CN108256681A - A kind of income level Forecasting Methodology, device, storage medium and system - Google Patents
A kind of income level Forecasting Methodology, device, storage medium and system Download PDFInfo
- Publication number
- CN108256681A CN108256681A CN201810036827.6A CN201810036827A CN108256681A CN 108256681 A CN108256681 A CN 108256681A CN 201810036827 A CN201810036827 A CN 201810036827A CN 108256681 A CN108256681 A CN 108256681A
- Authority
- CN
- China
- Prior art keywords
- income level
- neural network
- full connection
- connection neural
- forecasting methodology
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The present invention relates to user's income levels to predict field, a kind of income level Forecasting Methodology, device, storage medium and system;Its method includes collecting the user behavior data that user terminal is sent, and carry out numeralization to the behavioral data and handle the behavioural characteristic to be quantized;Each behavioural characteristic is converted into vector, and N layers of full Connection Neural Network are established according to the vector and all behavioural characteristics are combined to the high latitude space that is mapped to full Connection Neural Network;Learn the incidence relation between the behavioural characteristic of full Connection Neural Network, obtain the influence coefficient of the full Connection Neural Network in upper strata full Connection Neural Network to lower floor.The present invention provides a kind of income level Forecasting Methodology, device, storage medium and system and realizes, the behavioural characteristic that acquisition and processing to user behavior data are quantized, behavioural characteristic utilization rate is high, model accuracy rate is high, input cost is low and recommendation effect is good.
Description
Technical field
The present invention relates to user's income levels to predict field, and especially a kind of income level Forecasting Methodology, device, storage are situated between
Matter and system.
Background technology
In internet arena, we can be directed to user to be finely divided to user to realize the purpose of precision marketing
Earning power divides level and recommends different products & services contents.We can not obtain the true income water of each terminal user
It is flat, statistical modeling analysis is typically done according to the behavior of user, conventional model extremely relies on reason of the technical staff to behavioural characteristic
Solution, can not be by Feature Mapping to the space of more higher-dimension, it is difficult to play to greatest extent it is existing spy with value.As for most of normal
Scale there are poor performance, accuracy rate is low the problem of, cause marketing, recommendation effect it is poor, it is with high costs.
Invention content
For the defects in the prior art, the present invention provides a kind of income level Forecasting Methodology, device, storage medium and is
System, overcomes the problems such as user behavior characteristic use rate is low, model accuracy rate is low, input cost is high and recommendation effect is poor.
To achieve these goals, a kind of income level Forecasting Methodology provided by the invention, includes the following steps:
The user behavior data that user terminal is sent is collected, and carries out numeralization to the behavioral data and handle to be quantized
Behavioural characteristic;
Each behavioural characteristic is converted into vector, and N layers of full Connection Neural Network are established and by institute according to the vector
There is behavioural characteristic to be combined the high latitude space for being mapped to full Connection Neural Network;
Learn the incidence relation between the behavioural characteristic of full Connection Neural Network, obtain the full Connection Neural Network in upper strata to lower floor
The influence coefficient of full Connection Neural Network.
Further, the income level Forecasting Methodology further includes:The behavioral data is pre-processed.
Further, the income level Forecasting Methodology further includes:According to the numerical value change section of behavioural characteristic, normalizing is done
Change is handled.
Further, the income level Forecasting Methodology further includes:According to described N layers full Connection Neural Network addition M
Residual block.
Further, described N layers full Connection Neural Network is 150 layers of full Connection Neural Network, and the M residual block is 30
A residual block.
Further, the income level Forecasting Methodology further includes:Set the node in the full Connection Neural Network of each layer
Retain probability, the full Connection Neural Network includes multiple nodes.
A kind of income level prediction meanss, the income level prediction meanss include memory, processor and are stored in
In the memory and the program that can run on the processor, the memory and processor are electrically connected, wherein, the storage
Device is for storing computer program, and the computer program includes program instruction, and the processor is configured for described in calling
Program instruction, perform as mentioned income level Forecasting Methodology the step of.
A kind of computer readable storage medium, the computer-readable recording medium storage have computer program, the meter
Calculation machine program includes program instruction, and described program instruction makes the processor execution take in as mentioned when being executed by a processor
The step of horizontal forecast method.
A kind of income level forecasting system, the income level forecasting system include income level prediction meanss and user
End;The horizontal forecast device is connected by wireless or cable network with user terminal;The horizontal forecast device is performed such as institute
The step of income level Forecasting Methodology stated.
