CN109828825A - Abnormal deviation data examination method, device, computer equipment and storage medium - Google Patents
Abnormal deviation data examination method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109828825A CN109828825A CN201910012883.0A CN201910012883A CN109828825A CN 109828825 A CN109828825 A CN 109828825A CN 201910012883 A CN201910012883 A CN 201910012883A CN 109828825 A CN109828825 A CN 109828825A
- Authority
- CN
- China
- Prior art keywords
- data
- tested
- engine
- application container
- user
- 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 73
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 55
- 238000012360 testing method Methods 0.000 claims abstract description 50
- 238000012549 training Methods 0.000 claims description 67
- 238000013450 outlier detection Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 14
- 230000005856 abnormality Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 17
- 230000006870 function Effects 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 210000004209 hair Anatomy 0.000 description 3
- 238000007689 inspection Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005538 encapsulation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses abnormal deviation data examination method, device, computer equipment and storage mediums.This method comprises: receiving data to be tested if detecting the data to be tested of user terminal uploads;Judge to identify identical engine identification with the presence or absence of with the user object of data to be tested in each engine identification of the application container engine set constructed;If existing in each engine identification of the application container engine set constructed and identifying identical engine identification with the user object of data to be tested, it obtains and identifies application container engine corresponding to identical engine identification with the user object of data to be tested, as target application container engine, test data, which is treated, according to isolated forest model packaged in target application container engine carries out anomaly data detection, if data to be tested are abnormal data, data to be tested are saved to preset store path.The method achieve efficiently detecting exceptional value, and by way of disposing parallel, improve the efficiency of data processing.
Description
Technical field
The present invention relates to anomaly data detection technical field more particularly to a kind of abnormal deviation data examination method, device, calculating
Machine equipment and storage medium.
Background technique
Outlier detection is whether inspection data has typing mistake and the process containing the data for not conforming to convention, ignores different
The presence of constant value is very dangerous, and includes rejecting is not added into the process of calculation analysis of data, to result meeting exceptional value
Adverse effect is generated, payes attention to the appearance of exceptional value, analyzes its Producing reason, is usually become and is found the problem and then improve decision
Opportunity.Currently, generally by the way of artificial detection, this results in examining when exceptional value detects in the data to magnanimity
Inefficiency is surveyed, and accuracy is lower.
Summary of the invention
The embodiment of the invention provides a kind of abnormal deviation data examination method, device, computer equipment and storage mediums, it is intended to
When solution in the prior art detects exceptional value in the data of magnanimity, generally by the way of artificial detection, cause to detect
Inefficiency, and the problem that accuracy is lower.
In a first aspect, the embodiment of the invention provides a kind of abnormal deviation data examination methods comprising:
If detecting the data to be tested of user terminal uploads, the data to be tested are received;
Judge to whether there is and the data to be tested in each engine identification of the application container engine set constructed
User object identifies identical engine identification;
If there is the user couple with the data to be tested in each engine identification of the application container engine set constructed
As identifying identical engine identification, obtains and identified corresponding to identical engine identification with the user object of the data to be tested
Application container engine, as target application container engine, according to isolated forest packaged in the target application container engine
Model carries out anomaly data detection to the data to be tested, will be described to be tested if the data to be tested are abnormal data
Data are saved to preset store path;And
If there is no the users with the data to be tested in each engine identification of the application container engine set constructed
The identical engine identification of object identity is sent to there will be no the prompt information with user object mark respective application container engine
Corresponding user terminal.
Second aspect, the embodiment of the invention provides a kind of anomaly data detection devices comprising:
Data receipt unit to be tested, if being received described to be measured for detecting the data to be tested of user terminal uploads
Try data;
Identify judging unit, whether there is in each engine identification of the application container engine set for judge to have constructed with
The user object of the data to be tested identifies identical engine identification;
Abnormality detecting unit, if in each engine identification of the application container engine set for having constructed exist with it is described to
The user object of test data identifies identical engine identification, obtains identical with the user object of the data to be tested mark
Application container engine corresponding to engine identification, as target application container engine, according in the target application container engine
Packaged isolated forest model carries out anomaly data detection to the data to be tested, if the data to be tested are abnormal number
According to saving the data to be tested to preset store path;And
Prompt unit, if in each engine identification of the application container engine set for having constructed there is no with it is described to be measured
The user object for trying data identifies identical engine identification, and there will be no identify mentioning for respective application container engine with user object
Show that information is sent to corresponding user terminal.
The third aspect, the embodiment of the present invention provide a kind of computer equipment again comprising memory, processor and storage
On the memory and the computer program that can run on the processor, the processor execute the computer program
Abnormal deviation data examination method described in the above-mentioned first aspect of Shi Shixian.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, wherein the computer can
It reads storage medium and is stored with computer program, it is above-mentioned that the computer program when being executed by a processor executes the processor
Abnormal deviation data examination method described in first aspect.
The embodiment of the invention provides a kind of abnormal deviation data examination method, device, computer equipment and storage mediums.The party
If method detects the data to be tested of user terminal uploads, the data to be tested are received;Judge that the application container constructed is drawn
It holds up in each engine identification of set and identifies identical engine identification with the presence or absence of with the user object of the data to be tested;If
Exist in each engine identification of the application container engine set of building identical with the user object of the data to be tested mark
The user object of engine identification, acquisition and the data to be tested identifies application container corresponding to identical engine identification and draws
It holds up, as target application container engine, according to isolated forest model packaged in the target application container engine to described
Data to be tested carry out anomaly data detection, if the data to be tested are abnormal data, by the data to be tested save to
Preset store path.The method achieve efficiently detecting exceptional value, and by way of disposing parallel, improve at data
The efficiency of reason.
