CN111753967A - Big data processing system based on deep learning feedback and edge calculation - Google Patents
Big data processing system based on deep learning feedback and edge calculation Download PDFInfo
- Publication number
- CN111753967A CN111753967A CN202010637720.4A CN202010637720A CN111753967A CN 111753967 A CN111753967 A CN 111753967A CN 202010637720 A CN202010637720 A CN 202010637720A CN 111753967 A CN111753967 A CN 111753967A
- Authority
- CN
- China
- Prior art keywords
- data
- data processing
- task
- input
- module
- 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
- 238000012545 processing Methods 0.000 title claims abstract description 155
- 238000013135 deep learning Methods 0.000 title claims abstract description 43
- 238000004364 calculation method Methods 0.000 title claims description 31
- 238000004458 analytical method Methods 0.000 claims abstract description 36
- 230000008859 change Effects 0.000 claims abstract description 21
- 238000009826 distribution Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 9
- 238000013519 translation Methods 0.000 claims description 8
- 230000004048 modification Effects 0.000 claims description 5
- 238000012986 modification Methods 0.000 claims description 5
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 description 12
- 210000002569 neuron Anatomy 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 230000001537 neural effect Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 230000002441 reversible effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000013329 compounding Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Machine Translation (AREA)
Abstract
The invention provides a big data processing system based on deep learning feedback, which comprises a data task receiving module, a data task analyzing module, a deep learning feedback module, a data task distributing module, an edge computing terminal and a cloud data processing system. The data task receiving module receives data processing tasks input from a plurality of mutually independent input terminals; the data task analysis module analyzes at least one data input attribute of the output data processing task; the deep learning feedback module is connected with the data task distribution module, and the data task analysis module transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module. According to the technical scheme, the corresponding data processing mode can be matched based on the attribute parameters, particularly input change parameters, of the input data processing task through deep learning feedback, and the data task under the big data background can be efficiently processed.
Description
Technical Field
The invention belongs to the technical field of big data processing, and particularly relates to a big data processing system based on deep learning feedback and edge calculation.
Background
With the advent of the world of everything interconnection, devices in a network become complex and diverse, and the number thereof is drastically increased. Inevitably, the amount of data transmitted in the network is also increasing dramatically. In the cloud computing era, all data storage and computation are performed by a cloud server, which brings convenience and also generates many demands, for example, when transmission of a large amount of data is faced, the bandwidth of a network is insufficient; when a task with higher timeliness is processed, certain delay exists in the transmission of data in a network and the calculation of a cloud end, and the real-time performance is insufficient; there are security and privacy issues with cloud-side data. Many scholars have also made extensive and intensive discussions of these aspects in an attempt to alleviate and solve these problems. In this case, edge calculation takes place as it is.
The edge computing is originated in the field of media, and means that an open platform integrating network, computing, storage and application core capabilities is adopted on one side close to an object or a data source to provide nearest-end service nearby. The application program is initiated at the edge side, so that a faster network service response is generated, and the basic requirements of the industry in the aspects of real-time business, application intelligence, safety, privacy protection and the like are met. The edge computation is between the physical entity and the industrial connection, or on top of the physical entity. And the cloud computing still can access the historical data of the edge computing.
The chinese patent application with application number CN201810952746.0 proposes a system for processing ultra-large-scale DPI data based on edge calculation, comprising: the gateway is provided with a DPI plug-in unit and is used for carrying out DPI data acquisition and edge calculation; the cloud server is used for storing the DPI data and processing information which cannot be processed by edge calculation; and data transmission, namely connecting the gateway and the cloud server to transmit data. According to the processing system of the ultra-large-scale DPI data based on the edge calculation, the gateway carries out the edge calculation such as data acquisition, data cleaning, special recognition and reported information extraction, the cloud server processes the information which cannot be processed by the edge calculation, the calculation power of the edge calculation of the gateway is fully exerted, the uploading amount of the data is reduced, the occupation of bandwidth is reduced, and the calculation efficiency of the cloud server is improved, so that the efficiency of overall data processing is improved, meanwhile, the risk of data stealing is reduced, especially the private data of a user is improved, and the safety is improved.