The beneficial effects of the invention are as follows:The present invention provides a kind of income level Forecasting Methodology, device, storage medium and system
It realizes, the behavioural characteristic that the acquisition and processing to user behavior data are quantized, behavioural characteristic utilization rate is high, model is accurate
True rate is high, input cost is low and recommendation effect is good.
Description of the drawings
Fig. 1 is the flow chart of income level Forecasting Methodology first embodiment of the present invention;
Fig. 2 is the flow chart of income level Forecasting Methodology second embodiment of the present invention;
Fig. 3 is the block diagram of income level forecasting system embodiment of the present invention.
Specific embodiment
Specific embodiments of the present invention are described more fully below, it should be noted that the embodiments described herein is served only for illustrating
Illustrate, be not intended to restrict the invention.In the following description, in order to provide a thorough understanding of the present invention, a large amount of spies are elaborated
Determine details.It will be apparent, however, to one skilled in the art that:This hair need not be carried out using these specific details
It is bright.In other instances, in order to avoid obscuring the present invention, well known circuit, software or method are not specifically described.
Throughout the specification, meaning is referred to " one embodiment ", " embodiment ", " example " or " example "
It:It is comprised at least one embodiment of the present invention with reference to a particular feature, structure, or characteristic that the embodiment or example describe.
Therefore, the phrase " in one embodiment ", " in embodiment ", " example " occurred in each place of the whole instruction
Or " example " is not necessarily all referring to the same embodiment or example.Furthermore, it is possible to any appropriate combination and or sub-portfolio will be specific
Feature, structure or characteristic combine in one or more embodiments or example.In addition, those of ordinary skill in the art should manage
Solution, diagram is provided to the purpose of explanation provided herein, and diagram is not necessarily drawn to scale.
Income level Forecasting Methodology first embodiment:
As shown in Figure 1, income level Forecasting Methodology, includes the following steps:
S001 collects the user behavior data that user terminal is sent, and carries out numeralization to the behavioral data and handle to obtain
The behavioural characteristic of numeralization;
Each behavioural characteristic is converted into vector, and establish N layers of full Connection Neural Network according to the vector by S002, with
And all behavioural characteristics are combined to the high latitude space for being mapped to full Connection Neural Network;
S003 learns the incidence relation between the behavioural characteristic of full Connection Neural Network, obtains the full Connection Neural Network in upper strata
The influence coefficient of full Connection Neural Network to lower floor.
Income level Forecasting Methodology second embodiment:
As shown in Fig. 2, income level Forecasting Methodology, includes the following steps:
S101 collects the user behavior data that user terminal is sent.
Specifically, user terminal is uploaded to income level prediction meanss after user behavior data is collected, and it is described
Horizontal forecast device is the behavior hair according to each different user and each user during user behavior data is acquired
What raw priority time sequencing was acquired.
S102 pre-processes the behavioral data.
Specifically, the behavioral data is pre-processed after the user behavior data is collected, the pre- place
The mode of reason includes but not limited to kick out of exceptional value and null value;By the way that the user behavior data is pre-processed and can be reduced
Later stage operation improves the efficiency of income level prediction.
S103 carries out numeralization to the behavioral data and handles the behavioural characteristic to be quantized.
It is specifically, special by the behavior that user behavior can be transformed to numeralization by behavioral data numeralization processing
Sign is conducive to later stage calculating.
Each behavioural characteristic is converted into vector, and establish N layers of full Connection Neural Network according to the vector by S104, with
And all behavioural characteristics are combined to the high latitude space for being mapped to full Connection Neural Network.
Specifically, it by the i.e. N of artificial settings is positive integer that full Connection Neural Network, which is, and according to reality in the present embodiment
Operation time and the factors such as efficiency N is set as 150, that is, 150 layers of full Connection Neural Network are established, when full Connection Neural Network
All behavioural characteristics are carried out again after building up, the combination of the behavioural characteristic includes but the existing plus-minus between behavioural characteristic does not multiply
Division operation or steering, can in high latitude space by being placed in the high latitude space of combinatorial mapping to full Connection Neural Network
It is more convenient to find out best combination between each feature.
S105 according to the numerical value change section of behavioural characteristic, does normalized.
Specifically, by the numerical value change section according to the behavioural characteristic to the behavioural characteristic of all collected users into
Capable normalized in batches, can either improve computational efficiency, and can avoid simultaneously to the behavior of all collected users spy
Behavior characteristic information loses the defects of more after sign carries out unified normalized.
S106, the node set in the full Connection Neural Network of each layer retain probability.
Specifically, the full Connection Neural Network includes multiple nodes, the number of the node be by artificial settings,
It can be increased and decreased by constantly testing;The node can be used for representing a behavioural characteristic, by the way that the node is set to protect
Probability is stayed to lose some parameters at random, the weight of node can be reduced, improves generalization ability.