Detailed description of the invention
Technical solution in order to illustrate the embodiments of the present invention more clearly, below will be to needed in embodiment description
Attached drawing is briefly described, it should be apparent that, drawings in the following description are some embodiments of the invention, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the application scenarios schematic diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 2 is the flow diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 3 is another flow diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 4 is the sub-process schematic diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 5 is another flow diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 6 is another sub-process schematic diagram of abnormal deviation data examination method provided in an embodiment of the present invention;
Fig. 7 is the schematic block diagram of anomaly data detection device provided in an embodiment of the present invention;
Fig. 8 is another schematic block diagram of anomaly data detection device provided in an embodiment of the present invention;
Fig. 9 is the subelement schematic block diagram of anomaly data detection device provided in an embodiment of the present invention;
Figure 10 is another schematic block diagram of anomaly data detection device provided in an embodiment of the present invention;
Figure 11 is another subelement schematic block diagram of anomaly data detection device provided in an embodiment of the present invention;
Figure 12 is the schematic block diagram of computer equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " and "comprising" instruction
Described feature, entirety, step, operation, the presence of element and/or component, but one or more of the other feature, whole is not precluded
Body, step, operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
Fig. 1 and Fig. 2 are please referred to, Fig. 1 is the application scenarios signal of abnormal deviation data examination method provided in an embodiment of the present invention
Figure, Fig. 2 is the flow diagram of abnormal deviation data examination method provided in an embodiment of the present invention, the abnormal deviation data examination method application
In server, this method is executed by the application software being installed in server.
As shown in Fig. 2, the method comprising the steps of S110~S140.
If S110, the data to be tested for detecting user terminal uploads, the data to be tested are received.
In the present embodiment, for the clearer usage scenario for understanding technical solution, below to involved terminal
It is introduced.It wherein, in this application, is to carry out description technique scheme in the angle of server.
First is that server, server can receive the training dataset that multiple user terminals initially upload respectively, then needle
To the training dataset of each user terminal uploads, construct what isolated forest model and being respectively stored in was disposed in server respectively
(i.e. application container engine, docker are the application container engines of an open source to multiple docker containers, and developer can be packaged institute
Then the application of exploitation and dependence packet are published on the Linux machine of any prevalence into a transplantable container, can also
To realize virtualization) in, namely isolated forest model corresponding with each user, it is stored in docker corresponding with the user
In container, every docker container can be both distributed to a load end by server and disposed to realize that single thread will correspond to
The data to be tested of user terminal uploads carry out anomaly data detection, and multiple load ends can also be distributed to by server and are carried out together
When dispose to realize that the multiple groups of corresponding user terminal uploads data to be tested are carried out anomaly data detection by multi-threaded parallel.?
That is, being deployed with docker container and load end corresponding with docker container in server.
Second is that user terminal, for uploading training dataset or uploading data to be tested.
When server detects the data to be tested of user terminal uploads, server receives the data to be tested, with right
The data to be tested carry out abnormality detection.
In one embodiment, as shown in figure 3, before step S110 further include:
S101, training dataset is received, the isolated forest mould of outlier detection is used for according to the corresponding building of training dataset
Type;
S102, obtain the corresponding user object mark of isolated forest model corresponding with the training dataset, obtain and
The user object identifies corresponding application container engine, and the training dataset is isolated to forest model accordingly and is stored to right
The application container engine answered.
It in the present embodiment, need to be corresponding according to the training dataset after server has received the training dataset
Building is used for the isolated forest model of outlier detection.It is detected, can quickly be identified to be measured by isolated forest model
Try the exceptional value in data.
In one embodiment, as shown in figure 4, step S101 includes:
S1011, random acquisition data attribute and split values corresponding with the data attribute are concentrated from training data;
S1012, training dataset is divided to obtain multiple isolated trees according to shown data attribute and the split values;
S1013, it is combined to obtain the isolated forest model for outlier detection by multiple isolated trees.
In the present embodiment, such as from training dataset D={ d1, d2 .., dn } a data attribute A is randomly choosed
With a split values p1;Then each data object di is concentrated to training data, is drawn according to the split values p1 of data attribute A
Point.If di (A) is less than p, be placed on left subtree, it is on the contrary then in right subtree.Randomly choose a data attribute B and one again at this time
Split values p2;Then left subtree and right subtree are divided all in accordance with according to the split values p2 of data attribute B, is obtained and Zuo Zi
Set corresponding secondary left subtree and secondary right subtree, and secondary left subtree corresponding with right subtree and secondary right subtree.With this
Iteration, until meeting one of condition: (1) being left a data or a plurality of identical data in D;(2) isolated tree reaches
Maximum height.Since each isolated tree is during formation, be randomly derived data attribute and corresponding with data attribute
Split values are different, and which results in can include multiple isolated trees in isolated forest.
In one embodiment, as shown in figure 5, after step S110 further include:
S111, the mark number of user object corresponding to data to be tested is judged whether more than 1, if data to be tested institute is right
The user object mark number answered is less than 1, executes step S120;If user object corresponding to data to be tested identifies number
More than 1, step S112 is executed;
S112, each user object mark in the mark of user object corresponding to the data to be tested is sequentially obtained.
In the present embodiment, judge that the mark number of user object corresponding to data to be tested is to sentence whether more than 1
It is disconnected currently whether to there is multiple and different users to upload data to be tested, to further determine whether that multithreading carries out abnormal data inspection
It surveys.If the mark number of user object corresponding to data to be tested is less than 1, then it represents that only 1 user uploads number to be tested
According to, it need to only judge to identify identical application container engine with the user object of the data to be tested in the server at this time, with
The data to be tested are uploaded to and identify identical application container engine as mesh with the user object of the data to be tested
It marks application container engine and carries out anomaly data detection.If it is more than 1 that user object corresponding to data to be tested, which identifies number, table
It is shown with multiple users and uploads data to be tested respectively, need to only find uploaded respectively with each user in the server at this time
The user objects of data to be tested identify one-to-one application container engine, the data to be tested that each user is uploaded
It is uploaded to respectively with corresponding application container engine as target application container engine progress anomaly data detection.Due to starting
Application container engine identical with the mark number of user object corresponding to data to be tested, may be implemented to the more of abnormal data
Thread parallel detection, and the number disposed parallel is easily extended according to testing requirement.
It whether there is and the number to be tested in each engine identification for the application container engine set that S120, judgement have constructed
According to user object identify identical engine identification.