The chinese patent application with application number CN201910146350.1 proposes to add an equipment side offload mode and an equipment relay forwarding offload mode, that is, to offload a computation task to an adjacent device with sufficient computation resources based on a D2D link technology or to forward the computation task to an edge cloud server through an adjacent device relay for computation offload, which causes a problem of mode selection. The device side unloading mode and the device relay forwarding mode both relate to the node matching problem, and the social relationship is increased to adjust optimization variables such as the forwarding power and the calculation resource allocation by considering the influence of the mutual relationship between the devices on the node matching. In addition, in consideration of the mobility of equipment and the long-term performance of the system, the invention introduces a time-dependent long-term system optimization target to form a dynamic optimization problem, and realizes the improvement of the IoT calculation unloading performance by analyzing and solving a relevant algorithm.
However, the existing edge computing technology does not give different processing modes according to different types of input data, especially different input environments; in addition, the edge calculation in the prior art does not have the feedback learning capability, so that the task processing efficiency in a big data mode cannot be improved, and the user experience is reduced.
Disclosure of Invention
In order to solve the technical problems, the invention provides a big data processing system based on deep learning feedback, which comprises a data task receiving module, a data task analyzing module, a deep learning feedback module, a data task distributing module, an edge computing terminal and a cloud data processing system. The data task receiving module receives data processing tasks input from a plurality of mutually independent input terminals; the data task analysis module analyzes at least one data input attribute of the output data processing task; the deep learning feedback module is connected with the data task distribution module, and the data task analysis module transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module. According to the technical scheme, the corresponding data processing mode can be matched based on the attribute parameters, particularly input change parameters, of the input data processing task through deep learning feedback, and the data task under the big data background can be efficiently processed.
Specifically, the technical scheme of the invention is realized as follows:
a big data processing system based on deep learning feedback comprises a data task receiving module, a data task analyzing module, a deep learning feedback module, a data task distributing module, an edge computing terminal and a cloud data processing system;
the data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, and the data processing tasks have data input attributes;
it should be noted that, in the technical solution of the present invention, the data processing tasks received from the input terminals independent of each other mean that a large number of concurrent tasks to be processed are generated at the same time;
the data task analysis module is connected with the data task receiving module to receive the data processing task and analyzes and outputs at least one data input attribute of the data processing task;
as one of the prominent contributions of the present invention to the prior art, the data input attribute includes input environment parameters when the plurality of mutually independent input terminals input data processing tasks, and the input environment parameters include hardware parameters, software parameters and input process parameters of the input terminals; the input process parameters comprise change parameters during the input of the data processing task by the input terminal, and the change parameters comprise a time starting point when the input terminal starts to input the data processing task, a time ending point when the input of the current data processing task is ended and/or change operation between the time starting point and the time ending point;
as one of the prominent contributions of the present invention to the prior art, the deep learning feedback module and the data task distribution module are connected to the data task analysis module, and transmit the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module;
as a more specific technical means of the above outstanding contribution, the change operation comprises one or a combination of modification, deletion, rollback, interface switching and quiesce operation.
As a more specific technical means of the above outstanding contribution, the data input attribute includes an input mode of the input terminal inputting the data processing task, and the input mode includes one of or a combination of a keyboard input, a touch input, a mouse selection input, and a voice input.
The data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, the data processing tasks have data input attributes, and the data processing tasks specifically include:
the data task receiving module further comprises a data input sensing module, and the data input sensing module is used for sensing the input mode of the data processing task and the input environment parameters.
The data task analysis module is connected to the data task receiving module to receive the data processing task and analyze and output at least one data input attribute of the data processing task, and further includes:
and the data task analysis module analyzes the data size and the data type of the data processing task.