S107 adds M residual block according to N layers of full Connection Neural Network.
Specifically, residual block is by artificial settings, i.e. M is positive integer and M≤N.In practical application process, examine
Under the premise of considering the condition elements such as computational efficiency, by by every 5 layers of 150 layers of full Connection Neural Network full Connection Neural Network into
Row one residual block of addition, when the gradient of the influence coefficient of the full Connection Neural Network in upper strata full Connection Neural Network to lower floor is pair
The influence coefficient carries out derivation, when the value of derivation level off to substantially zero when or gradient disappear when, directly skip full connection god
Remaining middle layer through network obtains influencing the finally determining value of coefficient, and having, which reduces calculation amount, improves the excellent of computational efficiency
Point.
S108 learns the incidence relation between the behavioural characteristic of full Connection Neural Network, obtains the full Connection Neural Network in upper strata
The influence coefficient of full Connection Neural Network to lower floor.
Specifically, the incidence relation between the behavioural characteristic is determined according to score is judged, the judgement score
Calculation be:Square of predicted value and actual value difference;Need to illustrate be actual value is that existing fixation has
Value, predicted value and actual value difference square increasingly level off to 0, then predicted value and actual value are closer;The influence coefficient
For the coefficient of formula, trained process is exactly to learn the process of the influence coefficient.
S109 exports final result using activation primitive, is mapped in class categories.
Specifically, by the activation primitive can square must specifically be presented predicted value and actual value difference,
And pass through one judgment threshold of setting and be compared with the final result, if final result is more than the judgment threshold, user
Income level it is high, if final result is less than the judgment threshold, the income level of user is low.
The first embodiment of income level prediction meanss:
Income level prediction meanss, the income level prediction meanss include memory, processor and are stored in described
In memory and the program that can run on the processor, the memory and processor are electrically connected, wherein, the memory is used
In storage computer program, the computer program includes program instruction, and the processor is configured for calling described program
The step of instructing, performing income level Forecasting Methodology second embodiment as mentioned, specifically includes:
S201 collects the user behavior data that user terminal is sent.
S202 pre-processes the behavioral data.
S203 carries out numeralization to the behavioral data and handles the behavioural characteristic to be quantized.
Each behavioural characteristic is converted into vector, and establish N layers of full Connection Neural Network according to the vector by S204, with
And all behavioural characteristics are combined to the high latitude space for being mapped to full Connection Neural Network.
S205 according to the numerical value change section of behavioural characteristic, does normalized.
S206, the node set in the full Connection Neural Network of each layer retain probability.
S207 adds M residual block according to N layers of full Connection Neural Network.
S208 learns the incidence relation between the behavioural characteristic of full Connection Neural Network, obtains the full Connection Neural Network in upper strata
The influence coefficient of full Connection Neural Network to lower floor.
S209 exports final result using activation primitive, is mapped in class categories.
Specifically, memory may include high-speed random access memory (RAM:Random Access Memory),
Non-labile memory (non-volatile memory), for example, at least a magnetic disk storage may be further included.By extremely
A few communication interface (can be wired or wireless) realizes the communication between the system network element and at least one other network element
Connection can use internet, wide area network, local network, Metropolitan Area Network (MAN) etc..
Specifically, processor may be a kind of IC chip, there is the processing capacity of signal.During realization,
Each step of the above method can be completed by the integrated logic circuit of the hardware in processor or the instruction of software form.On
The processor stated can be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), net
Network processor (Network Processor, abbreviation NP) etc.;It can also be digital signal processor (DSP), application-specific integrated circuit
(ASIC), ready-made programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components.
It needs to illustrate, the processor is performing income level Forecasting Methodology second embodiment as mentioned
During step, specific mode and advantageous effect in income level Forecasting Methodology second embodiment it is recorded it is substantially similar
This is not repeated.
The first embodiment of computer readable storage medium:
Computer readable storage medium, the computer-readable recording medium storage have computer program, the computer
Program includes program instruction, and described program instruction makes the processor perform income level as mentioned when being executed by a processor
It the step of Forecasting Methodology second embodiment, specifically includes:
S301 collects the user behavior data that user terminal is sent.
S302 pre-processes the behavioral data.
S303 carries out numeralization to the behavioral data and handles the behavioural characteristic to be quantized.
Each behavioural characteristic is converted into vector, and establish N layers of full Connection Neural Network according to the vector by S304, with
And all behavioural characteristics are combined to the high latitude space for being mapped to full Connection Neural Network.
S305 according to the numerical value change section of behavioural characteristic, does normalized.
