of Invention
The embodiment of the invention provides a method and a device for detecting abnormal data fluctuation, which are used for improving the accuracy of the abnormal data fluctuation detection and the efficiency of analyzing the reason of the abnormal data fluctuation.
In a first aspect, an embodiment of the present invention provides a data abnormal fluctuation detection method, including:
acquiring data to be detected, wherein the data to be detected comprises at least one index data;
determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method;
and analyzing the abnormal fluctuation index data according to corresponding preset dimension information respectively, and generating corresponding abnormal fluctuation analysis results, wherein the preset dimension information comprises the dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
In a second aspect, an embodiment of the present invention further provides a data abnormal fluctuation detection apparatus, including:
the data acquisition module to be detected is used for acquiring data to be detected, wherein the data to be detected comprises at least one index data;
the abnormal fluctuation index data determining module is used for determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method;
and the abnormal fluctuation index data analysis module is used for analyzing the abnormal fluctuation index data according to corresponding preset dimension information respectively and generating corresponding abnormal fluctuation analysis results, wherein the preset dimension information comprises the dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
According to the embodiment of the invention, at least one abnormal fluctuation index data is automatically determined by at least one index data in the data to be detected according to a preset data fluctuation detection method, and the abnormal fluctuation index data is respectively analyzed according to corresponding preset dimension information to generate a corresponding abnormal fluctuation analysis result.
Drawings
Fig. 1 is a flowchart of a data abnormal fluctuation detection method according to an embodiment of the present invention;
fig. 2 is a flowchart of a data abnormal fluctuation detection method according to a second embodiment of the present invention;
fig. 3 is a schematic structural diagram of an index abnormal fluctuation analysis tree corresponding to UV index data according to a second embodiment of the present invention;
FIG. 4 is a diagram illustrating an abnormal fluctuation analysis report according to a second embodiment of the present invention;
fig. 5 is a schematic structural diagram of an index abnormal fluctuation analysis integrated tree T according to a second embodiment of the present invention;
fig. 6 is a schematic flow chart of an abnormal data fluctuation detection logic according to a third embodiment of the present invention;
fig. 7 is a schematic structural diagram of a data abnormal fluctuation detection apparatus according to a fourth embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the invention and are not limiting of the invention. It should be further noted that, for the convenience of description, only some of the structures related to the present invention are shown in the drawings, not all of the structures.
Example one
Fig. 1 is a flowchart of a data abnormal fluctuation detection method according to an embodiment of the present invention, where the method of this embodiment may be executed by a data abnormal fluctuation detection apparatus, the apparatus may be implemented in a hardware and/or software manner, and the apparatus may be built in a server as a part of the server.
As shown in fig. 1, the method for detecting data abnormal fluctuation provided by the embodiment of the present invention specifically includes:
s101, data to be detected are obtained, and the data to be detected comprise at least one index data.
The data to be detected in this embodiment may be various report data, for example, Key Performance Indicators (KPIs) data, various business data, and the like. Specifically, the data to be detected can be acquired from the database. The index data is generated by dividing data representing different attributes in the data to be detected.
S102, determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method.
The preset data fluctuation detection method may be a preset algorithm for detecting whether the index data is abnormal fluctuation index data, and specifically, whether the index data is abnormal fluctuation index data may be determined by detecting a fluctuation amount or a fluctuation rate of the index data.
Further, after determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method, the method may further include: and generating abnormal data fluctuation warning information.
Specifically, a data abnormal fluctuation warning message may be generated after each determination of one abnormal fluctuation index data, or a data abnormal fluctuation warning message may be generated after at least one abnormal fluctuation index data in at least one index data is determined. The data abnormal fluctuation warning information may be alarm sound information, or may be prompt information popped up through a prompt window, or other information for warning data abnormal, which is not limited in this embodiment.
S103, analyzing the abnormal fluctuation index data according to corresponding preset dimension information respectively, and generating corresponding abnormal fluctuation analysis results, wherein the preset dimension information comprises the dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
Specifically, in this operation, each abnormal fluctuation index data in the at least one abnormal fluctuation index data acquired in S102 is analyzed according to the corresponding preset dimension information, and a corresponding abnormal fluctuation analysis result is generated, that is, each abnormal fluctuation index data generates a corresponding abnormal fluctuation analysis result.
The dimension of the abnormal fluctuation index data is the level number of the abnormal fluctuation index data. For example, if the abnormal fluctuation index data is located at the first level and includes two second levels of abnormal wave sub-index data, and one second level of abnormal wave sub-index data includes two third levels of sub-index data, the number of levels of the abnormal fluctuation index data is 3, that is, the dimension of the abnormal fluctuation index is 3.
After the generating of the corresponding abnormal fluctuation analysis result, the method further comprises: and displaying the abnormal fluctuation analysis result. The abnormal fluctuation analysis result may be an abnormal fluctuation analysis report in the form of a web page.
The technical scheme provided by the embodiment is that at least one abnormal fluctuation index data is automatically determined by at least one index data in the data to be detected according to a preset data fluctuation detection method, and the abnormal fluctuation index data is analyzed according to corresponding preset dimension information respectively to generate a corresponding abnormal fluctuation analysis result, so that the technical problems that in the prior art, a data analyst judges whether each index data in the data to be detected generates abnormal fluctuation or not and the reason for generating abnormal fluctuation is analyzed are solved, the data abnormal fluctuation detection and analysis time is long, the cost is high, and the accuracy is low, and the accuracy of data abnormal fluctuation detection and the efficiency of data abnormal fluctuation reason analysis are improved.
Example two
Fig. 2 is a flowchart of a data abnormal fluctuation detection method according to a second embodiment of the present invention, and on the basis of the foregoing embodiments, this embodiment preferably specifically optimizes at least one abnormal fluctuation index data in the at least one index data determined according to a preset data fluctuation detection method as follows: acquiring a ring ratio fluctuation amount and a same ratio fluctuation amount of each index data in the at least one index data, wherein the ring ratio fluctuation amount is a change amplitude of a current day value and a yesterday value of the index data, and the same ratio fluctuation amount is a change amplitude of the current day value and a corresponding day value of a last week of the index data; and determining at least one abnormal fluctuation index data in the at least one index data according to the ring ratio fluctuation amount and the same ratio fluctuation amount of each index data and the ring ratio setting range and the same ratio setting range corresponding to each index data.
