CN107526666A - Alarm method, system, device and electronic equipment based on deep learning - Google Patents

Alarm method, system, device and electronic equipment based on deep learning Download PDF

Info

Publication number
CN107526666A
CN107526666A CN201710580182.8A CN201710580182A CN107526666A CN 107526666 A CN107526666 A CN 107526666A CN 201710580182 A CN201710580182 A CN 201710580182A CN 107526666 A CN107526666 A CN 107526666A
Authority
CN
China
Prior art keywords
alarm
deep learning
data
appraising
service data
Prior art date
Application number
CN201710580182.8A
Other languages
Chinese (zh)
Inventor
胡进伟
Original Assignee
阿里巴巴集团控股有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Priority to CN201710580182.8A priority Critical patent/CN107526666A/en
Publication of CN107526666A publication Critical patent/CN107526666A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/30Monitoring
    • G06F11/3065Monitoring arrangements determined by the means or processing involved in reporting the monitored data
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/30Monitoring
    • G06F11/32Monitoring with visual or acoustical indication of the functioning of the machine
    • G06F11/324Display of status information
    • G06F11/327Alarm or error message display
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
    • G06K9/46Extraction of features or characteristics of the image
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/62Methods or arrangements for recognition using electronic means
    • G06K9/6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
    • G06K9/6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting

Abstract

This application discloses the alarm method based on deep learning, system, device and electronic equipment.Methods described includes:History service data are obtained, the history service data reflection is without the regular traffic scene alarmed and/or the abnormal traffic scene that need to be alarmed;Deep learning is carried out using the history service data, obtains appraising model of alarming;The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;According to the estimation result, decide whether to alarm.

Description

Alarm method, system, device and electronic equipment based on deep learning

Technical field

The application is related to computer software technical field, more particularly to the alarm method based on deep learning, system, device And electronic equipment.

Background technology

System alarm is that modern IT enterprises are used to quickly find and solve the effective mechanism of service exception.

In the prior art, the alarm rule that usual human configuration is fixed, and according to these rule analysis daily records, once day Alarm threshold value in the result triggering rule of will analysis, system will send warning message, to notify person skilled to solve Certainly problem.

Based on prior art, it would be desirable to more intelligent alarm scheme.

The content of the invention

The embodiment of the present application provides the alarm method based on deep learning, system, device and electronic equipment, to solve Following technical problem:Based on prior art, it would be desirable to more intelligent alarm scheme.

In order to solve the above technical problems, what the embodiment of the present application was realized in:

A kind of alarm method based on deep learning that the embodiment of the present application provides, including:

History service data are obtained, the regular traffic scene and/or need to report that the history service data reflection need not alarm Alert abnormal traffic scene;

Deep learning is carried out using the history service data, obtains appraising model of alarming;

The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;

According to the estimation result, decide whether to alarm.

A kind of warning system based on deep learning that the embodiment of the present application provides, including:Data Analysis Platform, depth Practise platform, alarm platform;

The Data Analysis Platform obtains history service data, normal industry of the history service data reflection without alarm Business scene and/or the abnormal traffic scene that need to be alarmed;

The history service data that the deep learning platform is obtained using the Data Analysis Platform carry out depth Practise, obtain appraising model of alarming, and the business datum input alarm appraising model to be evaluated is handled, estimated Calculate result;

The estimation result that the alarm platform obtains according to the deep learning platform, decides whether to alarm.

A kind of warning device based on deep learning that the embodiment of the present application provides, including:

Acquisition module, obtain history service data, regular traffic scene of the history service data reflection without alarm And/or the abnormal traffic scene that need to be alarmed;

Study module, deep learning is carried out using the history service data, obtains appraising model of alarming;

Estimation block, the business datum input alarm appraising model to be evaluated is handled, obtains estimation result;

Alarm module, according to the estimation result, decide whether to alarm.