S306, the node set in the full Connection Neural Network of each layer retain probability.
S307 adds M residual block according to N layers of full Connection Neural Network.
S308 learns the incidence relation between the behavioural characteristic of full Connection Neural Network, obtains the full Connection Neural Network in upper strata
The influence coefficient of full Connection Neural Network to lower floor.
S309 exports final result using activation primitive, is mapped in class categories.
Specifically, computer readable storage medium may include caching (Cache), high-speed random access memory (RAM), example
Such as common double data rate Synchronous Dynamic Random Access Memory (DDR SDRAM), and may also include nonvolatile memory
(NVRAM), such as one or more read-only memory (ROM), disk storage equipment, flash memory (Flash) memory devices or its
His non-volatile solid state memory equipment such as CD (CD-ROM, DVD-ROM), floppy disk or data tape etc..
It needs to illustrate, the processor is performing income level Forecasting Methodology second embodiment as mentioned
During step, specific mode and advantageous effect in income level Forecasting Methodology second embodiment it is recorded it is substantially similar
This is not repeated.
The first embodiment of income level forecasting system:
As shown in figure 3, income level forecasting system, the income level forecasting system include income level prediction meanss and
User terminal;The horizontal forecast device is connected by wireless or cable network with user terminal;The horizontal forecast device performs
Such as the step of income level Forecasting Methodology second embodiment, specifically include:
S401 collects the user behavior data that user terminal is sent.
S402 pre-processes the behavioral data.
S403 carries out numeralization to the behavioral data and handles the behavioural characteristic to be quantized.
Each behavioural characteristic is converted into vector, and establish N layers of full Connection Neural Network according to the vector by S404, with
And all behavioural characteristics are combined to the high latitude space for being mapped to full Connection Neural Network.
S405 according to the numerical value change section of behavioural characteristic, does normalized.
S406, the node set in the full Connection Neural Network of each layer retain probability.
S407 adds M residual block according to N layers of full Connection Neural Network.
S408 learns the incidence relation between the behavioural characteristic of full Connection Neural Network, obtains the full Connection Neural Network in upper strata
The influence coefficient of full Connection Neural Network to lower floor.
S409 exports final result using activation primitive, is mapped in class categories.
It needs to illustrate, the income level prediction meanss are performing income level Forecasting Methodology the as mentioned
During the step of two embodiments, recorded in specific mode and advantageous effect and income level Forecasting Methodology second embodiment
It is substantially similar to be not repeated herein.
Finally it should be noted that:The above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations;To the greatest extent
Pipe is described in detail the present invention with reference to foregoing embodiments, it will be understood by those of ordinary skill in the art that:Its according to
Can so modify to the technical solution recorded in foregoing embodiments either to which part or all technical features into
Row equivalent replacement;And these modifications or replacement, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme should all cover in the claim of the present invention and the range of specification.
Claims (9)
1. a kind of income level Forecasting Methodology, which is characterized in that include the following steps:
The user behavior data that user terminal is sent is collected, and numeralization is carried out to the behavioral data and handles the row to be quantized
It is characterized;
Each behavioural characteristic is converted into vector, and N layers of full Connection Neural Network are established and by all rows according to the vector
It is characterized and is combined the high latitude space for being mapped to full Connection Neural Network;
Learn the incidence relation between the behavioural characteristic of full Connection Neural Network, obtain the full Connection Neural Network in upper strata and lower floor is connected entirely
Connect the influence coefficient of neural network.
2. income level Forecasting Methodology according to claim 1, which is characterized in that the income level Forecasting Methodology is also wrapped
It includes:The behavioral data is pre-processed.
3. income level Forecasting Methodology according to claim 1, which is characterized in that the income level Forecasting Methodology is also wrapped
It includes:According to the numerical value change section of behavioural characteristic, normalized is done.
4. income level Forecasting Methodology according to claim 1, which is characterized in that the income level Forecasting Methodology is also wrapped
It includes:M residual block is added according to described N layers full Connection Neural Network.
5. income level Forecasting Methodology according to claim 4, it is characterised in that:Described N layers full Connection Neural Network be
150 layers of full Connection Neural Network, the M residual block are 30 residual blocks.
6. income level Forecasting Methodology according to claim 1, it is characterised in that:The income level Forecasting Methodology is also wrapped
It includes:It sets the node in the full Connection Neural Network of each layer and retains probability, the full Connection Neural Network includes multiple nodes.
