CN109947079A - Region method for detecting abnormality and edge calculations equipment based on edge calculations - Google Patents
Region method for detecting abnormality and edge calculations equipment based on edge calculations Download PDFInfo
- Publication number
- CN109947079A CN109947079A CN201910214642.4A CN201910214642A CN109947079A CN 109947079 A CN109947079 A CN 109947079A CN 201910214642 A CN201910214642 A CN 201910214642A CN 109947079 A CN109947079 A CN 109947079A
- Authority
- CN
- China
- Prior art keywords
- random forest
- internet
- things equipment
- equipment
- abnormality detection
- 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
Abstract
The embodiment of the present application discloses a kind of region method for detecting abnormality and edge calculations equipment based on edge calculations, this method comprises: edge calculations equipment obtains the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment, wherein the multiple internet of things equipment belongs to same designated area；Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, to predict whether the specified region is abnormal, wherein, the abnormality detection Random Forest model includes the more random forest decision trees that the monitoring characteristic based on the multiple internet of things equipment is respectively trained.
Description
Technical field
This application involves computer software technical field more particularly to a kind of region abnormality detection sides based on edge calculations
Method and edge calculations equipment.
Background technique
With the development of IOT (Internet of Things), equipment accesses Internet of Things on a large scale, the mass data generated on the terminal device
While commercial value is provided, new challenge is also proposed to big data air control.First, becoming along with Internet of Things development
The initiation of gesture, transaction and payment will gradually get rid of dependence to single mobile device, from currently depending on PC or mobile phone terminal
The means of payment become wearable device such as intelligent watch bracelet, intelligent glasses etc. even take (noninductive) payment of equipment.Its
Secondary, along with the regulatory requirements higher and higher to individual privacy data, the acquisition and use of private data also will be by great
Limitation.
How to cope with the higher and higher regulatory requirements of individual privacy data and realize abnormality detection, it has also become is urgently to be resolved
Problem.
Summary of the invention
The purpose of the embodiment of the present application is to provide a kind of region method for detecting abnormality and edge calculations based on edge calculations
Equipment is carried out abnormality detection with the monitoring characteristic acquired by the internet of things equipment in region.
In order to solve the above technical problems, the embodiment of the present application is achieved in that
In a first aspect, proposing a kind of region method for detecting abnormality based on edge calculations, comprising:
Edge calculations equipment obtains the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment
According to wherein the multiple internet of things equipment belongs to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, with pre-
Survey whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that the monitoring characteristic of net equipment is respectively trained.
Second aspect proposes a kind of edge calculations equipment, comprising:
Module is obtained, the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained,
Wherein the multiple internet of things equipment belongs to same designated area；
Prediction module, using the monitoring characteristic of the multiple internet of things equipment as abnormality detection Random Forest model
Input, to predict whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes based on described
At least partly decision in more random forest decision trees that the monitoring characteristic of multiple internet of things equipment is respectively trained
Tree.
The third aspect proposes a kind of edge calculations equipment, which includes:
Processor；And
It is arranged to the memory of storage computer executable instructions, the executable instruction makes the place when executed
It manages device and executes following operation:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein described more
A internet of things equipment belongs to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, with pre-
Survey whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that the monitoring characteristic of net equipment is respectively trained.
Fourth aspect proposes a kind of computer readable storage medium, the computer-readable recording medium storage one
Or multiple programs, one or more of programs are when the edge calculations equipment for being included multiple application programs executes, so that institute
It states edge calculations equipment and executes following operation:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein described more
A internet of things equipment belongs to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, with pre-
Survey whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that the monitoring characteristic of net equipment is respectively trained.
As can be seen from the technical scheme provided by the above embodiments of the present application, the embodiment of the present application scheme at least has following one kind
Technical effect:
The monitoring characteristic of multiple internet of things equipment of same edge calculations equipment is accessed by acquisition, and is input to base
It is predicted in the abnormality detection forest model of the history monitoring characteristic training of multiple internet of things equipment, thus based on more
The current acquisition data of a internet of things equipment discriminate whether to be abnormal, to realize abnormality detection.
Detailed description of the invention
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The some embodiments recorded in application, for those of ordinary skill in the art, in the premise of not making the creative labor property
Under, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is region abnormality detection schematic diagram of a scenario of the embodiment of the present application based on edge calculations.
Fig. 2 is region method for detecting abnormality flow chart of the one embodiment based on edge calculations of the application.
Fig. 3 is the schematic diagram of the embodiment of the present application edge calculations equipment training abnormality detection Random Forest model.
Fig. 4 is the structural schematic diagram of one embodiment electronic equipment of the application.
Fig. 5 is the structural schematic diagram of one embodiment edge calculations equipment of the application.
Specific embodiment
The embodiment of the present application provides a kind of region method for detecting abnormality and edge calculations equipment based on edge calculations.
