CN107453921A - Smart city system artificial intelligence evaluation method based on nonlinear neural network - Google Patents
Smart city system artificial intelligence evaluation method based on nonlinear neural network Download PDFInfo
- Publication number
- CN107453921A CN107453921A CN201710803578.4A CN201710803578A CN107453921A CN 107453921 A CN107453921 A CN 107453921A CN 201710803578 A CN201710803578 A CN 201710803578A CN 107453921 A CN107453921 A CN 107453921A
- Authority
- CN
- China
- Prior art keywords
- smart city
- city system
- nonlinear
- artificial intelligence
- evaluation method
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- 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
- G06N3/048—Activation functions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of smart city system artificial intelligence evaluation method based on nonlinear neural network, solution is the technical problem that intelligent quantization evaluation can not be carried out to smart city system, by using establishing initial data base；Initial data base is pre-processed, obtains forecast database；According to forecast database, according to the nonlinear auto-companding network for carrying outside input, the nonlinear auto-companding smart city system performance prediction model p (t) on time series is established；According to nonlinear auto-companding smart city system performance prediction model p (t), excitation function and training algorithm, carry out real time nonlinear autoregression smart city system performance prediction, and performance prediction result is inputted to the technical scheme of initial data base, the problem is preferably resolved, available in smart city system performance evaluation.
Description
Technical field
The present invention relates to smart city system field, and in particular to a kind of smart city system based on nonlinear neural network
System artificial intelligence evaluation method.
Background technology
The concept of smart city is developed by digital city, and China quotes this conceptual entity of smart city extensively
2010, in simple terms, smart city was formed by digital city+Internet of Things+cloud computing, and this is smart city one certainly
Recapitulative understanding, in fact smart city be mainly embodied in " wisdom ", it allows the more harmonious, nature of life, energy-conservation and height
Effect！At the same time, with regard to particularly important, the research and development of intelligent city management can not only promote for the development of intelligent city management
Enter the development in city, moreover, the function of intelligent city management displaying can more allow people to receive such a theory.It is but right
In the evaluation of smart city system, clear and definite method is now had no at present.
The present invention provides a kind of smart city system artificial intelligence evaluation method based on nonlinear neural network, Neng Gouzhun
Really efficient evaluation smart city system makes the reaction time of the links of feedback from gathered data to terminal, and is combed with this
Manage out the degree of accuracy and the performance of each link of smart city system.
The content of the invention
The technical problems to be solved by the invention are can not to carry out intelligence to smart city system present in prior art
The technical problem of quantitatively evaluating.A kind of new smart city system artificial intelligence evaluation side based on nonlinear neural network is provided
Method, it is easy to use, intelligent, accurate that the smart city system artificial intelligence evaluation method based on nonlinear neural network has
Spend the characteristics of high.
In order to solve the above technical problems, the technical scheme used is as follows：
A kind of smart city system artificial intelligence evaluation method based on nonlinear neural network, smart city system are artificial
Intelligent Evaluation method is based on smart city system, and the smart city system includes at least one data information acquisition end, with number
It is believed that the control device that is communicated by I/O of breath collection terminal, it is controlled by the terminal of control device, and remote server and for counting
According to the data communication equipment of communication, the smart city system also includes time set, and the artificial intelligence evaluation method includes：
Step 1：Establish initial data base；
Step 2：Initial data base is pre-processed, obtains forecast database；
Step 3：According to forecast database, according to the nonlinear auto-companding network for carrying outside input, establish on the time
The nonlinear auto-companding smart city system performance prediction model p (t) of sequence, the outside input are outside impression factor u
(t), the nonlinear auto-companding network with outside input include input layer, input it is stagnant when, hidden layer, output layer and output
When stagnant：
P (t)=f (p (t-n) ... p (t-1), u (t-n) ... u (t-1), W)
P (t)=f [(p (t), u (t), W)]；
Step 4：According to nonlinear auto-companding smart city system performance prediction model p (t), excitation function and training are calculated
Method, real time nonlinear autoregression smart city system performance prediction is carried out, and performance prediction result is inputted into initial data base；
Wherein, t represents the time, and p represents data traffic, and n is that positive integer characterizes delay exponent number, and W represents weight matrix.