7. a kind of income level prediction meanss, it is characterised in that:The income level prediction meanss include memory, processor with
And the program that can be run in the memory and on the processor is stored in, the memory and processor are electrically connected, wherein,
The memory is for storing computer program, and the computer program includes program instruction, and the processor is configured for
Described program instruction is called, is performed such as the step of claim 1-6 any one of them income level Forecasting Methodologies.
8. a kind of computer readable storage medium, it is characterised in that:The computer-readable recording medium storage has computer journey
Sequence, the computer program include program instruction, and described program instruction makes the processor perform such as when being executed by a processor
The step of income level Forecasting Methodology described in claim 1-6 any one.
9. a kind of income level forecasting system, it is characterised in that:The income level forecasting system includes income level prediction dress
It puts and user terminal;The horizontal forecast device is connected by wireless or cable network with user terminal;The horizontal forecast device
It performs such as the step of claim 1-6 any one of them income level Forecasting Methodologies.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810036827.6A CN108256681A (en) | 2018-01-15 | 2018-01-15 | A kind of income level Forecasting Methodology, device, storage medium and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810036827.6A CN108256681A (en) | 2018-01-15 | 2018-01-15 | A kind of income level Forecasting Methodology, device, storage medium and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108256681A true CN108256681A (en) | 2018-07-06 |
Family
ID=62726728
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810036827.6A Pending CN108256681A (en) | 2018-01-15 | 2018-01-15 | A kind of income level Forecasting Methodology, device, storage medium and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108256681A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110197301A (en) * | 2019-05-27 | 2019-09-03 | 深圳乐信软件技术有限公司 | A kind of prediction technique of disposable income, device, server and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1795463A (en) * | 2003-05-22 | 2006-06-28 | 珀欣投资有限责任公司 | Customer revenue prediction method and system |
CN106104406A (en) * | 2014-03-06 | 2016-11-09 | 前进公司 | Neutral net and the method for neural metwork training |
CN107194715A (en) * | 2017-04-07 | 2017-09-22 | 广东精点数据科技股份有限公司 | The construction method of social action data model |
-
2018
- 2018-01-15 CN CN201810036827.6A patent/CN108256681A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1795463A (en) * | 2003-05-22 | 2006-06-28 | 珀欣投资有限责任公司 | Customer revenue prediction method and system |
CN106104406A (en) * | 2014-03-06 | 2016-11-09 | 前进公司 | Neutral net and the method for neural metwork training |
CN107194715A (en) * | 2017-04-07 | 2017-09-22 | 广东精点数据科技股份有限公司 | The construction method of social action data model |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110197301A (en) * | 2019-05-27 | 2019-09-03 | 深圳乐信软件技术有限公司 | A kind of prediction technique of disposable income, device, server and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102611938B1 (en) | Generate integrated circuit floorplans using neural networks | |
CN109710404B (en) | Task scheduling method in distributed system | |
CN107153874A (en) | Water quality prediction method and system | |
CN107169563A (en) | Processing system and method applied to two-value weight convolutional network | |
WO2021208079A1 (en) | Method and apparatus for obtaining power battery life data, computer device, and medium | |
McGarvey et al. | On the use of climate covariates in aquatic species distribution models: are we at risk of throwing out the baby with the bath water? | |
CN108120521A (en) | Coiling hot point of transformer temperature predicting method and terminal device | |
WO2017161646A1 (en) | Method for dynamically selecting optimal model by three-layer association for large data volume prediction | |
CN110348665A (en) | A kind of low-voltage platform area electric power system data quality evaluating method and device | |
Zhao et al. | Improved dynamic programming for reservoir flood control operation | |
CN107133190A (en) | The training method and training system of a kind of machine learning system | |
CN107346267A (en) | A kind of cpu performance optimization method and device based on NUMA architecture | |
CN105791447A (en) | Method and device for dispatching cloud resource orienting to video service | |
CN110287250A (en) | User gradation quantization method and device | |
CN105790866B (en) | Base station rankings method and device | |
CN112288172A (en) | Prediction method and device for line loss rate of transformer area | |
CN110598336B (en) | Water consumption prediction method and device for water heater, water heater and electronic equipment | |
CN106682414A (en) | Method and device for establishing timing sequence prediction model | |
CN108256681A (en) | A kind of income level Forecasting Methodology, device, storage medium and system | |
CN113837383B (en) | Model training method and device, electronic equipment and storage medium | |
CN107943756A (en) | A kind of computational methods and Related product | |
CN115018119A (en) | Power load prediction method and system | |
CN116541128A (en) | Load adjusting method, device, computing equipment and storage medium | |
CN101666853B (en) | Test scheduling system and test scheduling method | |
CN110415014A (en) | A kind of branded network health degree determines method, apparatus, terminal and storage medium |
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: 20180706 |