The a kind of electronic equipment that the embodiment of the present application provides, including:

At least one processor;And

The memory being connected with least one processor communication;Wherein,

The memory storage has can be by the instruction of at least one computing device, and the instruction is by described at least one Individual computing device, so that at least one processor can:

History service data are obtained, the regular traffic scene and/or need to report that the history service data reflection need not alarm Alert abnormal traffic scene;

Deep learning is carried out using the history service data, obtains appraising model of alarming;

The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;

According to the estimation result, decide whether to alarm.

Above-mentioned at least one technical scheme that the embodiment of the present application uses can reach following beneficial effect:Based on depth The alarm appraising model that acquistion is arrived, it can intelligently know regular traffic scene and/or abnormal traffic scene, advantageously reduce wrong report It is alert.

Brief description of the drawings

, below will be to embodiment or existing in order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments described in application, for those of ordinary skill in the art, do not paying the premise of creative labor Under, other accompanying drawings can also be obtained according to these accompanying drawings.

Fig. 1 is a kind of schematic flow sheet for alarm method based on deep learning that the embodiment of the present application provides;

Fig. 2 is that a kind of structure for warning system based on deep learning corresponding to Fig. 1 that the embodiment of the present application provides is shown It is intended to;

Fig. 3 is that a kind of work for warning system based on deep learning corresponding to Fig. 1 that the embodiment of the present application provides is former Manage schematic diagram;

Fig. 4 is a kind of working-flow schematic diagram corresponding to Fig. 3 that the embodiment of the present application provides;

Fig. 5 is that a kind of structure for warning device based on deep learning corresponding to Fig. 1 that the embodiment of the present application provides is shown It is intended to.

Embodiment

The embodiment of the present application provides the alarm method based on deep learning, system, device and electronic equipment.

In order that those skilled in the art more fully understand the technical scheme in the application, it is real below in conjunction with the application The accompanying drawing in example is applied, the technical scheme in the embodiment of the present application is clearly and completely described, it is clear that described implementation Example only some embodiments of the present application, rather than whole embodiments.It is common based on the embodiment in the application, this area The every other embodiment that technical staff is obtained under the premise of creative work is not made, it should all belong to the application protection Scope.

In order to make it easy to understand, the core concept of the scheme of the application is illustrated:The method for introducing deep learning is carried out The design of warning system, can be with the ability of imparting system intelligent Understanding business scenario, so as to advantageously reduce false alarm;Not only such as This, with adding up for data volume, the success rate of deep learning will gradually rise.

Below based on above core concept, the scheme of the application is described in detail.

Fig. 1 is a kind of schematic flow sheet for alarm method based on deep learning that the embodiment of the present application provides.The flow Possible executive agent includes but is not limited to can be as server or the following equipment of terminal:Personal computer, medium-sized calculating Machine, computer cluster, mobile phone, tablet personal computer, intelligent wearable device, vehicle device etc..

Flow in Fig. 1 may comprise steps of:

S102:Obtain history service data, history service data reflection without alarm regular traffic scene and/or The abnormal traffic scene that need to be alarmed.

In the embodiment of the present application, history service data can be daily record, such as, business diary, using daily record etc., daily record Typically non-structural data;History service data can also be the structured data in Service Database, such as, business datum table, Index etc..

In the embodiment of the present application, regular traffic scene is without alarm, and abnormal traffic scene needs to alarm, if for just Normal business has entered alarm, then belongs to false alarm.History service data can be analyzed by artificial or machine mode, with Determine the corresponding relation between history service data and regular traffic scene or abnormal traffic scene.

S104:Deep learning is carried out using the history service data, obtains appraising model of alarming.

In the embodiment of the present application, the alarm appraising model obtained using deep learning, the business to input can be passed through Data are estimated, whether belong to regular traffic scene according to the business datum of estimation result intelligently identified input, either It is no to belong to abnormal traffic scene, and then, can be using the output for appraising model of alarming as the foundation for currently whether needing to alarm.

Estimation result can be alarm appraising model output or alarm appraising model output before centre Status data, such as, data caused by the hidden layer of neutral net.

The embodiment of the present application is not limited the representation of estimation result, for example can be numerical value, vector, boolean Value or character string etc..

S106:The business datum input alarm appraising model to be evaluated is handled, obtains estimation result.