Further, the abnormal fluctuation index data are analyzed according to corresponding preset dimension information respectively, and corresponding abnormal fluctuation analysis results are generated, wherein the preset dimension information includes dimensions of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension, and specifically, the preset dimension information includes: respectively taking the abnormal fluctuation index data as current abnormal fluctuation index data, and generating an index abnormal fluctuation analysis tree corresponding to the current abnormal fluctuation index data according to preset dimension information corresponding to the current abnormal fluctuation index data; and analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree downwards layer by layer, and generating a corresponding abnormal fluctuation analysis result.
Correspondingly, as shown in fig. 2, the method for detecting data abnormal fluctuation provided in this embodiment specifically includes:
s201, data to be detected is obtained, and the data to be detected comprises at least one index data.
S202, obtaining a ring ratio fluctuation amount and a same ratio fluctuation amount of each index data in the at least one index data, wherein the ring ratio fluctuation amount is a change amplitude of a current date value and a yesterday value of the index data, and the same ratio fluctuation amount is a change amplitude of the current date value and a previous week corresponding date value of the index data.
S203, determining at least one abnormal fluctuation index data in the at least one index data according to the ring ratio fluctuation amount and the same ratio fluctuation amount of each index data and the ring ratio setting range and the same ratio setting range corresponding to each index data.
The operation may specifically include: respectively taking each index data as current index data, and judging whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range and/or whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range according to the ring ratio fluctuation amount and the same ratio fluctuation amount of the current index data and the ring ratio setting range and the same ratio setting range corresponding to the current index data; and when the ring ratio fluctuation amount of the current index data is out of the ring ratio set range and/or the same-ratio fluctuation amount is out of the same-ratio set range, determining that the current index data is abnormal fluctuation index data. Preferably, when the ring ratio fluctuation amount of the current index data is outside the ring ratio setting range and the same-ratio fluctuation amount is outside the same-ratio setting range, it is determined that the current index data is abnormal fluctuation index data.
Wherein, the lower limit value of the ring ratio setting range D1 is as follows:the upper limit value of the ring ratio setting range D1 is:
wherein n is a set number of cycles, uiThe amount of cyclic fluctuation, ud, corresponding to the cycle is set for the ith1Setting the average value of the ring ratio fluctuation quantity in the cycle number n as a set parameter value, wherein the value is more than or equal to 1 and less than or equal to 6.
Illustratively, if n is 4, and the date corresponding to the current index data is wednesday, the ring ratio fluctuation u is calculated1The variation amplitude of the third and second cycles in the first set cycle (for example, the current cycle) and the ring ratio fluctuation u2The variation amplitude of the second set period (such as the last period) of the third and second cycles and the ring ratio fluctuation u3The variation amplitude of the third and the second in the third set period (for example, the upper period) and the ring ratio fluctuation u4The variation of the third and second weeks in the fourth set week (for example, the last week) can be set. The setting parameter is more than or equal to 6 according to the experimental test value, and the experiment verifies that the range value can obtain a better ring ratio setting range and a same ratio setting range. The benefits of this arrangement are: said u isiAll the values are the current indexesThe date (such as wednesday) corresponding to the data is selected, and the ring ratio fluctuation amount of the current index data can be reflected more accurately by selecting each ring ratio fluctuation amount of the date corresponding to the set number of weeks.
The lower limit value of the set range D2 is:the upper limit value of the set range D2 is:
wherein n is a set number of cycles, ujThe same ratio fluctuation amount u corresponding to the cycle is set for the j-th cycled2The average value of the fluctuation amounts of the same ratio in the number of cycles n is set as a set parameter value.
Illustratively, if n is 4, and the date corresponding to the current index data is wednesday, the geometric fluctuation u is1The variation amplitude of the third and last week in the first set period (for example, the current period) and the fluctuation u of the same ratio can be set2May be three weeks in the second set week (e.g., the last week)Amplitude of variation of the last cycle of the cycle, the fluctuation amount u3The variation amplitude of the third week (for example, the last week) and the third week of the last week can be set to be the same as the fluctuation u4The variation width of the second week (for example, the upper week) and the upper week may be set to be within the fourth set week. The benefits of this arrangement are: said u isjThe values of the data are all the same-ratio fluctuation amounts of the dates (such as wednesday) corresponding to the current index data, and the same-ratio fluctuation amounts of the dates corresponding to the set week number are selected, so that the same-ratio fluctuation amounts of the current index data can be reflected more accurately.
In this embodiment, at least one abnormal fluctuation index data in the at least one index data is determined according to the same-ratio fluctuation amount and/or the ring-ratio fluctuation amount of at least one data index in the data to be detected, so that a technical effect of quickly and accurately determining whether the data to be detected has abnormal fluctuation can be achieved.
And S204, respectively using the abnormal fluctuation index data as current abnormal fluctuation index data, and generating an index abnormal fluctuation analysis tree corresponding to the current abnormal fluctuation index data according to preset dimension information corresponding to the current abnormal fluctuation index data, wherein the number of layers of the index abnormal fluctuation analysis tree corresponds to the dimension of the current abnormal fluctuation index data, each node of each layer of the index abnormal fluctuation analysis tree corresponds to at least one dimension index data of the corresponding dimension, and the node of the first layer of the index abnormal fluctuation analysis tree corresponds to the current abnormal fluctuation index data.
Illustratively, the current abnormal fluctuation index data is independent visitor (UV) index data in the to-be-detected bar service data, a dimension of the UV index data is 4, and fig. 3 is a schematic structural diagram of an index abnormal fluctuation analysis tree corresponding to the UV index data provided in the second embodiment of the present invention. As shown in fig. 3, according to the preset dimension information corresponding to the UV index data 30, based on a set mathematical model, an index abnormal fluctuation analysis tree corresponding to the UV index data is generated, the number of layers of the index abnormal fluctuation analysis tree corresponds to the dimension 4 of the UV index, the node of the first layer of the index abnormal fluctuation analysis tree corresponds to UV index data, each layer of nodes of the index abnormal fluctuation analysis tree corresponds to at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data, for example, each node of the second layer of the index abnormal fluctuation analysis tree corresponds to the 24-hour dimension index data 31, the platform dimension index data 32, the major-minor version dimension index data 33, the IP dimension index data 34, the page dimension index data 35, and the directory dimension index data 36 of the second dimension in the UA index data 30.