In the present embodiment, in order to ensure the data to be tested uploaded for different user are targetedly tested,
It needs to search in the server and identifies identical application container engine with the user object of the data to be tested, and by corresponding
Application container engine the data to be tested that user uploads are tested.Such as user 1 uploads the first training dataset, according to
The corresponding building of first training dataset is used for the first isolated forest model of outlier detection, builds first in server at this time and answers
With container engine to store the described first isolated forest model, i.e., set the user object mark of the first application container engine to
User 1.If detecting later, user 1 uploads data to be tested, need to first judge to set in server with the presence or absence of user object mark
It is set to the application container engine of user 1, if the application container engine for being set as user 1 is identified in server there are user object,
The isolated forest model for being set as being stored in the application container engine of user 1 can be identified by user object carries out abnormal number
According to detection.And so on, user N (wherein N is the positive integer greater than 1) uploads the first training dataset, according to the first training
The corresponding building of data set is used for the first isolated forest model of outlier detection, builds the first application container in server at this time and draws
It holds up to store the described first isolated forest model, i.e., sets user N for the user object mark of the first application container engine.It
If detecting afterwards, user N uploads data to be tested, need to first judge to be set as user N with the presence or absence of user object mark in server
Application container engine can pass through use if there are user object marks to be set as the application container engine of user N in server
Family object identity is set as the detection that the isolated forest model stored in the application container engine of user N carries out abnormal data.
If there is the use with the data to be tested in each engine identification of S130, the application container engine set constructed
It is right to obtain engine identification institute identical with the user object of the data to be tested mark for object identity identical engine identification in family
The application container engine answered is isolated as target application container engine according to packaged in the target application container engine
Forest model carries out anomaly data detection to the data to be tested, if the data to be tested are abnormal data, will it is described to
Test data is saved to preset store path.
In the present embodiment, if stored in the server and exist in the application container engine that has constructed with it is described to be measured
The user object for trying data identifies identical application container engine, indicates to upload the user terminals of the data to be tested basis
Its testing requirement uploads training dataset, and constructs isolated forest model in the corresponding application container engine of user terminal.
The data to be tested of the user terminal uploads are sent in corresponding application container engine at this time and are deposited by wherein encapsulation
The isolated forest model of storage carries out anomaly data detection to the data to be tested, if the data to be tested are abnormal data,
The data to be tested are saved to preset store path.
In one embodiment, as shown in fig. 6, step S130 includes:
Include in packaged isolated forest model in S131, the acquisition target application container engine is multiple isolated
Tree;
If being obtained in S132, multiple isolated trees with the presence of the test metadata for including in the isolated tree data to be tested
Depth value of the metadata in corresponding isolated tree is respectively tested in the data to be tested, using as corresponding with each test metadata
Target depth value set;
S133, the average value for obtaining each depth value in each target depth value set, using as with the data to be tested
In the one-to-one average depth value of each test metadata;
If S134, existing beyond pre- with respectively being tested in the data to be tested in the one-to-one average depth value of metadata
The average depth value for the depth threshold being first arranged will exceed the corresponding test metadata envelope of average depth value of the depth threshold
Dress is abnormal data set, and abnormal data set corresponding with the data to be tested is saved to preset store path.
In the present embodiment, by isolated forest model packaged in the target application container engine to described to be measured
It is each test metadata that will include in the data to be tested when being detected in examination data with the presence or absence of abnormal data
(such as the data to be tested be considered as 1 include 100 metadata data acquisition system, then wherein included each metadata
It is considered as a test metadata) it is input to the isolated forest model and carries out abnormality detection, once some test metadata
Average depth value exceed the depth threshold, which is added abnormal data set, until to the number to be tested
Metadata is respectively tested in completes whether average depth value exceeds the judgement of the depth threshold to determine whether to be added abnormal number
After set, abnormal data set corresponding with the data to be tested can be obtained.
When calculating mean depth of each test metadata in isolated forest model in each isolated tree, calculation formula is such as
Under:
Wherein, in formula 1, Score indicates the average depth value for the test metadata k for including in the data to be tested, m
Indicate to isolate the isolated tree number that there is test metadata k in multiple isolated trees that forest model includes, depth (i) expression is deposited
In the m isolated tree isolated tree of test metadata k in i-th of isolated tree test metadata k depth value.For example, can be with
It is 3 that depth threshold, which is arranged, if ScorekIf 3, then it is assumed that data to be tested are abnormal data.By isolating forest model
It is detected, can quickly identify the exceptional value in data to be tested.
If being not present and the data to be tested in each engine identification of S140, the application container engine set constructed
User object identifies identical engine identification, and there will be no the prompt information hairs with user object mark respective application container engine
It send to corresponding user terminal.
There is no identical application container engine is identified with the user objects of data to be tested even in server, need pair
The prompt information that respective application container engine is identified with user object is not present in the user terminal for uploading data to be tested, with
If informing, the user terminal has testing requirement, can upload training dataset first to obtain corresponding isolated forest model to carry out
Subsequent anomaly data detection.
This method improves at data the method achieve efficiently detecting exceptional value, and by way of disposing parallel
The efficiency of reason.
The embodiment of the present invention also provides a kind of anomaly data detection device, and the anomaly data detection device is aforementioned for executing
Any embodiment of abnormal deviation data examination method.Specifically, referring to Fig. 7, Fig. 7 is abnormal data provided in an embodiment of the present invention
The schematic block diagram of detection device.The anomaly data detection device 100 can be configured in server.
As shown in fig. 7, anomaly data detection device 100 includes data receipt unit 110 to be tested, mark judging unit
120, abnormality detecting unit 130, prompt unit 140.
Data receipt unit 110 to be tested, if for detecting the data to be tested of user terminal uploads, receive it is described to
Test data.
I.e. when server detects the data to be tested of user terminal uploads, server receives the data to be tested, with
The data to be tested are carried out abnormality detection.