The operation principle of the present invention：The performance data of smart city system is established into initial data base, by the time therein
Sequence vector and the data traffic corresponding to it are used as input, when time series vector includes input time, control device receives
Between, control device input time, terminal receive the time, terminal output time, the input feedback time.And then to it is following in a short time certain
The data traffic of carrying is predicted in period, so as to draw the reaction time of smart city system, and peak data flow
Most short reaction time corresponding to amount.
According to time space distribution, added for time-delay equation Recurrent neural network model representative how whole
End prediction.Time-delay equation Recurrent neural network has more preferable performance in the nonlinear data of processing sequential, and can overcome the disadvantages that branch
Hold the confinement problems of vector machine sample.Non linear autoregressive model full name is the nonlinear auto-companding mould with outside input
Type.Nonlinear auto-companding neutral net than total regression neutral net advantageously, nonlinear auto-companding neutral net can with it is complete
Recurrent neural networks are mutually changed, and it is defeated that nonlinear auto-companding introduces the outside closely related with time series in a model
Enter so that the precision of prediction is more accurate.
In such scheme, for optimization, further, the initial data base of establishing includes carrying out simulated experiment, and collection is real
Test data input remote server or database is generated according to history evaluation result.
Further, the delay exponent number n=30, the hiding layer number are 1, hidden layer neuron quantity l=20.
Further, the training algorithm is SCG algorithms.
Further, the excitation function is S type functions.
Further, the excitation function is T Elloit S functions.
Further, the excitation function is Tan-Sigmoid functions.
Further, the pretreatment includes carrying out data classification to database data, calculates the time difference and to data
Carry out data arrangement.
Further, the data classification includes being categorized as input time, control device reception time, control device input
Time, terminal receive time, terminal output time, input feedback time.
SVMs, i.e. SVM are the machine learning methods based on data.SVM is being solved for small sample and higher-dimension mould
There is outstanding behaviours in the problems such as formula identifies, and can be by the correlation machine Learning Studies such as application to Function Fitting.SVM energy
Enough solve the problems, such as local minimum and higher-dimension problem.Bayes classifier, footpath can be realized by the interior Product function for defining different
To a variety of learning algorithms such as Basis Function Method, multi-Layer Perceptron Neural Network.
Non linear autoregressive model than total regression neutral net advantageously, can be carried out mutual with total regression neutral net
Conversion.Unlike non linear autoregressive model NAR models, nonlinear auto-companding introduces close with time series in a model
The outside input that cut is closed, this make it that the precision of prediction is more accurate.
Its model is expressed as：
Y (t)=f (y (t-1), y (t-2) ..., y (t-ny),u(t-1),u(t-2),...,u(t-nu))
(n+1)th y (t) value is together decided on by preceding n y (t) value and preceding n u (t) value.Nonlinear auto-companding nerve net
Network mainly includes several parts when input layer, hidden layer, output layer and stagnant input and output, its basic structure such as Fig. 3.X (t) is god
Outside input through network, y (t) are the output of neutral net, 1:30 be delay exponent number n, and W represents the weights of link, and b represents threshold
Value.
Such as the network structure that Fig. 4 is nonlinear auto-companding neutral net, the model not only possesses good simulated performance,
Also possess prominent performance in estimated performance, can make accurate judgement to future trend, and with other nerve nets
Network compares, and it restrains faster, and regression nature is more preferable.
Beneficial effects of the present invention：
Effect one, valuation prediction models are built by using nonlinear neural network, can it is intelligent, easily evaluate intelligence
The performance parameter of intelligent city system；
Effect two, the degree of accuracy is high, reduces cost；
Effect three, by self-teaching amendment, improve the degree of accuracy.
Brief description of the drawings
The present invention is further described with reference to the accompanying drawings and examples.