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with the application reality
The attached drawing in example is applied, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described implementation
Example is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field is common
The application protection all should belong in technical staff's every other embodiment obtained without creative efforts
Range.
Fig. 1 is region abnormality detection schematic diagram of a scenario of the embodiment of the present application based on edge calculations.
As shown in Figure 1, internet of things equipment can acquire a variety of data, by taking smart phone as an example, smart phone can pass through acceleration
Degree meter acquisition speed data counts acquisition gesture data by near field, movement range data is acquired by gyroscope, pass through pressure gauge
Acquisition touch screen pressure data, by illumination meter acquire intensity of illumination data, by thermometer temperature collection data, pass through hygrometer
It acquires humidity data, RPC data, etc. is acquired by barometer.
After internet of things equipment acquires data, the corresponding characteristic of internet of things equipment can be formed, edge calculations is uploaded to and sets
In standby.Edge calculations equipment can provide IOT terminal and calculate service, and the characteristic uploaded to multiple internet of things equipment is processed
Processing, characteristic needed for forming abnormality detection model, is input in IOT termination decision model and carries out decision.In addition, edge
Characteristic can be also stored in edge calculations server-side by calculating equipment to be backed up, using the IOT as edge calculations equipment
The training sample source of termination decision model.
It should be understood that the limitation of storage capacity and computing capability due to edge calculations equipment, the simple edge calculations that rely on are set
The standby decision carried out abnormality detection, it is possible that biggish erroneous judgement, be based particularly on the decision trees of multiple internet of things equipment into
Row is when judging, adjudicates as exception and judgement when being that normal ratio is relatively close to.At this time, it may be necessary to be carried out by cloud/server-side different
The decision often detected.Wherein, cloud/server-side stores more sample datas, has bigger calculating capacity, court verdict is more
It is accurate.In this way, edge calculations equipment and cloud/server-side can respectively share the decision task of part, for example, respective 50%,
Or 4:6,7:3, etc..
In the following, being further described in conjunction with Fig. 1 to the technical solution of the embodiment of the present application.
Fig. 2 is region method for detecting abnormality flow chart of the one embodiment based on edge calculations of the application.The method of Fig. 1
It can be executed by edge calculations equipment.It should be understood that edge calculations (Edge computing) equipment of the embodiment of the present application, refers to and leans on
The network edge side apparatus of nearly data source header, can converged network, calculating, storage, application core ability etc. open platform, nearby
The crucial requirement for servicing and meeting security and privacy protection etc. is provided.In the embodiment of the present application, the method for Fig. 1 can wrap
It includes:
S210 obtains the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment.
It should be understood that the monitoring characteristic of the embodiment of the present application internet of things equipment acquisition, may include internet of things equipment sheet
At least one of the environmental data of operation characteristic data and the internet of things equipment acquisition of body.
The environmental data of internet of things equipment acquisition, the data acquired when may include internet of things equipment monitoring environment, for example,
The environment temperature of thermometer acquisition, the air humidity of hygrometer acquisition, the geostatic pressure data of geostatic pressure meter acquisition, microphone
The sonic data of acquisition, the image data, etc. of camera acquisition.
The operation characteristic data of internet of things equipment itself may include the operating characteristics that user's operation internet of things equipment generates
Data also may include other operation datas of internet of things equipment timing monitoring collection.What user's operation internet of things equipment generated
Operating characteristics data, such as the speed data acquired in smart phone by accelerometer, the posture number acquired by near field meter
According to, the movement range data that are acquired by gyroscope, the touch screen pressure data acquired by pressure gauge, acquired by illumination meter
Intensity of illumination data, the temperature data acquired by thermometer are acquired by the humidity data of hygrometer acquisition, by barometer
Data of RPC, etc.
In addition, timestamp information can also be carried in monitoring characteristic, for identifying the time of origin of acquisition data.
Optionally, the multiple internet of things equipment belongs to same designated area.It should be understood that in the embodiment of the present application, it is described
Edge calculations equipment can be monitored the internet of things equipment in specified region, to obtain the monitoring feature of internet of things equipment acquisition
Data.The specified region mentioned in the embodiment of the present application, for example, it may be some smart home, some company office space,
Some mansion, etc..Abnormality detection at this time, can be used for detecting whether the specified region is abnormal.
By taking smart home as an example, the characteristic of internet of things equipment acquisition, such as may include the unlatching thing of intelligent door and window
Part and/or close event, the voice data of microphone acquisition, the light intensity data of light intensity inductor acquisition, the opening thing of refrigerator
Part and/or close event, power-on event of TV, etc..
S220, using the monitoring characteristic of the multiple internet of things equipment as the defeated of abnormality detection Random Forest model
Enter, to predict whether the specified region is abnormal.Wherein, the abnormality detection Random Forest model includes based on described more
At least partly decision tree in more random forest decision trees that the monitoring characteristic of a internet of things equipment is respectively trained.