In the embodiment of the present application, business datum to be evaluated can be the new caused business for needing to carry out alarm monitoring Data;Business datum to be evaluated can also be:Alarm decision-making was carried out by other schemes (such as prior art), still Because whether the uncertain alarm result of decision is correct, it is therefore desirable to the business datum further verified.

S108:According to the estimation result, decide whether to alarm.

In the embodiment of the present application, when decision will alarm, specific actuation of an alarm can be by the execution of the flow in Fig. 1 Main body performs, and can also be performed by another main body.

The executive agent of action involved by Fig. 1 can be same equipment or same program or distinct device or Distinct program.For example step S102 executive agent is Data Analysis Platform, step S104, S106 executive agent is depth Learning platform, step S108 executive agent is alarm platform;For another example, the executive agent of each step of the flow in Fig. 1 is equal For same distributed type assemblies;Etc..

By Fig. 1 method, the alarm appraising model obtained based on deep learning, regular traffic scene can be intelligently known And/or abnormal traffic scene, advantageously reduce false alarm.

Further, in the case where alarm appraising model is practically applicable into intelligent alarm, can continue while practicality Current alarm appraising model is trained, further improved so as to advantageously allow the accuracy of alarm appraising model, It is more preferable compared to relatively-stationary alarm rule growth in the prior art.

Method based on Fig. 1, the embodiment of the present application additionally provide some specific embodiments of this method, and extension side Case, it is illustrated below.

In the embodiment of the present application, it is described to carry out deep learning using the history service data for step S104, obtain To alarm appraising model, can specifically include:Data cleansing processing is carried out to the history service data, obtains carrying label Training data, the label show that its corresponding training data belongs to the regular traffic scene or the abnormal traffic field Scape;Specified deep neural network is trained using the training data, obtains appraising model of alarming.

Training method in the preceding paragraph belongs to Training, in actual applications, can also be by the way of unsupervised It is trained.

Further, it is described that specified deep neural network is trained using the training data, it can specifically wrap Include:By the training data input deep neural network specified carries out feature extraction, regression model calculates, loss function Contrast, model adjustment.Usually, training process is that iteration is carried out, until training convergence, and then obtain appraising model of alarming.

The scheme of the application can depart from prior art and be independently operated, in this case, described to incite somebody to action for step S102 The business datum input alarm appraising model to be evaluated carries out before processing, can also carry out:Business is monitored;According to It is described to monitor obtained data, the business datum to be evaluated is obtained, for inputting at the alarm appraising model Reason.

The scheme of the application can also coordinate prior art to use.Such as the alarm result of decision to prior art Verified, to filter out the invalid alarm result of decision, so as to be advantageous to prevent false alarm.In this case, for step Rapid S102, it is described that the business datum input alarm appraising model to be evaluated is subjected to before processing, it can also carry out:Acquisition is treated The alarm result of decision of verification;Business datum corresponding to the alarm result of decision is obtained, as the business number to be evaluated According to, for input it is described alarm appraising model handled;Accordingly for step S108, it is described decide whether alarm before, It can also carry out:According to the estimation result, determine whether the alarm result of decision is null result;If so, it can not enter Row alarm.

In the embodiment of the present application, the history industry for carrying out deep learning is used in order to obtain preferable results of learning, in Fig. 1 Data of being engaged in are often magnanimity, can be based on distributed type assemblies, deep learning are carried out using history service data, with efficiently complete Into learning process.

Based on same thinking, the embodiment of the present application additionally provides a kind of alarm based on deep learning corresponding to Fig. 1 System, Fig. 2 are the structural representation of the system.

The system includes:Data Analysis Platform 201, deep learning platform 202, alarm platform 203;

The Data Analysis Platform 201 obtains history service data, and the history service data reflection is without alarming just Normal business scenario and/or the abnormal traffic scene that need to be alarmed;

The history service data that the deep learning platform 202 is obtained using the Data Analysis Platform 201 are carried out Deep learning, appraising model of alarming is obtained, and the business datum input alarm appraising model to be evaluated is handled, Obtain estimation result;

The estimation result that the alarm platform 203 obtains according to the deep learning platform 202, decides whether to report It is alert.