Wherein the set mathematical model includes a decision tree model and a hybrid online analytical processing (HOLAP) model.
S205, analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree downwards layer by layer, and generating a corresponding abnormal fluctuation analysis result.
In this operation, analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree downward layer by layer may specifically include: analyzing at least one dimension data corresponding to each node of a second layer of the index abnormal fluctuation analysis tree, and determining at least one dimension index data meeting preset conditions in the second layer; analyzing the next-layer dimension index data of the current dimension data by taking each dimension index data in the at least one dimension index data meeting the preset condition as the current dimension index data respectively to generate an analysis result meeting the preset condition; and if the analysis result does not comprise the leaf node of the index abnormal fluctuation analysis tree, taking the analysis result as new current dimension index data, analyzing the next layer of dimension index data of the new current dimension index data, and generating a new analysis result meeting preset conditions until the new current dimension index data is the leaf node of the index abnormal fluctuation analysis tree.
The preset condition may be that the ring ratio fluctuation rate of at least one dimension index data reaches a first set number and/or the same ratio fluctuation rate reaches a first set number; or the sum of the ring ratio fluctuation rates or the sum of the same ratio fluctuation rates of the plurality of dimension index data reaches a second set number, or the sum of the ring ratio fluctuation rates and the same ratio fluctuation rates of the plurality of dimension index data all reach the second set number, wherein the ring ratio fluctuation rate is the ratio of the ring ratio fluctuation amount of the dimension index data to the yesterday value, and the same ratio fluctuation rate is the ratio of the same ratio fluctuation amount of the dimension index data to the last week corresponding day value. For example, the first set number is 30%, and if the ring ratio fluctuation rate of one dimension index data reaches 30% and/or the same ratio fluctuation rate reaches 30%, the dimension index data is the dimension index data meeting the preset condition; the second set number is 50%, if the sum of the ring ratio fluctuation rates of the two dimensional index data can reach 50%, or the sum of the same ratio fluctuation rates can reach 50%, or both the sum of the ring ratio fluctuation rates and the same ratio fluctuation rates can reach 50%, the two dimensional index data are the dimensional index data meeting the preset condition. In addition, the first set number may be the same as the second set number.
For example, as shown in fig. 3, at least one dimension data corresponding to each node of the second layer of the index abnormal fluctuation analysis tree corresponding to the UV index data is analyzed, and the dimension index data (marked by a black emphasis frame in the figure) satisfying the preset condition in the second layer is determined as the main sub-version dimension index data 33, the page dimension index data 35, and the directory dimension index data 36; analyzing the main and sub version dimension index data 33, the page dimension index data 35, and the directory dimension index data 36, for example, analyzing the main and sub version dimension index data 33, specifically, analyzing the next layer dimension index data of the main and sub version dimension index data, and generating an analysis result meeting a preset condition, as shown in fig. 3, where the analysis result meeting the preset condition is the page dimension index data 331 in the third dimension; when the page dimension index data 331 in the third dimension is not a leaf node, analyzing the dimension index data of the next layer to generate an analysis result meeting a preset condition, as shown in fig. 3, at this time, the analysis result meeting the preset condition is 24-hour dimension index data 3311 in the fourth dimension, because the 24-hour dimension index data 3311 in the fourth dimension is a leaf node, the analysis is stopped, and the other dimension index data 3311 in the second layer meeting the preset condition is continuously analyzed by using the same method, for example, 24-hour dimension index data 351 in the third dimension of the page dimension index data 35, bar dimension index data 361 in the third dimension of the directory dimension index data 36, and sticker dimension index data 3611 in the fourth dimension are sequentially analyzed.
Furthermore, after an abnormal fluctuation analysis result corresponding to the abnormal fluctuation index data is generated, the abnormal fluctuation analysis result is displayed, so that a user can clearly check the reason of the abnormal fluctuation of the data through the abnormal fluctuation analysis result. The abnormal fluctuation analysis result may be an abnormal fluctuation analysis report in the form of a web page.
Fig. 4 is a schematic diagram of an abnormal fluctuation analysis report according to a second embodiment of the present invention, and as shown in fig. 4, in the abnormal fluctuation analysis report corresponding to the abnormal fluctuation index data UV index data, fluctuation conditions of the dimensional index data of each dimension corresponding to each node of each layer obtained by analyzing the corresponding index abnormal fluctuation analysis tree are shown. In fig. 4, the abnormal index is abnormal fluctuation index data, and the client daily life (number of cookies) is UV index data. As can be seen from fig. 4, a part of dimension index data in the second layer of the abnormal fluctuation analysis tree is shown: the 24-hour dimension index data (the 24-hour distribution map in fig. 4), the platform dimension index data, the main and sub-version dimension index data, and the Software Development Kit (SDK) 24-hour dimension index data (the SDK24 hour distribution map in fig. 4) in a layer below the main and sub-version dimension index data satisfying a preset condition. It should be noted that fig. 4 only shows the fluctuation of the dimension index data in a part of the analysis process by way of example.
It should be noted that, in this embodiment, only one abnormal fluctuation index data (for example, UV index data) in the data to be detected is exemplarily analyzed through the corresponding index abnormal fluctuation analysis tree to generate a corresponding abnormal fluctuation analysis result, and similarly, other abnormal fluctuation index data in the data to be detected may also be analyzed and processed by the same method.