The edge calculation terminals are multiple, each edge calculation terminal comprises a different edge calculation model, and at least one edge calculation model corresponds to the data type.
Further advantages of the invention will be apparent in the detailed description section in conjunction with the drawings attached hereto.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is an overall framework diagram of a big data processing system based on deep learning feedback according to an embodiment of the invention
FIG. 2 is a first embodiment of a data input aware module of the big data processing system of FIG. 1
FIG. 3 is a second embodiment of a data input aware module of the big data processing system of FIG. 1
FIGS. 4-5 are flow diagrams illustrating particular methods for performing translation data processing tasks using the system of FIG. 1
Detailed Description
The invention is further described with reference to the following drawings and detailed description.
Referring to fig. 1, fig. 1 is an overall framework diagram of a big data processing system based on deep learning feedback according to an embodiment of the present invention.
In fig. 1, the big data processing system includes a data task receiving module, a data task analyzing module, a deep learning feedback module, a data task distributing module, an edge computing terminal, and a cloud data processing system.
The data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, and the data processing tasks have data input attributes;
the data input attribute comprises an input mode of the data processing task input by the input terminal, and the input mode comprises one of or the combination of keyboard input, touch input, mouse selection input and voice input.
In fig. 1, the data task receiving module further includes a data input sensing module, and the data input sensing module is configured to sense an input mode of the data processing task and the input environment parameter.
The data task analysis module is connected with the data task receiving module to receive the data processing task and analyzes and outputs at least one data input attribute of the data processing task;
specifically, the data task analysis module analyzes the data size and the data type of the data processing task.
In fig. 1, there are a plurality of edge calculation terminals, each of the edge calculation terminals includes a different edge calculation model, and at least one of the edge calculation models corresponds to the data type.
The deep learning feedback module is connected with the data task distribution module and the data task analysis module, and transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module.
In this embodiment, the deep learning feedback module includes a feedforward deep neural network and a feedback deep neural network.
In the feedforward neural network, each neuron is divided into different groups according to the sequence of received information, and each group can be regarded as a neural layer. The neurons in each layer receive the output of the neurons of the previous layer and output to the neurons of the next layer. Information in the whole network is propagated towards one direction, information is not propagated in the reverse direction (and error reverse propagation is not the same thing), and the information can be represented by a directed acyclic graph. The feedforward neural network comprises a fully-connected feedforward neural network and a convolutional neural network. The feedforward neural network can be regarded as a function, and complex mapping from an input space to an output space is realized through multiple compounding of simple nonlinear functions.
The neurons in the feedback neural network can receive the signals of other neurons and can receive the feedback signals of the neurons. Compared with a feedforward neural network, neurons in the feedback neural network have a memory function and have different states at different moments. Information propagation in the feedback neural network can be one-way or two-way, and therefore can be represented by a directed cyclic graph or an undirected graph.
Common feedback neural networks include recurrent neural networks, Hopfield networks, and Boltzmann machines.
To further enhance the memory capacity of the memory network, external memory units and read/write mechanisms may be mapped to store intermediate states of some networks, which are called memory-enhanced networks, such as neural turing machines.
2-3 are two embodiments of the data input sensing module of the big data processing system of FIG. 1. in the present invention, the data input attribute includes input environment parameters when the plurality of independent input terminals input data processing tasks, and the input environment parameters include hardware parameters, software parameters and input process parameters of the input terminals; the input process parameters comprise change parameters during the input of the data processing task by the input terminal, and the change parameters comprise a time starting point when the input terminal starts to input the data processing task, a time ending point when the input of the current data processing task is ended and/or change operation between the time starting point and the time ending point;
the change operation comprises one or a combination of modification, deletion, return, interface switching and pause operation.
The data input attribute comprises an input mode of the data processing task input by the input terminal, and the input mode comprises one of or the combination of keyboard input, touch input, mouse selection input and voice input.