Alternatively, the history service data include the structured data in daily record and/or Service Database.

Alternatively, after the Data Analysis Platform 201 obtains history service data, also the history service data are carried out Data cleansing is handled, and obtains carrying the training data of label, and the label shows that its corresponding training data belongs to described normal Business scenario or the abnormal traffic scene;

The history service data that the deep learning platform 202 is obtained using the Data Analysis Platform 201 are carried out Deep learning, appraising model of alarming is obtained, is specifically included:

The deep learning platform 202 is trained using the training data to specified deep neural network, is obtained Alarm appraising model.

Alternatively, the deep learning platform 202 is instructed using the training data to specified deep neural network Practice, specifically include:

The training data input deep neural network specified is carried out feature and carried by the deep learning platform 202 Take, regression model calculates, loss function contrasts, model adjustment.

Alternatively, the deep learning platform 202 carries out deep learning using the history service data, specifically includes:

The deep learning platform 202 is based on distributed type assemblies, and deep learning is carried out using the history service data, In this case, deep learning platform 202 is specifically as follows distributed deep learning platform.

Alternatively, the Data Analysis Platform 201 inputs business datum to be evaluated in the deep learning platform 202 The alarm appraising model carries out before processing, and business is monitored, and according to the data for monitoring and obtaining, waits to estimate described in acquisition The business datum of calculation, handled for inputting the alarm appraising model.

Alternatively, the Data Analysis Platform 201 inputs business datum to be evaluated in the deep learning platform 202 The alarm appraising model carries out before processing, obtains the alarm result of decision to be verified, and it is corresponding to obtain the alarm result of decision Business datum, as the business datum to be evaluated, handled for inputting the alarm appraising model;

Before the alarm platform 203 decides whether alarm, always according to the estimation result, the alarm result of decision is determined Whether it is null result.

Based on to the explanation of the system, the embodiment of the present application additionally provides the principle schematic of the system above, such as Fig. 3 institutes Show.The part of the system, and the critical workflow being related to of each several part has been shown in particular in Fig. 3, is illustrated with reference to Fig. 4.

Fig. 4 is a kind of working-flow schematic diagram corresponding to Fig. 3 that the embodiment of the present application provides.

Workflow in Fig. 4 mainly includes the following steps that:

S402:Data Analysis Platform cleans to business diary, using the structured data in daily record and Service Database Afterwards, training data is obtained;

S404:Distributed deep learning platform carries out deep learning using training data, obtains appraising model of alarming, its In, learning process includes:The actions such as feature extraction, regression model calculate, loss function contrasts and model adjusts;

S406:Distributed deep learning platform also using obtained alarm appraising model, enters to business datum to be evaluated Row processing, obtains estimation result;

S408:Alarm platform decides whether to alarm according to estimation result.

Based on same thinking, the embodiment of the present application additionally provides a kind of alarm based on deep learning corresponding to Fig. 1 Device, Fig. 5 are the structural representation of the device, and the device can be located in Fig. 1 on the executive agent of flow, including:

Acquisition module 501, obtain history service data, regular traffic field of the history service data reflection without alarm Scape and/or the abnormal traffic scene that need to be alarmed;

Study module 502, deep learning is carried out using the history service data, obtains appraising model of alarming;

Estimation block 503, the business datum input alarm appraising model to be evaluated is handled, obtains estimation knot Fruit;

Alarm module 504, according to the estimation result, decide whether to alarm.

Based on same thinking, the embodiment of the present application additionally provides corresponding a kind of electronic equipment, including:

At least one processor;And

The memory being connected with least one processor communication;Wherein,

The memory storage has can be by the instruction of at least one computing device, and the instruction is by described at least one Individual computing device, so that at least one processor can:

History service data are obtained, the regular traffic scene and/or need to report that the history service data reflection need not alarm Alert abnormal traffic scene;

Deep learning is carried out using the history service data, obtains appraising model of alarming;

The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;

According to the estimation result, decide whether to alarm.