Fig. 5 shows a schematic structural diagram of an index abnormal fluctuation analysis comprehensive tree T provided in the second embodiment of the present invention, where the index abnormal fluctuation analysis comprehensive tree T is a sum of index abnormal fluctuation analysis trees corresponding to at least one abnormal fluctuation index data in data to be detected, as shown in fig. 5, an nth branch tree of the index abnormal fluctuation analysis comprehensive tree T represents an index abnormal fluctuation analysis tree corresponding to an nth abnormal fluctuation index data, exemplarily, a first abnormal fluctuation index data a1 is a UV index data, a second dimension index data a21 under the corresponding first abnormal fluctuation index data a1 may be a 24-hour dimension index data, a second dimension index data a22 may be a platform dimension index data, a second dimension index data A2N may be a dimension directory index data, a third dimension index data a31 may be a bar dimension index data under the directory dimension index data, the fourth dimension index data a41 may be post dimension index data under the bar dimension index data.
According to the technical scheme provided by the embodiment, at least one abnormal fluctuation index data is automatically determined according to the ring ratio fluctuation amount and/or the same ratio fluctuation amount by at least one index data in the data to be detected, and generating corresponding index abnormal fluctuation analysis trees according to the corresponding preset dimension information respectively by the abnormal fluctuation index data, the index abnormal fluctuation analysis tree is analyzed layer by layer to generate a corresponding abnormal fluctuation analysis result, so that the technical problems of long time, high cost and low accuracy of data abnormal fluctuation detection and analysis caused by judging whether each index data in the data to be detected is abnormally fluctuated by a data analyst and analyzing the reason of the abnormal fluctuation in the prior art are solved, the automation of data abnormal fluctuation detection and analysis is realized, meanwhile, the accuracy of data abnormal fluctuation detection and the efficiency of analyzing the reason of the data abnormal fluctuation are improved.
EXAMPLE III
Fig. 6 is a schematic flow chart of the abnormal data fluctuation detection logic provided by the third embodiment of the present invention. The present embodiment is optimized based on the above embodiments to provide a preferred embodiment, and please refer to the above embodiments for technical details not described in detail in the present embodiment. As shown in fig. 6, the logic flow for detecting abnormal data fluctuation provided by this embodiment includes:
s601, acquiring data to be detected 61, wherein the data to be detected 61 comprises at least one index data;
s602, detecting the at least one index data according to a preset data abnormal fluctuation detection method 62, and determining at least one abnormal fluctuation index data in the at least one index data;
s603, generating a corresponding index abnormal fluctuation analysis tree 64 by the abnormal fluctuation index data based on a decision tree model 631 or a mixed online analysis processing model 632;
s604, analyzing the index abnormal fluctuation analysis tree 64 to generate a corresponding abnormal fluctuation analysis result 65.
According to the technical scheme provided by the embodiment, at least one abnormal fluctuation index data is automatically determined by at least one index data in the data to be detected according to a preset data fluctuation detection method, and the abnormal fluctuation index data is generated and analyzed by a corresponding index abnormal fluctuation analysis tree based on a decision tree model or a mixed online analysis processing model, so that a corresponding abnormal fluctuation analysis result is obtained, the automation of data abnormal fluctuation detection and analysis is realized, the time for data abnormal fluctuation detection and analysis is saved, the increasing requirements of people on convenience of data abnormal fluctuation detection and analysis are met, and the accuracy of data abnormal fluctuation detection and the efficiency of analysis of data abnormal fluctuation reasons are improved.
Example four
Fig. 7 is a schematic structural diagram of a data abnormal fluctuation detection apparatus according to a fourth embodiment of the present invention, where the data abnormal fluctuation detection apparatus is disposed in a server, and as shown in fig. 7, the apparatus includes:
the data to be detected acquisition module 71 is configured to acquire data to be detected, where the data to be detected includes at least one index data;
an abnormal fluctuation index data determination module 72, configured to determine at least one abnormal fluctuation index data of the at least one index data according to a preset data fluctuation detection method;
and the abnormal fluctuation index data analysis module 73 is configured to analyze the abnormal fluctuation index data according to corresponding preset dimension information, and generate a corresponding abnormal fluctuation analysis result, where the preset dimension information includes a dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
Optionally, the abnormal fluctuation index data determination module 72 includes:
the index data fluctuation amount acquisition unit is used for acquiring a ring ratio fluctuation amount and a same ratio fluctuation amount of each index data in the at least one index data, wherein the ring ratio fluctuation amount is the change amplitude of the current day value and the yesterday value of the index data, and the same ratio fluctuation amount is the change amplitude of the current day value and the previous week corresponding day value of the index data;
and the abnormal fluctuation index data determining unit is used for determining at least one abnormal fluctuation index data in the at least one index data according to the ring ratio fluctuation amount and the same ratio fluctuation amount of each index data and the ring ratio setting range and the same ratio setting range corresponding to each index data.
Optionally, the abnormal fluctuation index data determining unit is specifically configured to:
respectively taking each index data as current index data, and judging whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range and/or whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range according to the ring ratio fluctuation amount and the same ratio fluctuation amount of the current index data and the ring ratio setting range and the same ratio setting range corresponding to the current index data;
and when the ring ratio fluctuation amount of the current index data is out of the ring ratio set range and/or the same-ratio fluctuation amount is out of the same-ratio set range, determining that the current index data is abnormal fluctuation index data.
Further, the lower limit value of the ring ratio setting range D1 is:
the upper limit value of the ring ratio setting range D1 is:
wherein n is a set number of cycles, uiSetting a corresponding cyclic fluctuation amount u for the ithd1Setting the average value of the ring ratio fluctuation quantity in the cycle number n as a set parameter value, wherein the value is more than or equal to 1 and less than or equal to 6;
the lower limit value of the set range D2 is:
the upper limit value of the set range D2 is:
wherein n is a set number of cycles, ujThe same ratio fluctuation amount u corresponding to the cycle is set for the j-th cycled2The average value of the fluctuation amounts of the same ratio in the number of cycles n is set as a set parameter value.