In one embodiment, as shown in figure 8, anomaly data detection device 100 further include:
Isolated forest model training unit 101 is used for for receiving training dataset according to the corresponding building of training dataset
The isolated forest model of outlier detection;
Category of model storage unit 102, it is corresponding for obtaining isolated forest model corresponding with the training dataset
User object mark obtains application container engine corresponding with user object mark, and the training dataset is corresponding
Isolated forest model is stored to corresponding application container engine.
It in the present embodiment, need to be corresponding according to the training dataset after server has received the training dataset
Building is used for the isolated forest model of outlier detection.It is detected, can quickly be identified to be measured by isolated forest model
Try the exceptional value in data.
In one embodiment, as shown in figure 9, isolated forest model training unit 101 includes:
Splitting parameter acquiring unit 1011, for from training data concentrate it is random obtain data attribute and with the data category
The corresponding split values of property;
Isolated tree training unit 1012, for being drawn training dataset according to shown data attribute and the split values
Get multiple isolated trees;
Isolated forest model acquiring unit 1013, for combining to obtain for the isolated of outlier detection by multiple isolated trees
Forest model.
In the present embodiment, such as from training dataset D={ d1, d2 .., dn } a data attribute A is randomly choosed
With a split values p1;Then each data object di is concentrated to training data, is drawn according to the split values p1 of data attribute A
Point.If di (A) is less than p, be placed on left subtree, it is on the contrary then in right subtree.Randomly choose a data attribute B and one again at this time
Split values p2;Then left subtree and right subtree are divided all in accordance with according to the split values p2 of data attribute B, is obtained and Zuo Zi
Set corresponding secondary left subtree and secondary right subtree, and secondary left subtree corresponding with right subtree and secondary right subtree.With this
Iteration, until meeting one of condition: (1) being left a data or a plurality of identical data in D;(2) isolated tree reaches
Maximum height.Since each isolated tree is during formation, be randomly derived data attribute and corresponding with data attribute
Split values are different, and which results in can include multiple isolated trees in isolated forest.
In one embodiment, as shown in Figure 10, anomaly data detection device 100 further include:
Number judging unit 111, for judge user object corresponding to data to be tested mark number whether more than 1,
If the mark number of user object corresponding to data to be tested is less than 1, the application container engine for judging to have constructed is executed
The step of in each engine identification of set with the presence or absence of identical engine identification is identified with the user object of the data to be tested;
If it is more than 1 that user object corresponding to data to be tested, which identifies number, execution is sequentially obtained corresponding to the data to be tested
In user object mark the step of each user object mark;
Sequentially acquiring unit 112, it is each in the mark of user object corresponding to the data to be tested for sequentially obtaining
User object mark.
In the present embodiment, judge that the mark number of user object corresponding to data to be tested is to sentence whether more than 1
It is disconnected currently whether to there is multiple and different users to upload data to be tested, to further determine whether that multithreading carries out abnormal data inspection
It surveys.If the mark number of user object corresponding to data to be tested is less than 1, then it represents that only 1 user uploads number to be tested
According to, it need to only judge to identify identical application container engine with the user object of the data to be tested in the server at this time, with
The data to be tested are uploaded to and identify identical application container engine as mesh with the user object of the data to be tested
It marks application container engine and carries out anomaly data detection.If it is more than 1 that user object corresponding to data to be tested, which identifies number, table
It is shown with multiple users and uploads data to be tested respectively, need to only find uploaded respectively with each user in the server at this time
The user objects of data to be tested identify one-to-one application container engine, the data to be tested that each user is uploaded
It is uploaded to respectively with corresponding application container engine as target application container engine progress anomaly data detection.Due to starting
Application container engine identical with the mark number of user object corresponding to data to be tested, may be implemented to the more of abnormal data
Thread parallel detection, and the number disposed parallel is easily extended according to testing requirement.
Judging unit 120 is identified, whether is deposited in each engine identification of the application container engine set for judging to have constructed
Identical engine identification is being identified with the user object of the data to be tested.
In the present embodiment, in order to ensure the data to be tested uploaded for different user are targetedly tested,
It needs to search in the server and identifies identical application container engine with the user object of the data to be tested, and by corresponding
Application container engine the data to be tested that user uploads are tested.Such as user 1 uploads the first training dataset, according to
The corresponding building of first training dataset is used for the first isolated forest model of outlier detection, builds first in server at this time and answers
With container engine to store the described first isolated forest model, i.e., set the user object mark of the first application container engine to
User 1.If detecting later, user 1 uploads data to be tested, need to first judge to set in server with the presence or absence of user object mark
It is set to the application container engine of user 1, if the application container engine for being set as user 1 is identified in server there are user object,
The isolated forest model for being set as being stored in the application container engine of user 1 can be identified by user object carries out abnormal number
According to detection.And so on, user N (wherein N is the positive integer greater than 1) uploads the first training dataset, according to the first training
The corresponding building of data set is used for the first isolated forest model of outlier detection, builds the first application container in server at this time and draws
It holds up to store the described first isolated forest model, i.e., sets user N for the user object mark of the first application container engine.It
If detecting afterwards, user N uploads data to be tested, need to first judge to be set as user N with the presence or absence of user object mark in server
Application container engine can pass through use if there are user object marks to be set as the application container engine of user N in server
Family object identity is set as the detection that the isolated forest model stored in the application container engine of user N carries out abnormal data.
Abnormality detecting unit 130, if existing and institute in each engine identification of the application container engine set for having constructed
The user object for stating data to be tested identifies identical engine identification, obtains and identifies phase with the user object of the data to be tested
Application container engine corresponding to same engine identification draws as target application container engine according to the target application container
Packaged isolated forest model carries out anomaly data detection to the data to be tested in holding up, if the data to be tested are different
Regular data saves the data to be tested to preset store path.
In the present embodiment, if stored in the server and exist in the application container engine that has constructed with it is described to be measured
The user object for trying data identifies identical application container engine, indicates to upload the user terminals of the data to be tested basis
Its testing requirement uploads training dataset, and constructs isolated forest model in the corresponding application container engine of user terminal.
The data to be tested of the user terminal uploads are sent in corresponding application container engine at this time and are deposited by wherein encapsulation
The isolated forest model of storage carries out anomaly data detection to the data to be tested, if the data to be tested are abnormal data,
The data to be tested are saved to preset store path.