Or by taking smart home as an example, it is assumed that having a stranger, 10:00 enters the smart home on weekdays, at this time intelligence
The data for the acquisitions such as characteristic, such as intelligent door and window, microphone that internet of things equipment acquires in household will obviously extremely in
Usual data, the history feature data based on internet of things equipment carry out the abnormality detection random forest that random forest training obtains
It is abnormality that model, which will be easily identified out this state,.
In the embodiment of the present application, the monitoring feature of multiple internet of things equipment of same edge calculations equipment is accessed by acquisition
Data, and the abnormality detection forest model for being input to the history monitoring characteristic training based on multiple internet of things equipment carries out
Prediction, so that the current acquisition data based on multiple internet of things equipment differentiate whether specified region is abnormal, to realize region
Abnormality detection.
It should be understood, of course, that if being determined as abnormality, it at this time can be based at the corresponding processing strategie of abnormality
Reason, for example, being sounded an alarm by alert device；Send a warning message to designated person, etc..The embodiment of the present application does not make this
Limitation.
Particularly, when multiple internet of things equipment is wearable device, the use that region is wearable device is specified to use
In the preset range of family periphery, edge calculations equipment is the intelligent terminal using user.At this point, abnormality detection random forest mould
Type may also include the random forest decision tree that the monitoring characteristic training based on intelligent terminal acquisition obtains.
Optionally, if exporting the random forest decision tree ratio of court verdict in the abnormality detection Random Forest model
Less than preset threshold, then the monitoring characteristic of the multiple internet of things equipment is reported to the cloud of the edge calculations equipment access
Server is held, is made decisions with the abnormality detection Random Forest model by the movement server, the movement server
Abnormality detection Random Forest model include the monitoring characteristic based on the multiple internet of things equipment be respectively trained it is more
Random forest decision tree；
The court verdict of the cloud server feedback is received, and is exported the court verdict as prediction result.
As shown in Figure 1, when edge calculations service accurately can not carry out decision, can report cloud/server end by cloud/
Server end carries out decision.It should be understood, of course, that also including that the monitoring based on multiple internet of things equipment is special in cloud/server end
The abnormality detection Random Forest model that sign data training obtains.
Furthermore, it is to be understood that the time span of the data of cloud/server end abnormality detection Random Forest model training can
It is longer with the data of the abnormality detection Random Forest model training than edge calculations equipment, for example, cloud/server end exception
The training data that Random Forest model uses 3 months is detected, the abnormality detection Random Forest model of edge calculations equipment uses 1 week
Training data, etc..
Furthermore, it is to be understood that the decision tree that cloud/server end abnormality detection Random Forest model includes can compare edge
The decision tree for calculating the abnormality detection Random Forest model of equipment is more, for example, cloud/server end abnormality detection is gloomy at random
Woods model includes 20 decision trees, and the abnormality detection Random Forest model of edge calculations equipment includes 6 decision trees, etc..
It should be understood, of course, that further, the method also includes:
The monitoring characteristic of the multiple internet of things equipment of acquisition is uploaded into the cloud server, to carry out institute
State the training of the abnormality detection Random Forest model of cloud server.
Furthermore, it is to be understood that edge calculations equipment can be based on scheduled duration range before current time in the embodiment of the present application
The monitoring characteristic that interior internet of things equipment reports, the corresponding random forest decision tree of training internet of things equipment, one random gloomy
Woods decision tree corresponds to an internet of things equipment.In other words, target internet of things equipment is in the more random forest decision trees
In corresponding target random forest decision tree be to be adopted based on the edge calculations equipment within the scope of scheduled duration before current time
The monitoring characteristic of the target internet of things equipment of collection carries out what Random Forest model training obtained.
Optionally, as one embodiment, when duration of the acquisition time apart from current time of the training sample is greater than
When scheduled duration, training sample weight factor in trained random forest decision tree is 0.
In the embodiment of the present application, by the way that the training of the sample data of scheduled duration will be greater than apart from current time duration interval
Weight is set as 0, thereby may be ensured that the abnormality detection Random Forest model being made of random forest decision tree based on nearest
Sample data is predicted.
Optionally, as another embodiment, at predetermined time intervals, based on institute within the scope of scheduled duration before current time
State the monitoring characteristic of multiple internet of things equipment, respectively to the corresponding random forest decision tree of the multiple internet of things equipment into
Row training, thereby may be ensured that the abnormality detection Random Forest model being made of random forest decision tree based on nearest sample number
According to being predicted.
Fig. 3 is the schematic diagram of the embodiment of the present application edge calculations equipment training abnormality detection Random Forest model.Such as Fig. 3 institute
Show, edge calculations equipment can be based on the monitoring characteristic that each internet of things equipment reports, Independent modeling, beta pruning, assessment etc..It is optional
Ground, the method that edge calculations equipment executes may also include that the random forest of abnormality detection Random Forest model described in maintenance management
Decision tree.