Fig. 1, smart city city system schematic diagram.
Fig. 2, the smart city system artificial intelligence evaluation method schematic flow sheet of nonlinear neural network.
Fig. 3, NARX neural network structure schematic diagram.
The network structure of Fig. 4, NARX neutral net.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that specific embodiment described herein is not used to limit only to explain the present invention
The fixed present invention.
Embodiment 1
The present embodiment provides a kind of smart city system artificial intelligence evaluation method based on nonlinear neural network, wisdom
City system artificial intelligence evaluation method is based on smart city system, and the smart city system includes at least one data message
Collection terminal, the control device to be communicated with data information acquisition end by I/O is controlled by the terminal of control device, and remotely takes
Business device and the data communication equipment for data communication, the smart city system also include time set, the artificial intelligence
Evaluation method includes：
Step 1：Establish initial data base；
Step 2：Initial data base is pre-processed, obtains forecast database；
Step 3：According to forecast database, according to the nonlinear auto-companding network for carrying outside input, establish on the time
The nonlinear auto-companding smart city system performance prediction model p (t) of sequence, the outside input are outside impression factor u
(t), the nonlinear auto-companding network with outside input include input layer, input it is stagnant when, hidden layer, output layer and output
When stagnant：
P (t)=f (p (t-n) ... p (t-1), u (t-n) ... u (t-1), W)
P (t)=f [(p (t), u (t), W)]；
Step 4：According to nonlinear auto-companding smart city system performance prediction model p (t), excitation function and training are calculated
Method, real time nonlinear autoregression smart city system performance prediction is carried out, and performance prediction result is inputted into initial data base；
Wherein, t represents the time, and p represents data traffic, and n is that positive integer characterizes delay exponent number, and W represents weight matrix.
Specifically, the initial data base of establishing includes carrying out simulated experiment, collection experimental data input remote server
Or database is generated according to history evaluation result.
Specifically, the pretreatment includes carrying out data classification to database data, calculates the time difference and data are entered
Row data arrangement.
Specifically, when the data classification includes being categorized as input time, control device reception time, control device input
Between, terminal receive the time, terminal output time, the input feedback time.
NARX neutral nets include input layer, input it is stagnant when, hidden layer, output layer and output it is stagnant when, before model use really
When determining hidden layer neuron quantity, inputting stagnant and when output is stagnant, neural network structure such as Fig. 3.In Fig. 3, y (t) represents nerve net
The input or output of network, W represent the weights of link；B then represents threshold value.
Fig. 4 is the network structure of NARX neutral nets, and the model not only possesses good simulated performance, in estimated performance
On also possess prominent performance, accurate judgement can be made to future trend, and compared with other neutral nets, it is received
Hold back faster, regression nature is more preferable.
In neutral net, the effect of hidden layer comes from its groundwork in the feature in input data is extracted, so that
The data-handling capacity of network can be strengthened by hiding layer number in appropriate increase.The increase for hiding layer number is improving its prediction
Neutral net is complicated while performance, amount of calculation will be added, increases the training time of neutral net, or even causes anti-pass
Error.On this basis, error is reduced by increase hidden layer nodes.Unnecessary hidden layer will put neutral net in one
Kind extremity, increases it and is absorbed in the probability of local minimum point, so as to be impacted to the performance of neutral net；Hidden layer is got over
More, training burden of the neutral net in training is also more, and the training time is also longer.
Therefore, preferably, the delay exponent number n=30, the hiding layer number are 1.By adjusting this layer of neuron
Quantity optimizes to network.
The selection of training algorithm will generally consider many factors, such as sample size, network complexity.Its performance evaluation
Reference index mainly includes convergence of algorithm speed and memory consumption.In the present embodiment, specifically, the training algorithm is
SCG algorithms.SCG algorithms all show original in many-side, and particularly large-scale has in the network training of a large amount of weights.In function
In fitting, SCG algorithms possess the speed almost to be matched in excellence or beauty with LM algorithms, and similar to RP algorithm speeds in pattern-recognition, in god
Through in the more small training of network anticipation error, SCG algorithms are generally more more stable than RP algorithm, the memory consumption of SCG algorithms
It is relatively mild.