In one embodiment, as shown in figure 11, abnormality detecting unit 130 includes:
Data positioning unit 131, for obtaining in isolated forest model packaged in the target application container engine
Including multiple isolated trees;
Target depth value set acquiring unit 132, if in multiple isolated trees with the presence of the isolated tree number to be tested
The test metadata for including in obtains the depth value that metadata is respectively tested in the data to be tested in corresponding isolated tree,
Using as target depth value set corresponding with each test metadata;
Average depth value acquiring unit 133, for obtaining the average value of each depth value in each target depth value set, with
As with the one-to-one average depth value of each test metadata in the data to be tested;
Abnormal data retransmission unit 134, if for respectively test metadata in the data to be tested and put down correspondingly
There is the average depth value beyond pre-set depth threshold in equal depth value, will exceed the mean depth of the depth threshold
It is worth corresponding test metadata and is encapsulated as abnormal data set, abnormal data set corresponding with the data to be tested is saved
To preset store path.
In the present embodiment, by isolated forest model packaged in the target application container engine to described to be measured
It is each test metadata that will include in the data to be tested when being detected in examination data with the presence or absence of abnormal data
(such as the data to be tested be considered as 1 include 100 metadata data acquisition system, then wherein included each metadata
It is considered as a test metadata) it is input to the isolated forest model and carries out abnormality detection, once some test metadata
Average depth value exceed the depth threshold, which is added abnormal data set, until to the number to be tested
Metadata is respectively tested in completes whether average depth value exceeds the judgement of the depth threshold to determine whether to be added abnormal number
After set, abnormal data set corresponding with the data to be tested can be obtained.Each test metadata is being calculated isolated
When mean depth in forest model in each isolated tree, calculation formula such as formula 1.
Prompt unit 140, if in each engine identification of the application container engine set for having constructed there is no with it is described
The user object of data to be tested identifies identical engine identification, and there will be no identify respective application container engine with user object
Prompt information be sent to corresponding user terminal.
There is no identical application container engine is identified with the user objects of data to be tested even in server, need pair
The prompt information that respective application container engine is identified with user object is not present in the user terminal for uploading data to be tested, with
If informing, the user terminal has testing requirement, can upload training dataset first to obtain corresponding isolated forest model to carry out
Subsequent anomaly data detection.
The arrangement achieves efficiently detecting exceptional value, and by way of disposing parallel, improve the effect of data processing
Rate.
Above-mentioned anomaly data detection device can be implemented as the form of computer program, which can such as scheme
It is run in computer equipment shown in 12.
Figure 12 is please referred to, Figure 12 is the schematic block diagram of computer equipment provided in an embodiment of the present invention.The computer is set
Standby 500 be server.Wherein, server can be independent server, be also possible to the server set of multiple server compositions
Group.
Refering to fig. 12, which includes processor 502, memory and the net connected by system bus 501
Network interface 505, wherein memory may include non-volatile memory medium 503 and built-in storage 504.
The non-volatile memory medium 503 can storage program area 5031 and computer program 5032.The computer program
5032 are performed, and processor 502 may make to execute abnormal deviation data examination method.
The processor 502 supports the operation of entire computer equipment 500 for providing calculating and control ability.
The built-in storage 504 provides environment for the operation of the computer program 5032 in non-volatile memory medium 503, should
When computer program 5032 is executed by processor 502, processor 502 may make to execute abnormal deviation data examination method.
The network interface 505 is for carrying out network communication, such as the transmission of offer data information.Those skilled in the art can
To understand, structure shown in Figure 12, only the block diagram of part-structure relevant to the present invention program, is not constituted to this hair
The restriction for the computer equipment 500 that bright scheme is applied thereon, specific computer equipment 500 may include than as shown in the figure
More or fewer components perhaps combine certain components or with different component layouts.
Wherein, the processor 502 is for running computer program 5032 stored in memory, to realize following function
Can: if detecting the data to be tested of user terminal uploads, receive the data to be tested;Judge that the application container constructed is drawn
It holds up in each engine identification of set and identifies identical engine identification with the presence or absence of with the user object of the data to be tested;If
Exist in each engine identification of the application container engine set of building identical with the user object of the data to be tested mark
The user object of engine identification, acquisition and the data to be tested identifies application container corresponding to identical engine identification and draws
It holds up, as target application container engine, according to isolated forest model packaged in the target application container engine to described
Data to be tested carry out anomaly data detection, if the data to be tested are abnormal data, by the data to be tested save to
Preset store path;And if in each engine identification of the application container engine set constructed there is no with it is described to be tested
The user object of data identifies identical engine identification, and there will be no the prompts with user object mark respective application container engine
Information is sent to corresponding user terminal.
In one embodiment, it if processor 502 is executing the data to be tested for detecting user terminal uploads, receives
It before the step of data to be tested, also performs the following operations: receiving training dataset, constructed according to training dataset is corresponding
Isolated forest model for outlier detection;Obtain the isolated corresponding user of forest model corresponding with the training dataset
Object identity obtains application container engine corresponding with user object mark, the training dataset is isolated accordingly
Forest model is stored to corresponding application container engine.
In one embodiment, processor 502 is executing the reception training dataset, constructs according to training dataset is corresponding
It when step for the isolated forest model of outlier detection, performs the following operations: being concentrated from training data and random obtain data
Attribute and split values corresponding with the data attribute;Training dataset is carried out according to shown data attribute and the split values
Division obtains multiple isolated trees;It is combined to obtain the isolated forest model for outlier detection by multiple isolated trees.
In one embodiment, it if processor 502 is executing the data to be tested for detecting user terminal uploads, receives
It after the step of data to be tested, also performs the following operations: judging the mark of user object corresponding to data to be tested
Whether number is more than 1, if the mark number of user object corresponding to data to be tested is less than 1, executes what the judgement had constructed
Draw in each engine identification of application container engine set with the presence or absence of identical with the user object of the data to be tested mark
The step of holding up mark;If it is more than 1 that user object corresponding to data to be tested, which identifies number, execution sequentially obtains described to be tested
In the mark of user object corresponding to data the step of each user object mark;It sequentially obtains corresponding to the data to be tested
User object mark in each user object mark.