Optionally, as one embodiment, abnormality detection Random Forest model described in edge calculations equipment maintenance and management
Random forest decision tree includes:
Based on the practical abnormal conditions in the predicting abnormality result and the historical time section in historical time section to described
The corresponding more random forest decision trees of multiple internet of things equipment are assessed；
Cut operator is carried out to the abnormality detection Random Forest model according to assessment result.
Optionally, as one embodiment, abnormality detection Random Forest model described in edge calculations equipment maintenance and management
Random forest decision tree includes:
When detecting that the target internet of things equipment in multiple internet of things equipment breaks down, by the target internet of things equipment
It is removed in the corresponding target random forest decision tree of the abnormality detection Random Forest model.
Optionally, as one embodiment, abnormality detection Random Forest model described in edge calculations equipment maintenance and management
Random forest decision tree includes:
When the monitoring characteristic for not receiving the target internet of things equipment in multiple internet of things equipment in preset time period,
The target internet of things equipment is removed in the corresponding target random forest decision tree of the abnormality detection Random Forest model.
Optionally, as one embodiment, abnormality detection Random Forest model described in edge calculations equipment maintenance and management
Random forest decision tree includes:
When detecting that newly-increased internet of things equipment accesses the edge calculations equipment, the newly-increased internet of things equipment is acquired
It monitors characteristic and trains the corresponding random forest decision tree of the newly-increased internet of things equipment；
If the number of training of the newly-increased internet of things equipment is greater than the first preset threshold, by the newly-increased Internet of Things
The corresponding random forest decision tree of equipment is added in the abnormality detection Random Forest model.
Optionally, as one embodiment, abnormality detection Random Forest model described in edge calculations equipment maintenance and management
Random forest decision tree includes:
When detecting that newly-increased internet of things equipment accesses the edge calculations equipment, the newly-increased internet of things equipment is acquired
It monitors characteristic and trains the corresponding random forest decision tree of the newly-increased internet of things equipment；
If it is pre- that the acquisition time duration of the monitoring characteristic for training the newly-increased internet of things equipment is greater than second
If threshold value, then the corresponding random forest decision tree of the newly-increased internet of things equipment is added to the abnormality detection random forest mould
In type.
Optionally, as one embodiment, the method also includes:
When predicting that the specified region is abnormal, the payment for accessing the payment devices of the edge calculations equipment is prevented
Operation；And/or
When predicting the specified region no exceptions, allow to access the branch of the payment devices of the edge calculations equipment
Pay operation；
Wherein, the delivery operation that the edge calculations equipment interconnection enters the payment devices carries out abnormality detection.
It is above-mentioned that this specification specific embodiment is described.Other embodiments are in the scope of the appended claims
It is interior.In some cases, the movement recorded in detail in the claims or step can be come according to the sequence being different from embodiment
It executes and desired result still may be implemented.In addition, process depicted in the drawing not necessarily require show it is specific suitable
Sequence or consecutive order are just able to achieve desired result.In some embodiments, multitasking and parallel processing be also can
With or may be advantageous.
Fig. 4 is the structural schematic diagram of one embodiment electronic equipment of the application.Referring to FIG. 4, in hardware view, the electricity
Sub- equipment includes processor, optionally further comprising internal bus, network interface, memory.Wherein, memory may be comprising interior
It deposits, such as high-speed random access memory (Random-Access Memory, RAM), it is also possible to further include non-volatile memories
Device (non-volatile memory), for example, at least 1 magnetic disk storage etc..Certainly, which is also possible that other
Hardware required for business.
Processor, network interface and memory can be connected with each other by internal bus, which can be ISA
(Industry Standard Architecture, industry standard architecture) bus, PCI (Peripheral
Component Interconnect, Peripheral Component Interconnect standard) bus or EISA (Extended Industry Standard
Architecture, expanding the industrial standard structure) bus etc..The bus can be divided into address bus, data/address bus, control always
Line etc..Only to be indicated with a four-headed arrow in Fig. 4, it is not intended that an only bus or a type of convenient for indicating
Bus.
Memory, for storing program.Specifically, program may include program code, and said program code includes calculating
Machine operational order.Memory may include memory and nonvolatile memory, and provide instruction and data to processor.
Processor is from the then operation into memory of corresponding computer program is read in nonvolatile memory, in logical layer
The region abnormal detector based on edge calculations is formed on face.Processor executes the program that memory is stored, and specifically uses
The operation below executing:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein described more
A internet of things equipment belongs to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, with pre-
Survey whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that the monitoring characteristic of net equipment is respectively trained.