The selection of excitation function all causes directly to influence on the performance structure and computational complexity of network.It mainly divides
For global kernel function and local kernel function.The neutral net that the overall situation is approached, its excitation function also select overall situation function more suitable.
Selection is easy to calculate functional value and the function of first derivative values.
Specifically, the excitation function is S type functions.
S type functions carry out including Log-Sigmoid functions, Tan-Sigmoid functions and Elloit S functions.Log-
The derivative codomain of Sigmoid functions makes error correction signal can only be between 0 to 1/4.And Tan-Sigmoid and Elloit S letters
Number output areas are -1 to 1, and revise signal span is between 0 to 1.When the timing of step-length one of training algorithm, excitation function
First derivative may decide that the frequency of right value update.The first derivative shape of three kinds of functions is similar, the amplitude only around 0 point
It is variant, less random number is often taken between neutral net initial weight, therefore, the amplitude difference of derivative then can be to weights
Convergence rate impacts.When Tan-Sigmoid functions or Elloit S functions are as excitation function, network convergence is faster.
Although the illustrative embodiment of the present invention is described above, in order to the technology of the art
Personnel are it will be appreciated that the present invention, but the present invention is not limited only to the scope of embodiment, to the common skill of the art
For art personnel, as long as long as various change in the spirit and scope of the invention that appended claim limits and determines, one
The innovation and creation using present inventive concept are cut in the row of protection.
Claims (9)
- A kind of 1. smart city system artificial intelligence evaluation method based on nonlinear neural network, it is characterised in that：Wisdom city City's system artificial intelligence evaluation method is based on smart city system, and the smart city system is adopted including at least one data message Collect end, the control device to be communicated with data information acquisition end by I/O, be controlled by the terminal of control device, and remote service Device and the data communication equipment for data communication, the smart city system also include time set, and the artificial intelligence is commented Valency method includes：Step 1：Establish initial data base；Step 2：Initial data base is pre-processed, obtains forecast database；Step 3：According to forecast database, according to the nonlinear auto-companding network for carrying outside input, establish on time series Nonlinear auto-companding smart city system performance prediction model p (t), the outside input is outside impression factor u (t), institute State with outside input nonlinear auto-companding network include input layer, input it is stagnant when, hidden layer, output layer and output it is stagnant when：P (t)=f (p (t-n) ... p (t-1), u (t-n) ... u (t-1), W)P (t)=f [(p (t), u (t), W)]；Step 4：According to nonlinear auto-companding smart city system performance prediction model p (t), excitation function and training algorithm, Real time nonlinear autoregression smart city system performance prediction is carried out, and performance prediction result is inputted into initial data base；Wherein, t represents the time, and p represents data traffic, and n is positive integer, and n is delay exponent number, and W represents weight matrix.
- 2. the smart city system artificial intelligence evaluation method according to claim 1 based on nonlinear neural network, its It is characterised by：The initial data base of establishing includes carrying out simulated experiment, collection experimental data input remote server or root Database is generated according to history evaluation result.
- 3. the smart city system artificial intelligence evaluation method according to claim 1 based on nonlinear neural network, its It is characterised by：The delay exponent number is 30, and it is 1 to hide layer number, hidden layer neuron quantity l=20.
- 4. the smart city system artificial intelligence evaluation method according to claim 3 based on nonlinear neural network, its It is characterised by：The training algorithm is SCG algorithms.
- 5. the smart city system artificial intelligence evaluation method according to claim 3 based on nonlinear neural network, its It is characterised by：The excitation function is S type functions.
- 6. the smart city system artificial intelligence evaluation method according to claim 5 based on nonlinear neural network, its It is characterised by：The excitation function is T Elloit S functions.
- 7. the smart city system artificial intelligence evaluation method according to claim 5 based on nonlinear neural network, its It is characterised by：The excitation function is Tan-Sigmoid functions.