In one embodiment, processor 502 is executing the orphan packaged by the target application container engine
Vertical forest model carries out anomaly data detection to the data to be tested, will be described if the data to be tested are abnormal data
When data to be tested are saved to the step of preset store path, performs the following operations: obtaining the target application container engine
In include in packaged isolated forest model multiple isolated trees;If described to be tested with the presence of isolated tree in multiple isolated trees
The test metadata for including in data obtains the depth that metadata is respectively tested in the data to be tested in corresponding isolated tree
Value, using as target depth value set corresponding with each test metadata;Obtain each depth in each target depth value set
The average value of value, using as with the one-to-one average depth value of each test metadata in the data to be tested;If with institute
It states respectively to test in data to be tested in the one-to-one average depth value of metadata and exist beyond pre-set depth threshold
Average depth value, the corresponding test metadata of average depth value that will exceed the depth threshold are encapsulated as abnormal data set,
Abnormal data set corresponding with the data to be tested is saved to preset store path.
It will be understood by those skilled in the art that the embodiment of computer equipment shown in Figure 12 is not constituted to computer
The restriction of equipment specific composition, in other embodiments, computer equipment may include components more more or fewer than diagram, or
Person combines certain components or different component layouts.For example, in some embodiments, computer equipment can only include depositing
Reservoir and processor, in such embodiments, the structure and function of memory and processor are consistent with embodiment illustrated in fig. 12,
Details are not described herein.
It should be appreciated that in embodiments of the present invention, processor 502 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable GateArray, FPGA) or other programmable logic devices
Part, discrete gate or transistor logic, discrete hardware components etc..Wherein, general processor can be microprocessor or
The processor is also possible to any conventional processor etc..
Computer readable storage medium is provided in another embodiment of the invention.The computer readable storage medium can be with
For non-volatile computer readable storage medium.The computer-readable recording medium storage has computer program, wherein calculating
If machine program performs the steps of the data to be tested for detecting user terminal uploads when being executed by processor, receive it is described to
Test data;Judge to whether there is and the data to be tested in each engine identification of the application container engine set constructed
User object identifies identical engine identification;If in each engine identification of the application container engine set constructed exist with it is described
The user object of data to be tested identifies identical engine identification, obtains identical as the user object mark of the data to be tested
Engine identification corresponding to application container engine, as target application container engine, according to the target application container engine
In packaged isolated forest model anomaly data detection is carried out to the data to be tested, if the data to be tested are abnormal
Data save the data to be tested to preset store path;And if the application container engine set constructed is each
There is no identical engine identification is identified with the user object of the data to be tested in engine identification, will be not present and user couple
As the prompt information of mark respective application container engine is sent to corresponding user terminal.
In one embodiment, if the data to be tested for detecting user terminal uploads, the data to be tested are received
Before, further includes: receive training dataset, the isolated forest mould of outlier detection is used for according to the corresponding building of training dataset
Type;The corresponding user object mark of isolated forest model corresponding with the training dataset is obtained, is obtained and the user couple
As identifying corresponding application container engine, the training dataset is isolated to forest model accordingly and is stored to corresponding application and is held
Device engine.
In one embodiment, the reception training dataset is used for outlier detection according to the corresponding building of training dataset
Isolated forest model, comprising: concentrated from training data and random obtain data attribute and division corresponding with the data attribute
Value;Training dataset is divided to obtain multiple isolated trees according to shown data attribute and the split values;By multiple isolated
Tree combination obtains the isolated forest model for outlier detection.
In one embodiment, if the data to be tested for detecting user terminal uploads, the data to be tested are received
Later, further includes: judge the mark number of user object corresponding to data to be tested whether more than 1, if data to be tested institute is right
The user object mark number answered is less than 1, executes each engine identification of the application container engine set for judging to have constructed
In with the presence or absence of identical engine identification is identified with the user objects of the data to be tested the step of;If data institute to be tested is right
The user object mark number answered is more than 1, and execution sequentially obtains every in the mark of user object corresponding to the data to be tested
The step of one user object identifies;Sequentially obtain each user object in the mark of user object corresponding to the data to be tested
Mark.
In one embodiment, the isolated forest model packaged by the target application container engine is to described
Data to be tested carry out anomaly data detection, if the data to be tested are abnormal data, by the data to be tested save to
Preset store path, comprising: include in the isolated forest model packaged by obtaining in the target application container engine is more
A isolated tree;If with the presence of the test metadata for including in the isolated tree data to be tested in multiple isolated trees, described in acquisition
Depth value of the metadata in corresponding isolated tree is respectively tested in data to be tested, using as mesh corresponding with each test metadata
Mark depth value set;The average value for obtaining each depth value in each target depth value set, using as with the data to be tested
In the one-to-one average depth value of each test metadata;If being corresponded with metadata is respectively tested in the data to be tested
Average depth value in exist beyond pre-set depth threshold average depth value, will exceed being averaged for the depth threshold
The corresponding test metadata of depth value is encapsulated as abnormal data set, will abnormal data set corresponding with the data to be tested
It saves to preset store path.