The side that the region abnormal detector based on edge calculations disclosed in the above-mentioned embodiment illustrated in fig. 2 such as the application executes
Method can be applied in processor, or be realized by processor.Processor may be a kind of IC chip, with signal
Processing capacity.During realization, each step of the above method can by the integrated logic circuit of the hardware in processor or
The instruction of person's software form is completed.Above-mentioned processor can be general processor, including central processing unit (Central
Processing Unit, CPU), network processing unit (Network Processor, NP) etc.；It can also be Digital Signal Processing
Device (Digital Signal Processor, DSP), specific integrated circuit (Application Specific Integrated
Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other can
Programmed logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute the application implementation
Disclosed each method, step and logic diagram in example.General processor can be microprocessor or the processor can also be with
It is any conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present application, can be embodied directly in hardware decoding
Processor executes completion, or in decoding processor hardware and software module combination execute completion.Software module can position
In random access memory, flash memory, read-only memory, programmable read only memory or electrically erasable programmable memory, register
In the storage medium of equal this fields maturation.The storage medium is located at memory, and processor reads the information in memory, in conjunction with it
Hardware completes the step of above method.
The method that the electronic equipment can also carry out Fig. 2, and realize the region abnormal detector based on edge calculations or side
Edge calculates equipment in Fig. 1, the function of embodiment illustrated in fig. 2, and details are not described herein for the embodiment of the present application.
Certainly, other than software realization mode, other implementations are not precluded in the electronic equipment of the application, for example patrol
Collect device or the mode of software and hardware combining etc., that is to say, that the executing subject of following process flow is not limited to each patrol
Unit is collected, hardware or logical device are also possible to.
The embodiment of the present application also proposed a kind of computer readable storage medium, the computer-readable recording medium storage one
A or multiple programs, the one or more program include instruction, and the instruction is when by the portable electronic including multiple application programs
When equipment executes, the method that the portable electronic device can be made to execute embodiment illustrated in fig. 2, and be specifically used for executing following behaviour
Make:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein described more
A internet of things equipment belongs to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, with pre-
Survey whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that the monitoring characteristic of net equipment is respectively trained.
Fig. 5 is the structural schematic diagram of one embodiment edge calculations equipment of the application.Referring to FIG. 5, in a kind of software
In embodiment, edge calculations equipment can include:
Module 510 is obtained, the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained
According to wherein the multiple internet of things equipment belongs to same designated area；
Prediction module 520, using the monitoring characteristic of the multiple internet of things equipment as abnormality detection random forest mould
The input of type, to predict whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes being based on
In the more random forest decision trees that the monitoring characteristic of the multiple internet of things equipment is respectively trained at least partly
Decision tree.
The method that the edge calculations equipment can also carry out Fig. 2, and realize the region abnormal detector based on edge calculations
Or edge calculations equipment, in Fig. 1, the function of embodiment illustrated in fig. 2, details are not described herein for the embodiment of the present application.
In short, being not intended to limit the protection scope of the application the foregoing is merely the preferred embodiment of the application.
Within the spirit and principles of this application, any modification, equivalent replacement, improvement and so on should be included in the application's
Within protection scope.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.It is a kind of typically to realize that equipment is computer.Specifically, computer for example may be used
Think personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media play
It is any in device, navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or these equipment
The combination of equipment.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence " including one ... ", it is not excluded that including described
There is also other identical elements in the process, method of element, commodity or equipment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
Claims (16)
1. a kind of region method for detecting abnormality based on edge calculations, comprising:
Edge calculations equipment obtains the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment,
Described in multiple internet of things equipment belong to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, to predict
State whether specified region is abnormal, wherein the abnormality detection Random Forest model includes setting based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that standby monitoring characteristic is respectively trained.
2. the method as described in claim 1,
The target internet of things equipment corresponding target random forest decision tree in the more random forest decision trees is base
The monitoring characteristic for the target internet of things equipment that the edge calculations equipment acquires within the scope of scheduled duration before current time
It is obtained according to Random Forest model training is carried out.
3. the method as described in claim 1, when the acquisition time of the training sample be greater than apart from the duration at current time it is pre-
When periodically long, weight factor of the training sample in trained random forest decision tree is 0.
4. the method as described in claim 1, the method also includes:
At predetermined time intervals, the monitoring characteristic based on the multiple internet of things equipment within the scope of scheduled duration before current time
According to being trained respectively to the corresponding random forest decision tree of the multiple internet of things equipment.
5. the method as described in claim 1, the method also includes:
If the random forest decision tree ratio for exporting court verdict in the abnormality detection Random Forest model is less than default threshold
The monitoring characteristic of the multiple internet of things equipment is then reported the cloud server of the edge calculations equipment access by value,
With by it is described movement server abnormality detection Random Forest model make decisions, it is described movement server abnormality detection with
Machine forest model includes the more random forests that the monitoring characteristic based on the multiple internet of things equipment is respectively trained
At least partly decision tree in decision tree；
The court verdict of the cloud server feedback is received, and is exported the court verdict as prediction result.