Based on same thinking, the embodiment of the present application additionally provides a kind of corresponding nonvolatile computer storage media, Computer executable instructions are stored with, the computer executable instructions are arranged to:

History service data are obtained, the regular traffic scene and/or need to report that the history service data reflection need not alarm Alert abnormal traffic scene;

Deep learning is carried out using the history service data, obtains appraising model of alarming;

The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;

According to the estimation result, decide whether to alarm.

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 action recorded in detail in the claims or step can be come according to different from the order in embodiment Perform and still can realize desired result.In addition, the process described in the accompanying drawings not necessarily require show it is specific suitable Sequence or consecutive order could realize desired result.In some embodiments, multitasking and parallel processing be also can With or be probably favourable.

Each embodiment in this specification is described by the way of progressive, identical similar portion between each embodiment Divide mutually referring to what each embodiment stressed is the difference with other embodiment.Especially for system, For device, electronic equipment, nonvolatile computer storage media embodiment, because it is substantially similar to embodiment of the method, institute With the fairly simple of description, the relevent part can refer to the partial explaination of embodiments of method.

System, device, electronic equipment, nonvolatile computer storage media and the method that the embodiment of the present application provides are pair Answer, therefore, system, device, electronic equipment, nonvolatile computer storage media also have similar with corresponding method beneficial Technique effect, due to the advantageous effects of method being described in detail above, therefore, repeat no more here correspondingly System, device, electronic equipment, the advantageous effects of nonvolatile computer storage media.

In the 1990s, the improvement for a technology can clearly distinguish be on hardware improvement (for example, Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).So And as the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit. Designer nearly all obtains corresponding hardware circuit by the way that improved method flow is programmed into hardware circuit.Cause This, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, PLD (Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable Gate Array, FPGA)) it is exactly such a integrated circuit, its logic function is determined by user to device programming.By designer Voluntarily programming comes a digital display circuit " integrated " on a piece of PLD, without asking chip maker to design and make Special IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " patrols Volume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development, And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language (Hardware Description Language, HDL), and HDL is also not only a kind of, but have many kinds, such as ABEL (Advanced Boolean Expression Language)、AHDL(Altera Hardware Description Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL (Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby Hardware Description Language) etc., VHDL (Very-High-Speed are most generally used at present Integrated Circuit Hardware Description Language) and Verilog.Those skilled in the art also should This understands, it is only necessary to method flow slightly programming in logic and is programmed into integrated circuit with above-mentioned several hardware description languages, Can is readily available the hardware circuit for realizing the logical method flow.

Controller can be implemented in any suitable manner, for example, controller can take such as microprocessor or processing Device and storage can by the computer of the computer readable program code (such as software or firmware) of (micro-) computing device Read medium, gate, switch, application specific integrated circuit (Application Specific Integrated Circuit, ASIC), the form of programmable logic controller (PLC) and embedded microcontroller, the example of controller include but is not limited to following microcontroller Device:ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320, are deposited Memory controller is also implemented as a part for the control logic of memory.It is also known in the art that except with Pure computer readable program code mode realized beyond controller, completely can be by the way that method and step is carried out into programming in logic to make Controller is obtained in the form of gate, switch, application specific integrated circuit, programmable logic controller (PLC) and embedded microcontroller etc. to come in fact Existing identical function.Therefore this controller is considered a kind of hardware component, and various for realizing to including in it The device of function can also be considered as the structure in hardware component.Or even, can be by for realizing that the device of various functions regards For that not only can be the software module of implementation method but also can be the structure in hardware component.

System, device, module or the unit that above-described embodiment illustrates, it can specifically be realized by computer chip or entity, Or realized by the product with certain function.One kind typically realizes that equipment is computer.Specifically, computer for example may be used Think personal computer, laptop computer, cell phone, camera phone, smart phone, personal digital assistant, media play It is any in device, navigation equipment, electronic mail equipment, game console, tablet PC, wearable device or these equipment The combination of equipment.

For convenience of description, it is divided into various units during description apparatus above with function to describe respectively.Certainly, this is being implemented The function of each unit can be realized in same or multiple softwares and/or hardware during application.