FIG. 2 shows a data perception scenario for keyboard input. The change parameters in the keyboard input process sensed by the data input sensing module comprise a time starting point for starting inputting a data processing task by the input terminal, a time ending point for ending inputting the current data processing task and/or change operations between the time starting point and the time ending point, wherein the change operations comprise modification, deletion, withdrawal, interface switching and pause operations.
Referring next to fig. 3-5, a detailed embodiment of the present invention, which is specifically applied to process concurrent translation data processing tasks, is shown for a big data processing system based on deep learning feedback.
In fig. 3, the data processing task is a translation data processing task; and the data type is the language type of the translation data processing task.
The data task analysis module is connected to the data task receiving module to receive the data processing task and analyze and output at least one data input attribute of the data processing task, and further includes:
and the data task analysis module analyzes the data size and the data type of the data processing task.
In fig. 3, if the data task analysis module analyzes that the data type of the data processing task is a language, the data processing task is sent to at least one edge computing terminal through the data distribution module, where each of the at least one edge computing terminal includes an edge computing model corresponding to the data type; the edge calculation model is a translation engine with the ability of translating the small languages.
Referring next to fig. 4 to 5, the deep learning feedback module is connected to the data task distribution module and the data task analysis module, and transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module, which specifically includes:
if the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, dividing the data processing task into at least a first part and a second part;
the deep learning feedback module acquires feedback parameters of the edge computing terminals, and sends one of the first part or the second part to the cloud data processing system based on the feedback parameters.
If the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, the data processing task is divided into at least a first part and a second part, and the method specifically comprises the following steps:
the data processing task input through the keyboard is divided into a first part, and the data processing task input through voice is taken as a second part.
If the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, the data processing task is divided into at least a first part and a second part, and the method specifically comprises the following steps:
and taking the data processing task not containing the change parameters as a first part, taking the data processing task containing the change parameters as a second part, sending the first part to the edge computing terminal, and sending the second part to the cloud data processing system.
The deep learning feedback module obtains feedback parameters of the edge computing terminals, and sends one of the first part or the second part to the cloud data processing system based on the feedback parameters, and the deep learning feedback module specifically includes:
the feedback parameters comprise the current residual computing power of the edge computing terminals and the performance feedback value for completing the last data processing task.
In one example, the data set to be processed by the last data processing task contains i data sets to be translated,
wherein, the performance feedback value Hd fed back by each edge computing terminal is computed as follows:
wherein, Ti is the processing time of the edge computing terminal to the ith data set Xi to be translated last time; di is the time delay for obtaining the ith data set Xi to be translated; li is the size of the ith data set; maxX is the size of the largest dataset of all datasets { X1, X2, …, Xs } processed last time; i is 1, 2, … …, s.
If the Hd is less than a predetermined value and the current remaining computing power of the edge computing terminal is below a predetermined threshold, the edge computing terminal does not accept new data processing tasks for a predetermined period of time.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A big data processing system based on deep learning feedback comprises a data task receiving module, a data task analyzing module, a deep learning feedback module, a data task distributing module, an edge computing terminal and a cloud data processing system;
the method is characterized in that:
the data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, and the data processing tasks have data input attributes;
the data task analysis module is connected with the data task receiving module to receive the data processing task and analyzes and outputs at least one data input attribute of the data processing task;
the deep learning feedback module and the data task distribution module are connected with the data task analysis module, and the data processing task is transmitted to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module;
the data input attribute comprises input environment parameters when the plurality of mutually independent input terminals input data processing tasks, wherein the input environment parameters comprise hardware parameters, software parameters and input process parameters of the input terminals; the input process parameters comprise change parameters during the input of the data processing task by the input terminal, and the change parameters comprise a time starting point when the input terminal starts to input the data processing task, a time ending point when the input of the current data processing task is ended and/or change operation between the time starting point and the time ending point;
the change operation comprises one or a combination of modification, deletion, return, interface switching and pause operation.