Optionally, the abnormal fluctuation index data analysis module 73 includes:
an index abnormal fluctuation analysis tree generation unit, configured to respectively use the abnormal fluctuation index data as current abnormal fluctuation index data, and generate an index abnormal fluctuation analysis tree corresponding to the current abnormal fluctuation index data based on a set mathematical model according to preset dimension information corresponding to the current abnormal fluctuation index data, where the number of layers of the index abnormal fluctuation analysis tree corresponds to a dimension of the current abnormal fluctuation index data, each node of each layer of the index abnormal fluctuation analysis tree corresponds to at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data, and a node of a first layer of the index abnormal fluctuation analysis tree corresponds to the current abnormal fluctuation index data;
and the index abnormal fluctuation analysis tree analysis unit is used for analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree to the lower layer by layer and generating a corresponding abnormal fluctuation analysis result.
Optionally, the index abnormal fluctuation analysis tree analysis unit is specifically configured to:
analyzing at least one dimension data corresponding to each node of a second layer of the index abnormal fluctuation analysis tree, and determining at least one dimension index data meeting preset conditions in the second layer;
analyzing the next-layer dimension index data of the current dimension data by taking each dimension index data in the at least one dimension index data meeting the preset condition as the current dimension index data respectively to generate an analysis result meeting the preset condition;
and if the analysis result does not comprise the leaf node of the index abnormal fluctuation analysis tree, taking the analysis result as new current dimension index data, analyzing the next layer of dimension index data of the new current dimension index data, and generating a new analysis result meeting preset conditions until the new current dimension index data is the leaf node of the index abnormal fluctuation analysis tree.
Optionally, the preset condition may be: the ring ratio fluctuation rate of at least one dimension index data reaches a first set number and/or the same ratio fluctuation rate reaches a first set number, the ring ratio fluctuation rate is the ratio of the ring ratio fluctuation amount of the dimension index data to the yesterday value, and the same ratio fluctuation rate is the ratio of the same ratio fluctuation amount of the dimension index data to the last week corresponding day value; or,
the sum of the ring ratio fluctuation rates or the sum of the same ratio fluctuation rates of the plurality of dimension index data reaches a second set number, or the sum of the ring ratio fluctuation rates and the sum of the same ratio fluctuation rates of the plurality of dimension index data all reach the second set number.
Optionally, the apparatus further comprises:
and the warning information generating module is used for generating data abnormal fluctuation warning information after determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method.
Optionally, the set data model includes a decision tree model and a hybrid online analysis processing model.
Optionally, the abnormal fluctuation analysis result is an abnormal fluctuation analysis report in a form of a web page, and the apparatus further includes:
and the abnormal fluctuation analysis result display module is used for displaying the abnormal fluctuation analysis result after generating the corresponding abnormal fluctuation analysis result.
The data abnormal fluctuation detection device provided by the embodiment of the invention can be used for executing the data abnormal fluctuation detection method provided by any embodiment of the invention, has corresponding functional modules and realizes the same beneficial effects.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (14)
1. A method for detecting abnormal fluctuation of data is characterized by comprising the following steps:
acquiring data to be detected, wherein the data to be detected comprises at least one index data;
determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method;
and analyzing the abnormal fluctuation index data according to corresponding preset dimension information respectively, and generating corresponding abnormal fluctuation analysis results, wherein the preset dimension information comprises the dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
2. The method according to claim 1, wherein the determining at least one abnormal fluctuation index data among the at least one index data according to a preset data fluctuation detection method includes:
acquiring a ring ratio fluctuation amount and a same ratio fluctuation amount of each index data in the at least one index data, wherein the ring ratio fluctuation amount is a change amplitude of a current day value and a yesterday value of the index data, and the same ratio fluctuation amount is a change amplitude of the current day value and a corresponding day value of a last week of the index data;
and determining at least one abnormal fluctuation index data in the at least one index data according to the ring ratio fluctuation amount and the same ratio fluctuation amount of each index data and the ring ratio setting range and the same ratio setting range corresponding to each index data.
3. The method according to claim 2, wherein the determining at least one abnormal fluctuation index data in the at least one index data according to the amount of ring ratio fluctuation and the amount of same ratio fluctuation of each index data, and the corresponding ring ratio setting range and same ratio setting range of each index data comprises:
respectively taking each index data as current index data, and judging whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range and/or whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range according to the ring ratio fluctuation amount and the same ratio fluctuation amount of the current index data and the ring ratio setting range and the same ratio setting range corresponding to the current index data;
and when the ring ratio fluctuation amount of the current index data is out of the ring ratio set range and/or the same-ratio fluctuation amount is out of the same-ratio set range, determining that the current index data is abnormal fluctuation index data.
4. The method of claim 3, wherein:
the lower limit value of the ring ratio setting range D1 is:
the upper limit value of the ring ratio setting range D1 is:
wherein n is a set number of cycles, uiSetting a corresponding cyclic fluctuation amount u for the ithd1Setting the average value of the ring ratio fluctuation quantity in the cycle number n as a set parameter value, wherein the value is more than or equal to 1 and less than or equal to 6;
the lower limit value of the set range D2 is:
the upper limit value of the set range D2 is:
wherein n is a set number of cycles, ujThe same ratio fluctuation amount u corresponding to the cycle is set for the j-th cycled2The average value of the fluctuation amounts of the same ratio in the number of cycles n is set as a set parameter value.
5. The method according to claim 1, wherein the analyzing the abnormal fluctuation index data according to the corresponding preset dimension information, and generating corresponding abnormal fluctuation analysis results comprises:
respectively taking the abnormal fluctuation index data as current abnormal fluctuation index data, and generating an index abnormal fluctuation analysis tree corresponding to the current abnormal fluctuation index data according to preset dimension information corresponding to the current abnormal fluctuation index data, wherein the number of layers of the index abnormal fluctuation analysis tree corresponds to the dimension of the current abnormal fluctuation index data, each node of each layer of the index abnormal fluctuation analysis tree corresponds to at least one dimension index data of the corresponding dimension, and the node of the first layer of the index abnormal fluctuation analysis tree corresponds to the current abnormal fluctuation index data;
and analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree downwards layer by layer, and generating a corresponding abnormal fluctuation analysis result.