It should be understood by those skilled in the art that, embodiments of the invention can be provided as method, system or computer program Product.Therefore, the embodiment of the present invention can use complete hardware embodiment, complete software embodiment or combine software and hardware side The form of the embodiment in face.Moreover, the embodiment of the present invention can use wherein includes computer available programs in one or more Implement in the computer-usable storage medium (including but is not limited to magnetic disk storage, CD-ROM, optical memory etc.) of code The form of computer program product.

Present specification is with reference to method according to embodiments of the present invention, equipment (system) and computer program product Flow chart and/or block diagram describe.It should be understood that can be by every in computer program instructions implementation process figure and/or block diagram One flow and/or the flow in square frame and flow chart and/or block diagram and/or the combination of square frame.These computers can be provided Processor of the programmed instruction to all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing devices To produce a machine so that produce use by the instruction of computer or the computing device of other programmable data processing devices In the dress for realizing the function of being specified in one flow of flow chart or multiple flows and/or one square frame of block diagram or multiple square frames Put.

These computer program instructions, which may be alternatively stored in, can guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works so that the instruction being stored in the computer-readable memory, which produces, to be included referring to Make the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one square frame of block diagram or The function of being specified in multiple square frames.

These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted Series of operation steps is performed on calculation machine or other programmable devices to produce computer implemented processing, so as in computer or The instruction performed on other programmable devices is provided for realizing in one flow of flow chart or multiple flows and/or block diagram one The step of function of being specified in individual square frame or multiple square frames.

In a typical configuration, computing device includes one or more processors (CPU), input/output interface, net Network interface and internal memory.

Internal memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or The forms such as Nonvolatile memory, such as read-only storage (ROM) or flash memory (flash RAM).Internal memory is computer-readable medium Example.

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 instruction, data structure, the module of program or other data. The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moved State random access memory (DRAM), other kinds of random access memory (RAM), read-only storage (ROM), electric erasable Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage (CD-ROM), Digital versatile disc (DVD) or other optical storages, magnetic cassette tape, the storage of tape magnetic rigid disk or other magnetic storage apparatus Or any other non-transmission medium, the information that can be accessed by a computing device available for storage.Define, calculate according to herein Machine computer-readable recording medium does not include temporary computer readable media (transitory media), such as data-signal and carrier wave of modulation.

It should also be noted that, term " comprising ", "comprising" or its any other variant are intended to nonexcludability Comprising so that process, method, commodity or equipment including a series of elements not only include those key elements, but also wrapping Include the other element being not expressly set out, or also include for this process, method, commodity or equipment intrinsic want Element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that wanted including described Other identical element also be present in the process of element, method, commodity or equipment.

The application can be described in the general context of computer executable instructions, such as program Module.Usually, program module includes performing particular task or realizes routine, program, object, the group of particular abstract data type Part, data structure etc..The application can also be put into practice in a distributed computing environment, in these DCEs, by Task is performed and connected remote processing devices by communication network.In a distributed computing environment, program module can be with In the local and remote computer-readable storage medium including storage device.

Each embodiment in this specification is described by the way of progressive, identical similar portion between each embodiment Divide mutually referring to what each embodiment stressed is the difference with other embodiment.It is real especially for system For applying example, because it is substantially similar to embodiment of the method, so description is fairly simple, related part is referring to embodiment of the method Part explanation.

Embodiments herein is the foregoing is only, is not limited to the application.For those skilled in the art For, the application can have various modifications and variations.All any modifications made within spirit herein and principle, it is equal Replace, improve etc., it should be included within the scope of claims hereof.