- 8. the smart city system artificial intelligence evaluation method according to claim 1 based on nonlinear neural network, its It is characterised by：The pretreatment includes carrying out data classification to database data, calculates the time difference and carries out data to data Arrangement.
- 9. the smart city system artificial intelligence evaluation method according to claim 8 based on nonlinear neural network, its It is characterised by：The data classification includes being categorized as input time, control device reception time, control device input time, end Terminate the time receiving between, terminal output time, the input feedback time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710803578.4A CN107453921A (en) | 2017-09-08 | 2017-09-08 | Smart city system artificial intelligence evaluation method based on nonlinear neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710803578.4A CN107453921A (en) | 2017-09-08 | 2017-09-08 | Smart city system artificial intelligence evaluation method based on nonlinear neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107453921A true CN107453921A (en) | 2017-12-08 |
Family
ID=60495936
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710803578.4A Pending CN107453921A (en) | 2017-09-08 | 2017-09-08 | Smart city system artificial intelligence evaluation method based on nonlinear neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107453921A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108964969A (en) * | 2018-05-07 | 2018-12-07 | 中国铁路总公司 | The high-speed railway signal system method for predicting of hybrid neural networks and AR model |
| CN110658720A (en) * | 2019-09-27 | 2020-01-07 | 电子科技大学 | Novel exhibition stand system based on neural network |
| CN112905945A (en) * | 2019-11-19 | 2021-06-04 | 中移物联网有限公司 | Charging method, charging device and readable storage medium |
| CN113064220A (en) * | 2021-06-03 | 2021-07-02 | 四川九通智路科技有限公司 | Visibility measuring system and measuring method based on nonlinear autoregressive neural network |
| CN113836785A (en) * | 2021-07-28 | 2021-12-24 | 南京尔顺科技发展有限公司 | Municipal regional intelligent water supply system and artificial intelligent control optimization method thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080052095A1 (en) * | 2006-08-22 | 2008-02-28 | Steven Neil | System and method for facilitating a low cost real estate transaction using a Multiple Listing Service (MLS) |
| CN104951836A (en) * | 2014-03-25 | 2015-09-30 | 上海市玻森数据科技有限公司 | Posting predication system based on nerual network technique |
| CN106302522A (en) * | 2016-09-20 | 2017-01-04 | 华侨大学 | A kind of network safety situations based on neutral net and big data analyze method and system |
| CN107067076A (en) * | 2017-05-27 | 2017-08-18 | 重庆大学 | A kind of passenger flow forecasting based on time lag NARX neutral nets |
| CN107085732A (en) * | 2017-05-12 | 2017-08-22 | 淮阴工学院 | Cowshed environment ammonia intelligent monitor system based on wireless sensor network |
| CN107103394A (en) * | 2017-05-27 | 2017-08-29 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic based on neutral net |
| CN107122526A (en) * | 2017-04-06 | 2017-09-01 | 大连大学 | Test section Mach number modeling method based on differential mode character subset Integrated Algorithm |
-
2017
- 2017-09-08 CN CN201710803578.4A patent/CN107453921A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080052095A1 (en) * | 2006-08-22 | 2008-02-28 | Steven Neil | System and method for facilitating a low cost real estate transaction using a Multiple Listing Service (MLS) |
| CN104951836A (en) * | 2014-03-25 | 2015-09-30 | 上海市玻森数据科技有限公司 | Posting predication system based on nerual network technique |
| CN106302522A (en) * | 2016-09-20 | 2017-01-04 | 华侨大学 | A kind of network safety situations based on neutral net and big data analyze method and system |
| CN107122526A (en) * | 2017-04-06 | 2017-09-01 | 大连大学 | Test section Mach number modeling method based on differential mode character subset Integrated Algorithm |
| CN107085732A (en) * | 2017-05-12 | 2017-08-22 | 淮阴工学院 | Cowshed environment ammonia intelligent monitor system based on wireless sensor network |