2. The big data processing system based on deep learning feedback as claimed in claim 1, wherein:
the data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, the data processing tasks have data input attributes, and the data processing tasks specifically include:
the data input attribute comprises an input mode of the data processing task input by the input terminal, and the input mode comprises one of or the combination of keyboard input, touch input, mouse selection input and voice input.
3. The big data processing system based on deep learning feedback as claimed in claim 1, wherein:
the data task receiving module is used for receiving data processing tasks input from a plurality of mutually independent input terminals, the data processing tasks have data input attributes, and the data processing tasks specifically include:
the data task receiving module further comprises a data input sensing module, and the data input sensing module is used for sensing the input mode of the data processing task and the input environment parameters.
4. The big data processing system based on deep learning feedback as claimed in claim 2, wherein:
the data task analysis module is connected to the data task receiving module to receive the data processing task and analyze and output at least one data input attribute of the data processing task, and further includes:
and the data task analysis module analyzes the data size and the data type of the data processing task.
5. The big data processing system based on deep learning feedback as claimed in claim 4, wherein:
the edge calculation terminals are multiple, each edge calculation terminal comprises a different edge calculation model, and at least one edge calculation model corresponds to the data type.
6. The big data processing system based on deep learning feedback as claimed in claim 5, wherein:
the deep learning feedback module is connected with the data task distribution module and the data task analysis module, and transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module, and the deep learning feedback module specifically includes:
the data processing task is a translation data processing task;
the data type is the language type of the translation data processing task;
if the data type of the data processing task analyzed by the data task analysis module is in the Chinese language, the data processing task is sent to at least one edge computing terminal through the data distribution module, and each edge computing terminal comprises an edge computing model corresponding to the data type; the edge calculation model is a translation engine with the ability of translating the small languages.
7. The big data processing system based on deep learning feedback as claimed in claim 5, wherein:
the deep learning feedback module is connected with the data task distribution module and the data task analysis module, and transmits the data processing task to the edge computing terminal and/or the cloud data processing system based on the data input attribute output by the data task analysis module, and the deep learning feedback module specifically includes:
if the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, dividing the data processing task into at least a first part and a second part;
the deep learning feedback module acquires feedback parameters of the edge computing terminals, and sends one of the first part or the second part to the cloud data processing system based on the feedback parameters.
8. The big data processing system based on deep learning feedback of claim 7, wherein:
if the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, the data processing task is divided into at least a first part and a second part, and the method specifically comprises the following steps:
the data processing task input through the keyboard is divided into a first part, and the data processing task input through voice is taken as a second part.
9. The big data processing system based on deep learning feedback of claim 7, wherein:
if the data task analysis module analyzes that the data size of the data processing task exceeds a first preset value, the data processing task is divided into at least a first part and a second part, and the method specifically comprises the following steps:
and taking the data processing task not containing the change parameters as a first part, taking the data processing task containing the change parameters as a second part, sending the first part to the edge computing terminal, and sending the second part to the cloud data processing system.