6. The method according to claim 5, wherein analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from a first layer of the index abnormal fluctuation analysis tree to a lower layer by layer comprises:
analyzing at least one dimension data corresponding to each node of a second layer of the index abnormal fluctuation analysis tree, and determining at least one dimension index data meeting preset conditions in the second layer;
analyzing the next-layer dimension index data of the current dimension data by taking each dimension index data in the at least one dimension index data meeting the preset condition as the current dimension index data respectively to generate an analysis result meeting the preset condition;
and if the analysis result does not comprise the leaf node of the index abnormal fluctuation analysis tree, taking the analysis result as new current dimension index data, analyzing the next layer of dimension index data of the new current dimension index data, and generating a new analysis result meeting preset conditions until the new current dimension index data is the leaf node of the index abnormal fluctuation analysis tree.
7. The method according to claim 6, wherein the preset condition is:
the ring ratio fluctuation rate of at least one dimension index data reaches a first set number and/or the same ratio fluctuation rate reaches a first set number, the ring ratio fluctuation rate is the ratio of the ring ratio fluctuation amount of the dimension index data to the yesterday value, and the same ratio fluctuation rate is the ratio of the same ratio fluctuation amount of the dimension index data to the last week corresponding day value; or,
the sum of the ring ratio fluctuation rates or the sum of the same ratio fluctuation rates of the plurality of dimension index data reaches a second set number, or the sum of the ring ratio fluctuation rates and the sum of the same ratio fluctuation rates of the plurality of dimension index data all reach the second set number.
8. An abnormal fluctuation data detection apparatus, comprising:
the data acquisition module to be detected is used for acquiring data to be detected, wherein the data to be detected comprises at least one index data;
the abnormal fluctuation index data determining module is used for determining at least one abnormal fluctuation index data in the at least one index data according to a preset data fluctuation detection method;
and the abnormal fluctuation index data analysis module is used for analyzing the abnormal fluctuation index data according to corresponding preset dimension information respectively and generating corresponding abnormal fluctuation analysis results, wherein the preset dimension information comprises the dimension of the abnormal fluctuation index data and at least one dimension index data corresponding to each dimension.
9. The apparatus of claim 8, wherein the abnormal fluctuation index data determination module comprises:
the index data fluctuation amount acquisition unit is used for acquiring a ring ratio fluctuation amount and a same ratio fluctuation amount of each index data in the at least one index data, wherein the ring ratio fluctuation amount is the change amplitude of the current day value and the yesterday value of the index data, and the same ratio fluctuation amount is the change amplitude of the current day value and the previous week corresponding day value of the index data;
and the abnormal fluctuation index data determining unit is used for determining at least one abnormal fluctuation index data in the at least one index data according to the ring ratio fluctuation amount and the same ratio fluctuation amount of each index data and the ring ratio setting range and the same ratio setting range corresponding to each index data.
10. The apparatus according to claim 9, wherein the abnormal fluctuation index data determination unit is specifically configured to:
respectively taking each index data as current index data, and judging whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range and/or whether the ring ratio fluctuation amount of the current index data is out of the ring ratio setting range according to the ring ratio fluctuation amount and the same ratio fluctuation amount of the current index data and the ring ratio setting range and the same ratio setting range corresponding to the current index data;
and when the ring ratio fluctuation amount of the current index data is out of the ring ratio set range and/or the same-ratio fluctuation amount is out of the same-ratio set range, determining that the current index data is abnormal fluctuation index data.
11. The apparatus of claim 10, wherein:
the lower limit value of the ring ratio setting range D1 is:
the upper limit value of the ring ratio setting range D1 is:
wherein n is a set number of cycles, uiSetting a corresponding cyclic fluctuation amount u for the ithd1Setting the average value of the ring ratio fluctuation quantity in the cycle number n as a set parameter value, wherein the value is more than or equal to 1 and less than or equal to 6;
the lower limit value of the set range D2 is:
the upper limit value of the set range D2 is:
wherein n is a set number of cycles, ujThe same ratio fluctuation amount u corresponding to the cycle is set for the j-th cycled2The average value of the fluctuation amounts of the same ratio in the number of cycles n is set as a set parameter value.
12. The apparatus of claim 8, wherein the abnormal fluctuation index data analysis module comprises:
an index abnormal fluctuation analysis tree generation unit, configured to use the abnormal fluctuation index data as current abnormal fluctuation index data, and generate an index abnormal fluctuation analysis tree corresponding to the current abnormal fluctuation index data according to preset dimension information corresponding to the current abnormal fluctuation index data, where the number of layers of the index abnormal fluctuation analysis tree corresponds to a dimension of the current abnormal fluctuation index data, each node of each layer of the index abnormal fluctuation analysis tree corresponds to at least one dimension index data of the corresponding dimension, and a node of a first layer of the index abnormal fluctuation analysis tree corresponds to the current abnormal fluctuation index data;
and the index abnormal fluctuation analysis tree analysis unit is used for analyzing at least one dimension index data corresponding to each dimension of the current abnormal fluctuation index data corresponding to each node of each layer from the first layer of the index abnormal fluctuation analysis tree to the lower layer by layer and generating a corresponding abnormal fluctuation analysis result.
13. The apparatus according to claim 12, wherein the index abnormal fluctuation analysis tree analysis unit is specifically configured to:
analyzing at least one dimension data corresponding to each node of a second layer of the index abnormal fluctuation analysis tree, and determining at least one dimension index data meeting preset conditions in the second layer;
analyzing the next-layer dimension index data of the current dimension data by taking each dimension index data in the at least one dimension index data meeting the preset condition as the current dimension index data respectively to generate an analysis result meeting the preset condition;
and if the analysis result does not comprise the leaf node of the index abnormal fluctuation analysis tree, taking the analysis result as new current dimension index data, analyzing the next layer of dimension index data of the new current dimension index data, and generating a new analysis result meeting preset conditions until the new current dimension index data is the leaf node of the index abnormal fluctuation analysis tree.