6. method as claimed in claim 5, the method also includes:
The monitoring characteristic of the multiple internet of things equipment of acquisition is uploaded into the cloud server, to carry out the cloud
Hold the training of the abnormality detection Random Forest model of server.
7. the method as described in claim 1, the method also includes:
The random forest decision tree of abnormality detection Random Forest model described in maintenance management.
8. the method for claim 7, the random forest decision tree of abnormality detection Random Forest model described in maintenance management
Include:
Based on the practical abnormal conditions in the predicting abnormality result and the historical time section in historical time section to the multiple
The corresponding more random forest decision trees of internet of things equipment are assessed；
Cut operator is carried out to the abnormality detection Random Forest model according to assessment result.
9. the method for claim 7, the random forest decision tree of abnormality detection Random Forest model described in maintenance management
Include:
When detecting that the target internet of things equipment in multiple internet of things equipment breaks down, by the target internet of things equipment in institute
The corresponding target random forest decision tree of abnormality detection Random Forest model is stated to remove.
10. the method for claim 7, the random forest decision tree of abnormality detection Random Forest model described in maintenance management
Include:
When the monitoring characteristic for not receiving the target internet of things equipment in multiple internet of things equipment in preset time period, by institute
Target internet of things equipment is stated to remove in the corresponding target random forest decision tree of the abnormality detection Random Forest model.
11. the method for claim 7, the random forest decision tree of abnormality detection Random Forest model described in maintenance management
Include:
When detecting that newly-increased internet of things equipment accesses the edge calculations equipment, the monitoring of the newly-increased internet of things equipment is acquired
Characteristic trains the corresponding random forest decision tree of the newly-increased internet of things equipment；
If the number of training of the newly-increased internet of things equipment is greater than the first preset threshold, by the newly-increased internet of things equipment
Corresponding random forest decision tree is added in the abnormality detection Random Forest model.
12. the method for claim 7, the random forest decision tree of abnormality detection Random Forest model described in maintenance management
Include:
When detecting that newly-increased internet of things equipment accesses the edge calculations equipment, the monitoring of the newly-increased internet of things equipment is acquired
Characteristic trains the corresponding random forest decision tree of the newly-increased internet of things equipment；
If the acquisition time duration of the monitoring characteristic for training the newly-increased internet of things equipment is greater than the second default threshold
The corresponding random forest decision tree of the newly-increased internet of things equipment is then added to the abnormality detection Random Forest model by value
In.
13. such as method of any of claims 1-11, the method also includes:
When predicting that the specified region is abnormal, the payment behaviour for accessing the payment devices of the edge calculations equipment is prevented
Make；And/or
When predicting the specified region no exceptions, allow to access the payment behaviour of the payment devices of the edge calculations equipment
Make,
Wherein, the delivery operation that the edge calculations equipment interconnection enters the payment devices carries out abnormality detection.
14. a kind of edge calculations equipment, comprising:
Module is obtained, the monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein
The multiple internet of things equipment belongs to same designated area；
Prediction module, using the monitoring characteristic of the multiple internet of things equipment as the defeated of abnormality detection Random Forest model
Enter, to predict whether the specified region is abnormal, wherein the abnormality detection Random Forest model includes based on described more
At least partly decision tree in more random forest decision trees that the monitoring characteristic of a internet of things equipment is respectively trained.
15. a kind of edge calculations equipment, comprising:
Processor；And
It is arranged to the memory of storage computer executable instructions, the executable instruction makes the processor when executed
Execute following operation:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein the multiple object
Networked devices belong to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, to predict
State whether specified region is abnormal, wherein the abnormality detection Random Forest model includes setting based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that standby monitoring characteristic is respectively trained.