It is apparent to those skilled in the art that for convenience of description and succinctly, foregoing description is set
The specific work process of standby, device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
Those of ordinary skill in the art may be aware that unit described in conjunction with the examples disclosed in the embodiments of the present disclosure and algorithm
Step can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and software
Interchangeability generally describes each exemplary composition and step according to function in the above description.These functions are studied carefully
Unexpectedly the specific application and design constraint depending on technical solution are implemented in hardware or software.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In several embodiments provided by the present invention, it should be understood that disclosed unit and method, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only logical function partition, there may be another division manner in actual implementation, can also will be with the same function
Unit set is at a unit, such as multiple units or components can be combined or can be integrated into another system or some
Feature can be ignored, or not execute.In addition, shown or discussed mutual coupling, direct-coupling or communication connection can
Be through some interfaces, the indirect coupling or communication connection of device or unit, be also possible to electricity, mechanical or other shapes
Formula connection.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.Some or all of unit therein can be selected to realize the embodiment of the present invention according to the actual needs
Purpose.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, is also possible to two or more units and is integrated in one unit.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in one storage medium.Based on this understanding, technical solution of the present invention is substantially in other words to existing
The all or part of part or the technical solution that technology contributes can be embodied in the form of software products, should
Computer software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be
Personal computer, server or network equipment etc.) execute all or part of step of each embodiment the method for the present invention
Suddenly.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), magnetic disk or
The various media that can store program code such as person's CD.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of abnormal deviation data examination method characterized by comprising
If detecting the data to be tested of user terminal uploads, the data to be tested are received;
Judge in each engine identification of the application container engine set constructed with the presence or absence of the user with the data to be tested
The identical engine identification of object identity;
If there is the user object mark with the data to be tested in each engine identification of the application container engine set constructed
Know identical engine identification, obtains and identify application corresponding to identical engine identification with the user object of the data to be tested
Container engine, as target application container engine, according to isolated forest model packaged in the target application container engine
Anomaly data detection is carried out to the data to be tested, if the data to be tested are abnormal data, by the data to be tested
It saves to preset store path;And
If there is no the user objects with the data to be tested in each engine identification of the application container engine set constructed
Identical engine identification is identified, it is corresponding there will be no being sent to the prompt information of user object mark respective application container engine
User terminal.
2. abnormal deviation data examination method according to claim 1, which is characterized in that if described detect user terminal uploads
Data to be tested, before receiving the data to be tested, further includes:
Training dataset is received, the isolated forest model of outlier detection is used for according to the corresponding building of training dataset;
The corresponding user object mark of isolated forest model corresponding with the training dataset is obtained, is obtained and the user couple
As identifying corresponding application container engine, the training dataset is isolated to forest model accordingly and is stored to corresponding application and is held
Device engine.
3. abnormal deviation data examination method according to claim 2, which is characterized in that the reception training dataset, according to
The corresponding building of training dataset is used for the isolated forest model of outlier detection, comprising:
Random acquisition data attribute and split values corresponding with the data attribute are concentrated from training data;
Training dataset is divided to obtain multiple isolated trees according to shown data attribute and the split values;
It is combined to obtain the isolated forest model for outlier detection by multiple isolated trees.
4. abnormal deviation data examination method according to claim 1, which is characterized in that if described detect user terminal uploads
Data to be tested, after receiving the data to be tested, further includes:
The mark number of user object corresponding to data to be tested is judged whether more than 1, if user corresponding to data to be tested
Whether object identity number is less than 1, execute and deposit in each engine identification for judging the application container engine set constructed
In the step of identifying identical engine identification with the user object of the data to be tested;If user corresponding to data to be tested
Object identity number is more than 1, and execution sequentially obtains each user couple in the mark of user object corresponding to the data to be tested
As the step of identifying;
Sequentially obtain each user object mark in the mark of user object corresponding to the data to be tested.
5. abnormal deviation data examination method according to claim 1, which is characterized in that described according to the target application container
Packaged isolated forest model carries out anomaly data detection to the data to be tested in engine, if the data to be tested are
Abnormal data saves the data to be tested to preset store path, comprising:
Obtain the multiple isolated trees for including in isolated forest model packaged in the target application container engine;
If being obtained described to be tested in multiple isolated trees with the presence of the test metadata for including in the isolated tree data to be tested
Depth value of the metadata in corresponding isolated tree is respectively tested in data, using as target depth corresponding with each test metadata
Value set;
The average value for obtaining each depth value in each target depth value set, using as with each test in the data to be tested
The one-to-one average depth value of metadata;
If existing with respectively being tested in the data to be tested in the one-to-one average depth value of metadata beyond pre-set
The average depth value of depth threshold, the corresponding test metadata of average depth value that will exceed the depth threshold are encapsulated as exception
Data acquisition system saves abnormal data set corresponding with the data to be tested to preset store path.
6. a kind of anomaly data detection device characterized by comprising
Data receipt unit to be tested, if receiving the number to be tested for detecting the data to be tested of user terminal uploads
According to;
Identify judging unit, whether there is in each engine identification of the application container engine set for judge to have constructed with it is described
The user object of data to be tested identifies identical engine identification;
Abnormality detecting unit, if in each engine identification of the application container engine set for having constructed exist with it is described to be tested
The user object of data identifies identical engine identification, obtains and identifies identical engine with the user object of the data to be tested
The corresponding application container engine of mark is sealed as target application container engine according in the target application container engine
The isolated forest model of dress carries out anomaly data detection to the data to be tested, if the data to be tested are abnormal data,
The data to be tested are saved to preset store path;And
Prompt unit, if being not present and the number to be tested in each engine identification of the application container engine set for having constructed
According to user object identify identical engine identification, there will be no the prompt letters with user object mark respective application container engine
Breath is sent to corresponding user terminal.
7. anomaly data detection device according to claim 6, which is characterized in that further include:
Isolated forest model training unit is used for abnormal point according to the corresponding building of training dataset for receiving training dataset
The isolated forest model of detection;
Category of model storage unit, for obtaining the isolated corresponding user object of forest model corresponding with the training dataset
Mark obtains application container engine corresponding with user object mark, and the training dataset is isolated forest accordingly
Model is stored to corresponding application container engine.
8. anomaly data detection device according to claim 6, which is characterized in that the abnormality detecting unit, comprising:
Data positioning unit, for obtaining in the target application container engine include in packaged isolated forest model more
A isolated tree;
Target depth value set acquiring unit, if in multiple isolated trees with the presence of including in the isolated tree data to be tested
Test metadata, obtain the depth value that metadata is respectively tested in the data to be tested in corresponding isolated tree, using as with
The corresponding target depth value set of each test metadata;
Average depth value acquiring unit, for obtaining the average value of each depth value in each target depth value set, using as with
The one-to-one average depth value of each test metadata in the data to be tested;
Abnormal data retransmission unit, if for respectively test the one-to-one average depth value of metadata in the data to be tested
Middle to there is the average depth value for exceeding pre-set depth threshold, the average depth value that will exceed the depth threshold is corresponding
Test metadata is encapsulated as abnormal data set, and abnormal data set corresponding with the data to be tested is saved to preset
Store path.