14. The apparatus according to claim 13, wherein the preset condition is:
the ring ratio fluctuation rate of at least one dimension index data reaches a first set number and/or the same ratio fluctuation rate reaches a first set number, the ring ratio fluctuation rate is the ratio of the ring ratio fluctuation amount of the dimension index data to the yesterday value, and the same ratio fluctuation rate is the ratio of the same ratio fluctuation amount of the dimension index data to the last week corresponding day value; or,
the sum of the ring ratio fluctuation rates or the sum of the same ratio fluctuation rates of the plurality of dimension index data reaches a second set number, or the sum of the ring ratio fluctuation rates and the sum of the same ratio fluctuation rates of the plurality of dimension index data all reach the second set number.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510920803.3A CN105447323A (en) | 2015-12-11 | 2015-12-11 | Data abnormal fluctuations detecting method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510920803.3A CN105447323A (en) | 2015-12-11 | 2015-12-11 | Data abnormal fluctuations detecting method and apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105447323A true CN105447323A (en) | 2016-03-30 |
Family
ID=55557493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510920803.3A Pending CN105447323A (en) | 2015-12-11 | 2015-12-11 | Data abnormal fluctuations detecting method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105447323A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107026771A (en) * | 2017-04-20 | 2017-08-08 | 北京京东尚科信息技术有限公司 | Method, device, electronic equipment and the computer-readable recording medium of threshold adaptive |
| CN107819745A (en) * | 2017-10-25 | 2018-03-20 | 北京京东尚科信息技术有限公司 | The defence method and device of abnormal flow |
| CN107957988A (en) * | 2016-10-18 | 2018-04-24 | 阿里巴巴集团控股有限公司 | For determining the method, apparatus and electronic equipment of data exception reason |
| CN108009715A (en) * | 2017-11-28 | 2018-05-08 | 邢加和 | It is a kind of automatically analyze index fluctuation root because method |
| CN108346011A (en) * | 2018-05-15 | 2018-07-31 | 阿里巴巴集团控股有限公司 | Index fluction analysis method and 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 |
| CN108959493A (en) * | 2018-06-25 | 2018-12-07 | 阿里巴巴集团控股有限公司 | Detection method, device and the equipment of Indexes Abnormality fluctuation |
| CN109033104A (en) * | 2017-06-09 | 2018-12-18 | 微软技术许可有限责任公司 | To automatically analyzing for the difference between cube |
| 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 |
| CN109558295A (en) * | 2018-11-15 | 2019-04-02 | 新华三信息安全技术有限公司 | A kind of performance indicator method for detecting abnormality and device |
| CN109597744A (en) * | 2018-11-16 | 2019-04-09 | 北京奇虎科技有限公司 | Data unusual fluctuation analysis method and device |
| CN109634804A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | Data unusual fluctuation analysis method and device |
| CN109726068A (en) * | 2017-10-31 | 2019-05-07 | 北京国双科技有限公司 | A kind of data detection method and device |
| CN109882834A (en) * | 2019-03-27 | 2019-06-14 | 新奥数能科技有限公司 | The operation data monitoring method and device of boiler plant |
| CN110046291A (en) * | 2019-01-28 | 2019-07-23 | 阿里巴巴集团控股有限公司 | Visualization disassembling method, device and the equipment of achievement data |
| CN110995524A (en) * | 2019-10-28 | 2020-04-10 | 北京三快在线科技有限公司 | Flow data monitoring method and device, electronic equipment and computer readable medium |
| CN111833135A (en) * | 2019-08-01 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Order data analysis method and device and electronic equipment |
| CN111901171A (en) * | 2020-07-29 | 2020-11-06 | 腾讯科技（深圳）有限公司 | Anomaly detection and attribution method, device, equipment and computer readable storage medium |
| CN113220796A (en) * | 2020-01-21 | 2021-08-06 | 北京达佳互联信息技术有限公司 | Abnormal business index analysis method and device |
| CN113553208A (en) * | 2021-07-19 | 2021-10-26 | 神策网络科技（北京）有限公司 | Method for determining abnormal dimension of data abnormal point |
| CN114298863A (en) * | 2022-03-11 | 2022-04-08 | 浙江万胜智能科技股份有限公司 | Data acquisition method and system of intelligent meter reading terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102081623A (en) * | 2009-11-30 | 2011-06-01 | 中国移动通信集团浙江有限公司 | Method and system for detecting database abnormality |
| CN103532776A (en) * | 2013-09-30 | 2014-01-22 | 广东电网公司电力调度控制中心 | Service flow detection method and system |
| CN104268080A (en) * | 2014-09-25 | 2015-01-07 | 北京金山安全软件有限公司 | Software exception handling method and device |
-
2015
- 2015-12-11 CN CN201510920803.3A patent/CN105447323A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102081623A (en) * | 2009-11-30 | 2011-06-01 | 中国移动通信集团浙江有限公司 | Method and system for detecting database abnormality |
| CN103532776A (en) * | 2013-09-30 | 2014-01-22 | 广东电网公司电力调度控制中心 | Service flow detection method and system |
| CN104268080A (en) * | 2014-09-25 | 2015-01-07 | 北京金山安全软件有限公司 | Software exception handling method and device |
Non-Patent Citations (2)
| Title |
|---|
| 孙淑娟: "智能运维在百度日常业务监控中的探索", 《HTTP://OS.51CTO.COM/ART/201508/488470.HTM》 * |
| 郑黎明等: "多维多层次网络流量异常检测研究", 《计算机研究与发展》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107957988A (en) * | 2016-10-18 | 2018-04-24 | 阿里巴巴集团控股有限公司 | For determining the method, apparatus and electronic equipment of data exception reason |
| CN107026771B (en) * | 2017-04-20 | 2019-12-27 | 北京京东尚科信息技术有限公司 | Threshold value self-adaption method and device, electronic equipment and readable medium |
| CN107026771A (en) * | 2017-04-20 | 2017-08-08 | 北京京东尚科信息技术有限公司 | Method, device, electronic equipment and the computer-readable recording medium of threshold adaptive |
| US11308134B2 (en) | 2017-06-09 | 2022-04-19 | Microsoft Technology Licensing, Llc | Automatic analysis of difference between multi-dimensional datasets |
| CN109033104B (en) * | 2017-06-09 | 2022-05-13 | 微软技术许可有限责任公司 | Automatic analysis of differences between multi-dimensional datasets |