16. a kind of computer readable storage medium, the computer-readable recording medium storage one or more program, described one
A or multiple programs are when the edge calculations equipment for being included multiple application programs executes, so that the edge calculations equipment executes
It operates below:
The monitoring characteristic for accessing multiple internet of things equipment acquisition of the edge calculations equipment is obtained, wherein the multiple object
Networked devices belong to same designated area；
Using the monitoring characteristic of the multiple internet of things equipment as the input of abnormality detection Random Forest model, to predict
State whether specified region is abnormal, wherein the abnormality detection Random Forest model includes setting based on the multiple Internet of Things
At least partly decision tree in more random forest decision trees that standby monitoring characteristic is respectively trained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910214642.4A CN109947079A (en) | 2019-03-20 | 2019-03-20 | Region method for detecting abnormality and edge calculations equipment based on edge calculations |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910214642.4A CN109947079A (en) | 2019-03-20 | 2019-03-20 | Region method for detecting abnormality and edge calculations equipment based on edge calculations |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109947079A true CN109947079A (en) | 2019-06-28 |
Family
ID=67010408
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910214642.4A Pending CN109947079A (en) | 2019-03-20 | 2019-03-20 | Region method for detecting abnormality and edge calculations equipment based on edge calculations |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947079A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110633667A (en) * | 2019-09-11 | 2019-12-31 | 沈阳航空航天大学 | Action prediction method based on multitask random forest |
| CN110749462A (en) * | 2019-07-19 | 2020-02-04 | 华瑞新智科技（北京）有限公司 | Industrial equipment fault detection method and system based on edge calculation |
| CN111091278A (en) * | 2019-12-04 | 2020-05-01 | 湃方科技(天津)有限责任公司 | Edge detection model construction method and device for mechanical equipment anomaly detection |
| CN111427752A (en) * | 2020-06-09 | 2020-07-17 | 北京东方通科技股份有限公司 | Regional anomaly monitoring method based on edge calculation |
| CN111556469A (en) * | 2020-04-08 | 2020-08-18 | 深圳供电局有限公司 | Computing system and computing method for artificial intelligence edge |
| CN111596629A (en) * | 2020-06-02 | 2020-08-28 | 曲阜师范大学 | Cloud-edge-collaborative industrial data fusion method and security controller |
| CN112448861A (en) * | 2019-08-30 | 2021-03-05 | 北京国双科技有限公司 | Edge device processing method and device, storage medium and processor |
| CN112910723A (en) * | 2021-01-15 | 2021-06-04 | 广州穗能通能源科技有限责任公司 | Edge terminal management method, device, equipment and storage medium |
| CN112925680A (en) * | 2021-02-23 | 2021-06-08 | 重庆川仪自动化股份有限公司 | Pipe network monitoring method, system, medium and electronic terminal |
| CN113139817A (en) * | 2021-04-28 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Data classification method, data classification device, medium, and electronic apparatus |
| CN113191455A (en) * | 2021-05-26 | 2021-07-30 | 平安国际智慧城市科技股份有限公司 | Edge computing box election method and device, electronic equipment and medium |
| CN113468162A (en) * | 2021-09-01 | 2021-10-01 | 蘑菇物联技术（深圳）有限公司 | Method, apparatus, and medium for processing for exception data |
| CN114492613A (en) * | 2022-01-20 | 2022-05-13 | 中国人民解放军国防科技大学 | Internet of things and non-Internet of things equipment identification method, system, terminal and readable storage medium |
| CN114567568A (en) * | 2022-03-01 | 2022-05-31 | 北京中电普华信息技术有限公司 | Electric power Internet of things data processing method and device based on edge calculation |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107851057A (en) * | 2015-06-19 | 2018-03-27 | 阿普泰克科技公司 | The Dynamic Execution of forecast model |
| AU2016325186A1 (en) * | 2016-09-09 | 2018-03-29 | Beijing Guodiantong Corporation | Bus load forecasting method |
| CN108334033A (en) * | 2018-02-28 | 2018-07-27 | 中国科学院重庆绿色智能技术研究院 | Punching machine group failure prediction method and its system based on Internet of Things and machine learning |
| CN108512827A (en) * | 2018-02-09 | 2018-09-07 | 世纪龙信息网络有限责任公司 | The identification of abnormal login and method for building up, the device of supervised learning model |
| CN108681288A (en) * | 2018-06-28 | 2018-10-19 | 上海电器科学研究所（集团）有限公司 | A kind of novel maintenance system based on cloud platform |
-
2019
- 2019-03-20 CN CN201910214642.4A patent/CN109947079A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107851057A (en) * | 2015-06-19 | 2018-03-27 | 阿普泰克科技公司 | The Dynamic Execution of forecast model |
| AU2016325186A1 (en) * | 2016-09-09 | 2018-03-29 | Beijing Guodiantong Corporation | Bus load forecasting method |
| CN108512827A (en) * | 2018-02-09 | 2018-09-07 | 世纪龙信息网络有限责任公司 | The identification of abnormal login and method for building up, the device of supervised learning model |
| CN108334033A (en) * | 2018-02-28 | 2018-07-27 | 中国科学院重庆绿色智能技术研究院 | Punching machine group failure prediction method and its system based on Internet of Things and machine learning |
| CN108681288A (en) * | 2018-06-28 | 2018-10-19 | 上海电器科学研究所（集团）有限公司 | A kind of novel maintenance system based on cloud platform |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110749462A (en) * | 2019-07-19 | 2020-02-04 | 华瑞新智科技（北京）有限公司 | Industrial equipment fault detection method and system based on edge calculation |