9. a kind of computer equipment, including memory, processor and it is stored on the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 5 when executing the computer program
Any one of described in abnormal deviation data examination method.
10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey
Sequence, the computer program execute the processor as described in any one of claim 1 to 5 different
Regular data detection method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910012883.0A CN109828825A (en) | 2019-01-07 | 2019-01-07 | Abnormal deviation data examination method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910012883.0A CN109828825A (en) | 2019-01-07 | 2019-01-07 | Abnormal deviation data examination method, device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109828825A true CN109828825A (en) | 2019-05-31 |
Family
ID=66860208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910012883.0A Pending CN109828825A (en) | 2019-01-07 | 2019-01-07 | Abnormal deviation data examination method, device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109828825A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533108A (en) * | 2019-09-02 | 2019-12-03 | 四川长虹电器股份有限公司 | A kind of sales volume rejecting outliers method based on isolated forest algorithm |
CN110807488A (en) * | 2019-11-01 | 2020-02-18 | 北京芯盾时代科技有限公司 | Anomaly detection method and device based on user peer-to-peer group |
CN112069359A (en) * | 2020-09-01 | 2020-12-11 | 上海熙菱信息技术有限公司 | Method for dynamically filtering abnormal data of comparison result of snapshot object |
CN114861746A (en) * | 2021-12-15 | 2022-08-05 | 平安科技(深圳)有限公司 | Anti-fraud identification method and device based on big data and related equipment |
CN115065597A (en) * | 2022-06-14 | 2022-09-16 | 中国电信股份有限公司 | Container resource allocation method, device, medium and equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106569895A (en) * | 2016-10-24 | 2017-04-19 | 华南理工大学 | Construction method of multi-tenant big data platform based on container |
CN108921440A (en) * | 2018-07-11 | 2018-11-30 | 平安科技(深圳)有限公司 | Pollutant method for monitoring abnormality, system, computer equipment and storage medium |
CN108958881A (en) * | 2018-05-31 | 2018-12-07 | 平安科技(深圳)有限公司 | Data processing method, device and computer readable storage medium |
-
2019
- 2019-01-07 CN CN201910012883.0A patent/CN109828825A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106569895A (en) * | 2016-10-24 | 2017-04-19 | 华南理工大学 | Construction method of multi-tenant big data platform based on container |
CN108958881A (en) * | 2018-05-31 | 2018-12-07 | 平安科技(深圳)有限公司 | Data processing method, device and computer readable storage medium |
CN108921440A (en) * | 2018-07-11 | 2018-11-30 | 平安科技(深圳)有限公司 | Pollutant method for monitoring abnormality, system, computer equipment and storage medium |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110533108A (en) * | 2019-09-02 | 2019-12-03 | 四川长虹电器股份有限公司 | A kind of sales volume rejecting outliers method based on isolated forest algorithm |
CN110807488A (en) * | 2019-11-01 | 2020-02-18 | 北京芯盾时代科技有限公司 | Anomaly detection method and device based on user peer-to-peer group |
CN112069359A (en) * | 2020-09-01 | 2020-12-11 | 上海熙菱信息技术有限公司 | Method for dynamically filtering abnormal data of comparison result of snapshot object |
CN112069359B (en) * | 2020-09-01 | 2024-03-19 | 上海熙菱信息技术有限公司 | Method for dynamically filtering abnormal data of snapshot object comparison result |
CN114861746A (en) * | 2021-12-15 | 2022-08-05 | 平安科技(深圳)有限公司 | Anti-fraud identification method and device based on big data and related equipment |
CN115065597A (en) * | 2022-06-14 | 2022-09-16 | 中国电信股份有限公司 | Container resource allocation method, device, medium and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109828825A (en) | Abnormal deviation data examination method, device, computer equipment and storage medium | |
EP3139297B1 (en) | Malware determination device, malware determination system, malware determination method, and program | |
WO2021151334A1 (en) | Model deployment method and apparatus, and device and storage medium | |
CN106294743A (en) | The recommendation method and device of application function | |
CN109614608A (en) | Electronic device, text information detection method and storage medium | |
CN106548213B (en) | Blood vessel recognition methods and device | |
CN109473169A (en) | A kind of methods for the diagnosis of diseases, device and terminal device | |
CN104427547B (en) | Business and network associate method of testing, apparatus and system | |
CN102360405A (en) | Remote medical health monitoring data submitting and sorting method and system | |
CN109165170A (en) | A kind of method and system automating request for test | |
CN104866764B (en) | A kind of Android phone malware detection method based on object reference figure | |
CN107146160A (en) | Customer insured's analysis on the health status method and server | |
CN107360191A (en) | A kind of file acquisition method, device and storage device | |
RU2013140689A (en) | IDENTIFICATION OF MEDICAL CONCEPTS FOR THE SELECTION OF THE VISUALIZATION PROTOCOL | |
CN106682508B (en) | The checking and killing method and device of virus | |
CN106254141B (en) | A kind of method and terminal of smart home management | |
CN106294106B (en) | The test method and device of Web application system | |
CN110321144A (en) | Recognition methods, device, equipment and the storage medium of offline source code | |
CN111540020B (en) | Method and device for determining target behavior, storage medium and electronic device | |
CN103617261B (en) | Picture content attribute identification method and system | |
CN110196920A (en) | The treating method and apparatus and storage medium and electronic device of text data | |
CN109426598A (en) | A kind of hard disk state indicator lamp control method and device | |
CN114117080A (en) | Medical advice information processing method and device, storage medium and electronic device | |
CN109616113A (en) | Sound control method, phonetic controller and speech control system | |
CN109285009A (en) | It brushes single recognition methods and brushes single identification device |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190531 |
|
RJ01 | Rejection of invention patent application after publication |