| CN109033104A (en) * | 2017-06-09 | 2018-12-18 | 微软技术许可有限责任公司 | To automatically analyzing for the difference between cube |
| CN107819745A (en) * | 2017-10-25 | 2018-03-20 | 北京京东尚科信息技术有限公司 | The defence method and device of abnormal flow |
| CN109726068A (en) * | 2017-10-31 | 2019-05-07 | 北京国双科技有限公司 | A kind of data detection method and device |
| CN108009715A (en) * | 2017-11-28 | 2018-05-08 | 邢加和 | It is a kind of automatically analyze index fluctuation root because method |
| CN108346011A (en) * | 2018-05-15 | 2018-07-31 | 阿里巴巴集团控股有限公司 | Index fluction analysis method and device |
| WO2019218746A1 (en) * | 2018-05-15 | 2019-11-21 | 阿里巴巴集团控股有限公司 | Method and device for analyzing indicator fluctuation |
| 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 |
| CN108959493A (en) * | 2018-06-25 | 2018-12-07 | 阿里巴巴集团控股有限公司 | Detection method, device and the equipment of Indexes Abnormality fluctuation |
| CN109241043B (en) * | 2018-08-13 | 2022-10-14 | 蜜小蜂智慧（北京）科技有限公司 | Data quality detection method and device |
| 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 |
| CN109558295A (en) * | 2018-11-15 | 2019-04-02 | 新华三信息安全技术有限公司 | A kind of performance indicator method for detecting abnormality and device |
| CN109597744A (en) * | 2018-11-16 | 2019-04-09 | 北京奇虎科技有限公司 | Data unusual fluctuation analysis method and device |
| CN109634804A (en) * | 2018-11-16 | 2019-04-16 | 北京奇虎科技有限公司 | Data unusual fluctuation analysis method and device |
| CN110046291A (en) * | 2019-01-28 | 2019-07-23 | 阿里巴巴集团控股有限公司 | Visualization disassembling method, device and the equipment of achievement data |
| CN109882834A (en) * | 2019-03-27 | 2019-06-14 | 新奥数能科技有限公司 | The operation data monitoring method and device of boiler plant |
| CN111833135A (en) * | 2019-08-01 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Order data analysis method and device and electronic equipment |
| CN110995524A (en) * | 2019-10-28 | 2020-04-10 | 北京三快在线科技有限公司 | Flow data monitoring method and device, electronic equipment and computer readable medium |
| CN110995524B (en) * | 2019-10-28 | 2022-06-14 | 北京三快在线科技有限公司 | Flow data monitoring method and device, electronic equipment and computer readable medium |
| CN113220796A (en) * | 2020-01-21 | 2021-08-06 | 北京达佳互联信息技术有限公司 | Abnormal business index analysis method and device |
| CN111901171A (en) * | 2020-07-29 | 2020-11-06 | 腾讯科技（深圳）有限公司 | Anomaly detection and attribution method, device, equipment and computer readable storage medium |
| CN111901171B (en) * | 2020-07-29 | 2023-10-20 | 腾讯科技（深圳）有限公司 | Anomaly detection and attribution method, apparatus, device, and computer-readable storage medium |
| CN113553208A (en) * | 2021-07-19 | 2021-10-26 | 神策网络科技（北京）有限公司 | Method for determining abnormal dimension of data abnormal point |
| CN114298863A (en) * | 2022-03-11 | 2022-04-08 | 浙江万胜智能科技股份有限公司 | Data acquisition method and system of intelligent meter reading terminal |
| CN114298863B (en) * | 2022-03-11 | 2022-08-02 | 浙江万胜智能科技股份有限公司 | Data acquisition method and system of intelligent meter reading terminal |
| WO2023168950A1 (en) * | 2022-03-11 | 2023-09-14 | 浙江万胜智能科技股份有限公司 | Data collection method and system for smart meter-reading terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105447323A (en) | Data abnormal fluctuations detecting method and apparatus | |
| US20220292527A1 (en) | Methods of assessing long-term indicators of sentiment | |
| US10248528B2 (en) | System monitoring method and apparatus | |
| Reed et al. | Increased threat of tropical cyclones and coastal flooding to New York City during the anthropogenic era | |
| Li et al. | Building sparse models for traffic flow prediction: An empirical comparison between statistical heuristics and geometric heuristics for Bayesian network approaches | |
| Uhe et al. | Comparison of methods: Attributing the 2014 record European temperatures to human influences | |
| US8768873B2 (en) | Space-time-node engine signal structure | |
| Lopes et al. | Stellar cycles from photometric data: CoRoT stars | |
| Süveges | Extreme-value modelling for the significance assessment of periodogram peaks | |
| von Trentini et al. | Comparing internal variabilities in three regional single model initial-condition large ensembles (SMILE) over Europe | |
| CN111680855A (en) | Automatic risk detection and early warning method and system for whole process of project | |
| Hintz et al. | Collecting and processing of barometric data from smartphones for potential use in numerical weather prediction data assimilation | |
| CN108761568B (en) | Environment monitoring method, device, system and server | |
| Massah et al. | Confidence intervals for time averages in the presence of long‐range correlations, a case study on Earth surface temperature anomalies | |
| Montero-Dorta et al. | On the influence of halo mass accretion history on galaxy properties and assembly bias | |
| US20190086234A1 (en) | Systems and methods for displaying resource savings | |
| De Mulder et al. | An open software package for data reconciliation and gap filling in preparation of Water and Resource Recovery Facility Modeling | |
| Diehl et al. | Albero: A visual analytics approach for probabilistic weather forecasting | |
| Sheoran et al. | Efficacy and application of the window-sliding ARIMA for daily and weekly wind speed forecasting | |
| CN112131381A (en) | Method and device for identifying high-alarm-level place, electronic equipment and storage medium | |
| Shiogama et al. | MIROC6 Large Ensemble (MIROC6-LE): Experimental design and initial analyses | |
| Cannavò et al. | On a possible unified scaling law for volcanic eruption durations | |
| CN114235108A (en) | Method and device for detecting abnormal state of gas flowmeter based on data analysis | |
| CN116756616A (en) | Data processing method, device, computer readable medium and electronic equipment | |
| CN116228501A (en) | Pollution discharge exceeding area industry determining method and device, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160330 |
|
| RJ01 | Rejection of invention patent application after publication |