Claims (16)

1. a kind of alarm method based on deep learning, including:
Obtain history service data, history service data reflection without alarm regular traffic scene and/or need to alarm Abnormal traffic scene;
Deep learning is carried out using the history service data, obtains appraising model of alarming;
The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;
According to the estimation result, decide whether to alarm.
2. the method as described in claim 1, the history service data include the structure number in daily record and/or Service Database According to.
3. the method as described in claim 1, described to carry out deep learning using the history service data, alarm estimation is obtained Model, specifically include:
Data cleansing processing is carried out to the history service data, obtains carrying the training data of label, the label shows it Corresponding training data belongs to the regular traffic scene or the abnormal traffic scene;
Specified deep neural network is trained using the training data, obtains appraising model of alarming.
4. method as claimed in claim 3, described that specified deep neural network is trained using the training data, Specifically include:
By the training data input deep neural network specified carries out feature extraction, regression model calculates, loss letter Number contrast, model adjustment.
5. the method as described in claim 1, described to carry out deep learning using the history service data, specifically include:
Based on distributed type assemblies, deep learning is carried out using the history service data.
6. the method as described in claim 1, described by the business datum input alarm appraising model to be evaluated Before reason, methods described also includes:
Business is monitored;
According to the data for monitoring and obtaining, the business datum to be evaluated is obtained, for inputting the alarm estimation mould Type is handled.
7. the method as described in claim 1, described by the business datum input alarm appraising model to be evaluated Before reason, methods described also includes:
Obtain the alarm result of decision to be verified;
Business datum corresponding to the alarm result of decision is obtained, as the business datum to be evaluated, for inputting institute Alarm appraising model is stated to be handled;
It is described decide whether alarm before, methods described also includes:
According to the estimation result, determine whether the alarm result of decision is null result.
8. a kind of warning system based on deep learning, including:Data Analysis Platform, deep learning platform, alarm platform;
The Data Analysis Platform obtains history service data, regular traffic field of the history service data reflection without alarm Scape and/or the abnormal traffic scene that need to be alarmed;
The history service data that the deep learning platform is obtained using the Data Analysis Platform carry out deep learning, obtain Handled to alarm appraising model, and by the business datum input alarm appraising model to be evaluated, obtain estimation knot Fruit;
The estimation result that the alarm platform obtains according to the deep learning platform, decides whether to alarm.
9. system as claimed in claim 8, the history service data include the structure number in daily record and/or Service Database According to.
10. system as claimed in claim 8, after the Data Analysis Platform obtains history service data, also to the history Business datum carries out data cleansing processing, obtains carrying the training data of label, and the label shows its corresponding training data Belong to the regular traffic scene or the abnormal traffic scene;
The history service data that the deep learning platform is obtained using the Data Analysis Platform carry out deep learning, obtain To alarm appraising model, specifically include:
The deep learning platform is trained using the training data to specified deep neural network, obtains alarm estimation Model.
11. system as claimed in claim 10, the deep learning platform is using the training data to specified depth god It is trained, specifically includes through network:
The training data input deep neural network specified is carried out feature extraction, returned by the deep learning platform Model calculates, loss function contrasts, model adjustment.
12. system as claimed in claim 8, the deep learning platform carries out depth using the history service data Practise, specifically include:
The deep learning platform is based on distributed type assemblies, and deep learning is carried out using the history service data.
13. system as claimed in claim 8, the Data Analysis Platform is in the deep learning platform by business to be evaluated Appraising model of being alarmed described in data input carries out before processing, and business is monitored, and according to the data for monitoring and obtaining, obtains The business datum to be evaluated, handled for inputting the alarm appraising model.
14. system as claimed in claim 8, the Data Analysis Platform is in the deep learning platform by business to be evaluated Appraising model of being alarmed described in data input carries out before processing, obtains the alarm result of decision to be verified, obtains the alarm decision-making As a result corresponding business datum, as the business datum to be evaluated, for inputting at the alarm appraising model Reason;
Before the alarm platform decides whether alarm, always according to the estimation result, determine the alarm result of decision whether be Null result.
15. a kind of warning device based on deep learning, including:
Acquisition module, obtains history service data, history service data reflection without alarm regular traffic scene and/or The abnormal traffic scene that need to be alarmed;
Study module, deep learning is carried out using the history service data, obtains appraising model of alarming;
Estimation block, the business datum input alarm appraising model to be evaluated is handled, obtains estimation result;
Alarm module, according to the estimation result, decide whether to alarm.
16. a kind of electronic equipment, including:
At least one processor;And
The memory being connected with least one processor communication;Wherein,
The memory storage has can be by the instruction of at least one computing device, and the instruction is by least one place Manage device to perform, so that at least one processor can:
Obtain history service data, history service data reflection without alarm regular traffic scene and/or need to alarm Abnormal traffic scene;
Deep learning is carried out using the history service data, obtains appraising model of alarming;
The business datum input alarm appraising model to be evaluated is handled, obtains estimation result;
According to the estimation result, decide whether to alarm.
CN201710580182.8A 2017-07-17 2017-07-17 Alarm method, system, device and electronic equipment based on deep learning CN107526666A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710580182.8A CN107526666A (en) 2017-07-17 2017-07-17 Alarm method, system, device and electronic equipment based on deep learning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710580182.8A CN107526666A (en) 2017-07-17 2017-07-17 Alarm method, system, device and electronic equipment based on deep learning