| CN110749462B (en) * | 2019-07-19 | 2021-05-07 | 华瑞新智科技（北京）有限公司 | Industrial equipment fault detection method and system based on edge calculation |
| CN112448861B (en) * | 2019-08-30 | 2022-07-29 | 北京国双科技有限公司 | Edge device processing method and device, storage medium and processor |
| CN112448861A (en) * | 2019-08-30 | 2021-03-05 | 北京国双科技有限公司 | Edge device processing method and device, storage medium and processor |
| CN110633667A (en) * | 2019-09-11 | 2019-12-31 | 沈阳航空航天大学 | Action prediction method based on multitask random forest |
| CN110633667B (en) * | 2019-09-11 | 2021-11-26 | 沈阳航空航天大学 | Action prediction method based on multitask random forest |
| CN111091278A (en) * | 2019-12-04 | 2020-05-01 | 湃方科技(天津)有限责任公司 | Edge detection model construction method and device for mechanical equipment anomaly detection |
| CN111091278B (en) * | 2019-12-04 | 2023-09-08 | 湃方科技(天津)有限责任公司 | Edge detection model construction method and device for mechanical equipment anomaly detection |
| CN111556469A (en) * | 2020-04-08 | 2020-08-18 | 深圳供电局有限公司 | Computing system and computing method for artificial intelligence edge |
| CN111596629A (en) * | 2020-06-02 | 2020-08-28 | 曲阜师范大学 | Cloud-edge-collaborative industrial data fusion method and security controller |
| CN111427752A (en) * | 2020-06-09 | 2020-07-17 | 北京东方通科技股份有限公司 | Regional anomaly monitoring method based on edge calculation |
| CN112910723B (en) * | 2021-01-15 | 2023-01-31 | 南方电网大数据服务有限公司 | Edge terminal management method, device, equipment and storage medium |
| CN112910723A (en) * | 2021-01-15 | 2021-06-04 | 广州穗能通能源科技有限责任公司 | Edge terminal management method, device, equipment and storage medium |
| CN112925680A (en) * | 2021-02-23 | 2021-06-08 | 重庆川仪自动化股份有限公司 | Pipe network monitoring method, system, medium and electronic terminal |
| CN113139817A (en) * | 2021-04-28 | 2021-07-20 | 北京沃东天骏信息技术有限公司 | Data classification method, data classification device, medium, and electronic apparatus |
| CN113191455A (en) * | 2021-05-26 | 2021-07-30 | 平安国际智慧城市科技股份有限公司 | Edge computing box election method and device, electronic equipment and medium |
| CN113468162A (en) * | 2021-09-01 | 2021-10-01 | 蘑菇物联技术（深圳）有限公司 | Method, apparatus, and medium for processing for exception data |
| CN114492613A (en) * | 2022-01-20 | 2022-05-13 | 中国人民解放军国防科技大学 | Internet of things and non-Internet of things equipment identification method, system, terminal and readable storage medium |
| CN114567568A (en) * | 2022-03-01 | 2022-05-31 | 北京中电普华信息技术有限公司 | Electric power Internet of things data processing method and device based on edge calculation |
| CN114567568B (en) * | 2022-03-01 | 2024-04-05 | 北京中电普华信息技术有限公司 | Electric power Internet of things data processing method and device based on edge calculation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109947079A (en) | Region method for detecting abnormality and edge calculations equipment based on edge calculations | |
| CN102110211B (en) | For the method and apparatus of Administrative Security event | |
| CN106850346B (en) | Method and device for monitoring node change and assisting in identifying blacklist and electronic equipment | |
| EP2453381B1 (en) | System for an engine for forecasting cyber threats and method for forecasting cyber threats using the system | |
| US20150067153A1 (en) | Remote monitoring of data facility in real-time using wireless sensor network | |
| US10831785B2 (en) | Identifying security breaches from clustering properties | |
| CN110866820A (en) | Real-time monitoring system, method, equipment and storage medium for banking business | |
| CN103368904A (en) | Mobile terminal, and system and method for suspicious behavior detection and judgment | |
| US20180062909A1 (en) | Log monitoring system | |
| CN109309579A (en) | Log recording processing method, device, computer equipment and storage medium | |
| CN111131290B (en) | Flow data processing method and device | |
| US11245747B2 (en) | Monitoring a building management system | |
| CN103345439B (en) | A kind of full link monitoring method of health state of information system and device | |
| CN110021150B (en) | Data processing method, device and equipment | |
| CN114862288A (en) | Intelligent water affair secondary pressurization management system | |
| CN116821750A (en) | Data security risk monitoring traceability system based on artificial intelligence | |
| CN106021054A (en) | Method and apparatus for testing upgrading and downgrading stability of BMC | |
| CN109873790A (en) | Network security detection method, device and computer readable storage medium | |
| US9207105B2 (en) | System and method for incident detection with spatiotemporal thresholds estimated via nonparametric quantile regression | |
| CN108920326A (en) | Determine system time-consuming abnormal method, apparatus and electronic equipment | |
| CN111163073A (en) | Flow data processing method and device | |
| CN107193721B (en) | Method and device for generating log | |
| CN116011927A (en) | Supervision inspection method and device, electronic equipment and storage medium | |
| KR101973728B1 (en) | Integration security anomaly symptom monitoring system | |
| CN113076451B (en) | Abnormal behavior identification and risk model library establishment method and device and electronic equipment |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200921 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200921 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right |