| CN107067076A (en) * | 2017-05-27 | 2017-08-18 | 重庆大学 | A kind of passenger flow forecasting based on time lag NARX neutral nets |
| CN107103394A (en) * | 2017-05-27 | 2017-08-29 | 重庆大学 | A kind of real-time passenger flow forecasting of track traffic based on neutral net |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108964969A (en) * | 2018-05-07 | 2018-12-07 | 中国铁路总公司 | The high-speed railway signal system method for predicting of hybrid neural networks and AR model |
| CN108964969B (en) * | 2018-05-07 | 2021-12-07 | 中国铁路总公司 | High-speed railway signal system flow prediction method based on hybrid neural network and AR model |
| CN110658720A (en) * | 2019-09-27 | 2020-01-07 | 电子科技大学 | Novel exhibition stand system based on neural network |
| CN112905945A (en) * | 2019-11-19 | 2021-06-04 | 中移物联网有限公司 | Charging method, charging device and readable storage medium |
| CN112905945B (en) * | 2019-11-19 | 2023-08-15 | 中移物联网有限公司 | Charging method, device and readable storage medium |
| CN113064220A (en) * | 2021-06-03 | 2021-07-02 | 四川九通智路科技有限公司 | Visibility measuring system and measuring method based on nonlinear autoregressive neural network |
| CN113836785A (en) * | 2021-07-28 | 2021-12-24 | 南京尔顺科技发展有限公司 | Municipal regional intelligent water supply system and artificial intelligent control optimization method thereof |
| CN113836785B (en) * | 2021-07-28 | 2024-02-13 | 南京尔顺科技发展有限公司 | Municipal area intelligent water supply system and artificial intelligent control optimization method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deng et al. | Multi-scale convolutional neural network with time-cognition for multi-step short-term load forecasting | |
| CN107453921A (en) | Smart city system artificial intelligence evaluation method based on nonlinear neural network | |
| Deng et al. | A novel collaborative optimization algorithm in solving complex optimization problems | |
| Ferreira et al. | An approach to reservoir computing design and training | |
| CN106570597A (en) | Content popularity prediction method based on depth learning under SDN architecture | |
| CN107665230A (en) | Training method and device for the users' behavior model of Intelligent housing | |
| CN108154430A (en) | A kind of credit scoring construction method based on machine learning and big data technology | |
| Cinar et al. | Development of future energy scenarios with intelligent algorithms: case of hydro in Turkey | |
| CN105447510B (en) | Fluctuating wind speed prediction technique based on artificial bee colony optimization LSSVM | |
| CN107067076A (en) | A kind of passenger flow forecasting based on time lag NARX neutral nets | |
| CN109445935A (en) | A kind of high-performance big data analysis system self-adaption configuration method under cloud computing environment | |
| CN108876001A (en) | A kind of Short-Term Load Forecasting Method based on twin support vector machines | |
| Zhang et al. | Prediction for network traffic of radial basis function neural network model based on improved particle swarm optimization algorithm | |
| CN110147389A (en) | Account number treating method and apparatus, storage medium and electronic device | |
| Araghinejad et al. | Application of data-driven models in drought forecasting | |
| Ramesh et al. | Hybrid artificial neural networks using customer churn prediction | |
| Ren et al. | Multivariate time series predictor with parameter optimization and feature selection based on modified binary salp swarm algorithm | |
| CN111382840B (en) | HTM design method based on cyclic learning unit and oriented to natural language processing | |
| Yao et al. | Umim: Utility-maximization incentive mechanism for mobile crowd sensing | |
| Wu et al. | Ensemble model of intelligent paradigms for stock market forecasting | |
| CN107301478A (en) | A kind of cable run short-term load forecasting method | |
| Wei et al. | Research on group behavior model based on neural network computing | |
| CN111105127B (en) | Modular product design evaluation method based on data driving | |
| Cheng et al. | Research on prediction method based on ARIMA-BP combination model | |
| CN103886169A (en) | Link prediction algorithm based on AdaBoost |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171208 |