Publications (1)

Publication Number Publication Date
CN107526666A true CN107526666A (en) 2017-12-29

Family

ID=60748938

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710580182.8A CN107526666A (en) 2017-07-17 2017-07-17 Alarm method, system, device and electronic equipment based on deep learning

Country Status (1)

Country Link
CN (1) CN107526666A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106095639A (en) * 2016-05-30 2016-11-09 中国农业银行股份有限公司 A kind of cluster subhealth state method for early warning and system
CN106294066A (en) * 2016-08-01 2017-01-04 北京百度网讯科技有限公司 Alert data processing method and device
CN106407082A (en) * 2016-09-30 2017-02-15 国家电网公司 Method and device for alarming information system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106095639A (en) * 2016-05-30 2016-11-09 中国农业银行股份有限公司 A kind of cluster subhealth state method for early warning and system
CN106294066A (en) * 2016-08-01 2017-01-04 北京百度网讯科技有限公司 Alert data processing method and device
CN106407082A (en) * 2016-09-30 2017-02-15 国家电网公司 Method and device for alarming information system

Similar Documents

Publication Publication Date Title
Samek et al. Evaluating the visualization of what a deep neural network has learned
Wang et al. Salient object detection with recurrent fully convolutional networks
CN105912990B (en) The method and device of Face datection
US9405748B2 (en) Probabilistically finding the connected components of an undirected graph
Grindrod et al. Communicability across evolving networks
Miłkowski Explaining the computational mind
CN105550184B (en) A kind of information acquisition method and device
CN107358157B (en) Face living body detection method and device and electronic equipment
Liang et al. Interpretable structure-evolving lstm
JP2016508277A (en) Efficient query processing using histograms in the columnar database
Shukla et al. Efficient algorithms for discrete wavelet transform: with applications to denoising and fuzzy inference systems
US8266149B2 (en) Clustering with similarity-adjusted entropy
Alevizos et al. Probabilistic complex event recognition: A survey
CN104699601B (en) For the method and system for the injection for performing state machine driving
CN108256164A (en) Boolean logic in state machine lattice
Acampora et al. On the temporal granularity in fuzzy cognitive maps
WO2015176305A1 (en) Human-shaped image segmentation method
CN107704269A (en) A kind of method and system based on block chain generation block
US10387769B2 (en) Hybrid memory cell unit and recurrent neural network including hybrid memory cell units
CN107863147A (en) The method of medical diagnosis based on depth convolutional neural networks
CN105893349B (en) Classification tag match mapping method and device
CN104583935A (en) Instruction insertion in state machine engines
CN104487956B (en) The method and system of state vector data in being held up for use state power traction
CN107608862B (en) Monitoring alarm method, monitoring alarm device and computer readable storage medium
US10207111B2 (en) Analyzer for behavioral analysis and parameterization of neural stimulation

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
TA01 Transfer of patent application right

Effective date of registration: 20191209

Address after: P.O. Box 31119, grand exhibition hall, hibiscus street, 802 West Bay Road, Grand Cayman, ky1-1205, Cayman Islands

Applicant after: Innovative advanced technology Co., Ltd

Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands

Applicant before: Alibaba Group Holding Co., Ltd.