10. The big data processing system based on deep learning feedback of claim 7, wherein:
the deep learning feedback module obtains feedback parameters of the edge computing terminals, and sends one of the first part or the second part to the cloud data processing system based on the feedback parameters, and the deep learning feedback module specifically includes:
the feedback parameters comprise the current residual computing power of the edge computing terminals and the performance feedback value for completing the last data processing task.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010637720.4A CN111753967A (en) | 2020-07-05 | 2020-07-05 | Big data processing system based on deep learning feedback and edge calculation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010637720.4A CN111753967A (en) | 2020-07-05 | 2020-07-05 | Big data processing system based on deep learning feedback and edge calculation |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111753967A true CN111753967A (en) | 2020-10-09 |
Family
ID=72679247
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010637720.4A Pending CN111753967A (en) | 2020-07-05 | 2020-07-05 | Big data processing system based on deep learning feedback and edge calculation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111753967A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107819840A (en) * | 2017-10-31 | 2018-03-20 | 北京邮电大学 | Distributed mobile edge calculations discharging method in the super-intensive network architecture |
CN110764833A (en) * | 2019-04-19 | 2020-02-07 | 中国联合网络通信集团有限公司 | Task unloading method, device and system based on edge calculation |
CN110855512A (en) * | 2018-08-21 | 2020-02-28 | 苏州迈科网络安全技术股份有限公司 | Ultra-large-scale DPI data processing system based on edge calculation |
CN111199740A (en) * | 2019-12-31 | 2020-05-26 | 重庆大学 | Unloading method for accelerating automatic voice recognition task based on edge calculation |
CN111245903A (en) * | 2019-12-31 | 2020-06-05 | 烽火通信科技股份有限公司 | Joint learning method and system based on edge calculation |
-
2020
- 2020-07-05 CN CN202010637720.4A patent/CN111753967A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107819840A (en) * | 2017-10-31 | 2018-03-20 | 北京邮电大学 | Distributed mobile edge calculations discharging method in the super-intensive network architecture |
CN110855512A (en) * | 2018-08-21 | 2020-02-28 | 苏州迈科网络安全技术股份有限公司 | Ultra-large-scale DPI data processing system based on edge calculation |
CN110764833A (en) * | 2019-04-19 | 2020-02-07 | 中国联合网络通信集团有限公司 | Task unloading method, device and system based on edge calculation |
CN111199740A (en) * | 2019-12-31 | 2020-05-26 | 重庆大学 | Unloading method for accelerating automatic voice recognition task based on edge calculation |
CN111245903A (en) * | 2019-12-31 | 2020-06-05 | 烽火通信科技股份有限公司 | Joint learning method and system based on edge calculation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhou et al. | Edge intelligence: Paving the last mile of artificial intelligence with edge computing | |
US20190095780A1 (en) | Method and apparatus for generating neural network structure, electronic device, and storage medium | |
US20200110761A1 (en) | System and method for providing an artificially-intelligent graph database | |
GB2604552A (en) | Fusing multimodal data using recurrent neural networks | |
US20240086720A1 (en) | Federated learning method, apparatus, and system | |
CN113196303A (en) | Inappropriate neural network input detection and processing | |
Kaur et al. | Dynamic resource allocation for big data streams based on data characteristics (5 V s) | |
CN104104702A (en) | Cloud computation architecture facing household internet of things | |
US10726123B1 (en) | Real-time detection and prevention of malicious activity | |
CN112732802A (en) | Enterprise data mining system and method based on big data | |
US11861210B2 (en) | Data processing method and apparatus for solid state disk | |
CN117155845A (en) | Internet of things data interaction method and system | |
CN113515519A (en) | Method, device and equipment for training graph structure estimation model and storage medium | |
CN111753967A (en) | Big data processing system based on deep learning feedback and edge calculation | |
Aiswarya et al. | Latency Reduction in Medical IoT Using Fuzzy Systems by Enabling Optimized Fog Computing | |
CN111800505A (en) | Big data acquisition and processing system under control of field remote terminal unit | |
CN110728118A (en) | Cross-data-platform data processing method, device, equipment and storage medium | |
US20220360796A1 (en) | Method and apparatus for recognizing action, device and medium | |
US11321954B1 (en) | Computer-vision techniques for time-series recognition and analysis | |
CN112560213B (en) | System modeling method and system based on model system engineering and hyper-network theory | |
Pardeshi | Analysis of data handling challenges in edge computing | |
CN114268625B (en) | Feature selection method, device, equipment and storage medium | |
CN114897147A (en) | Backbone network generation method, device, equipment and storage medium | |
WO2021077284A1 (en) | Neural network operating system and method | |
Kim et al. | Estimating unreliable objects and system reliability in P2P networks |
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 |