CN110110927A - A kind of network courses order demand prediction technique based on grey neural network - Google Patents
A kind of network courses order demand prediction technique based on grey neural network Download PDFInfo
- Publication number
- CN110110927A CN110110927A CN201910373767.1A CN201910373767A CN110110927A CN 110110927 A CN110110927 A CN 110110927A CN 201910373767 A CN201910373767 A CN 201910373767A CN 110110927 A CN110110927 A CN 110110927A
- Authority
- CN
- China
- Prior art keywords
- network
- order
- neural network
- demand
- courses
- 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
- 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
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
- G06Q50/2057—Career enhancement or continuing education service
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Technology (AREA)
- Finance (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Accounting & Taxation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The network courses order demand prediction technique based on grey neural network that the present invention relates to a kind of, by being carried out according to network courses order data over the years and order demand dimensionality exponent using data modeling, the network order over the years is carried out from four demand dimensions using data modeling the present invention respectively according to time series;It is fitted to obtain network courses order demand index over the years according to application data modeling and grey neural network modeling;The weighting parameter and threshold parameter of grey neural network are initialized, learning rate is set, generates network architecture parameters by presetting random idea;Grey neural network training is carried out by cyclic iterative method;Order demand predictive information is obtained by training;Order is predicted by trained model, obtains difference, is tested into new data.Grey Neural Network Model of the present invention can be accurate, convenient, easily predicts Small Sample Database.
Description
Technical field
The present invention relates to grey neural network field more particularly to a kind of network courses orders based on grey neural network
Needing forecasting method.
Background technique
With China's expanding economy, a variety of scientific and technological industry online training such as IT, IC are gradually risen.Network is trained online
For instruction, many because being known as of market demand are influenced, for example improve in graduation recruitment season, information-based industry tide, on-job technical ability,
And training course quality etc..Online scientific and technical training business includes many Classic Courses, for example is suitable for university research organization
SCI writing training, the SPSS software training suitable for data statistics subject and the reality suitable for graduating student's employment interview
It fights training course etc., but for the course demand of market, lacks the network courses that analyze that a kind of method is capable of system
Relation between supply and demand.
Network courses market supply dynamic and market demand dynamic are closely bound up, and the order demand of product is vulnerable to annual fixed
Graduation recruit seasonal effect, be mainly manifested in that the autumn recruits, spring university enrollment and enterprise's position are transferred and promoted etc., therefore network courses are in city
Demand side has many uncertainties, it would be desirable to seek a kind of method to carry out the market demand of prediction network courses, with
Just Speeding up development corresponding course product and reasonably optimizing Course Training system, this, which is also that those skilled in the art are urgently to be resolved, asks
Topic.
Summary of the invention
The present invention solves the deficiency of gray model and Application of Neural Network, some optimization algorithms such as genetic algorithm also by with
In the modeling and optimization of grey neural network.Solve the problems, such as how to predict the network courses market demand.
A kind of network courses order demand prediction technique based on grey neural network, which is characterized in that including walking as follows
It is rapid:
S1: carrying out according to network courses order data over the years and order demand dimensionality exponent using data modeling, described to go through
Year network order is carried out from four demand dimensions using data modeling respectively according to time series;
S2: it is fitted to obtain network courses order demand over the years according to application data modeling and grey neural network modeling
Index;
S3: initializing the weighting parameter and threshold parameter of grey neural network, sets learning rate, by default
Random idea generates network architecture parameters;
S4: grey neural network training is carried out by cyclic iterative method;
S5: order demand predictive information is obtained by training;
S6: predicting order by trained model, obtain difference, tests into new data.
Preferred: the demand dimension is four vocational need, the market characteristics, selling price fluctuation and graduation recruitment season dimensions
Degree.
Preferred: the application data modeling mainly passes through time-sequencing method and is based on time demand rule and demand dimension,
To the day order data of order, all order datas, year order data, laterally ratio is modeled order.The grey neural network
Structure is 1-1-5-1 type, and network first tier M1 has 1 node, time series T, and network second layer M2 has a node, network
Third layer M3 has 5 nodes, and 2-5 input 4 vocational need dimension vocational needs, the market characteristics, sale price lattice waves respectively
Dynamic and graduation recruitment season, normalization data.
Preferred: the loop iteration valve optimizes order error amount, is ordered by making network by iteration undated parameter
Single network model.
Preferred: the neural metwork training is by obtaining the output valve of network middle layer, the threshold parameter of intermediate node
Value, and error correction is done according to predicted value of the label true value to the training network, thus further to the ginseng of network node
Number weight parameter is updated.
Compared with prior art, the invention has the following beneficial technical effects for Detailed description of the invention:
The present invention is based on the prediction of the network courses order demand of grey neural network, grey neural network is a kind of fusion ash
The innovation intelligence computation method of color system model and neural network, makes full use of the similitude of Grey System Model and neural network
And complementarity, solve the deficiency of gray model and Application of Neural Network respectively.Some optimization algorithms such as genetic algorithm also by with
In the modeling and optimization of grey neural network.Grey Neural Network Model of the present invention can be accurate, convenient, easily predicts sample
Notebook data.
Detailed description of the invention
Fig. 1 is algorithm flow chart of the invention;
Specific embodiment
The invention will be described in further detail with reference to the accompanying drawing, and the explanation of the invention is not limited.
As shown, the invention discloses a kind of network courses order demand prediction technique based on grey neural network,
Technical solution of the present invention includes the following steps:
S1: carrying out according to network courses order data over the years and order demand dimensionality exponent using data modeling, described to go through
Year network order is carried out from four demand dimensions using data modeling respectively according to time series;
S2: it is fitted to obtain network courses order demand over the years according to application data modeling and grey neural network modeling
Index;
S3: initializing the weighting parameter and threshold parameter of grey neural network, sets learning rate, by default
Random idea generates network architecture parameters;
S4: grey neural network training is carried out by cyclic iterative method;
S5: order demand predictive information is obtained by training;
S6: predicting order by trained model, obtain difference, tests into new data.
Demand dimension is vocational need, the market characteristics, selling price fluctuation and graduation recruitment season four dimensions.
Mainly pass through time-sequencing method using data modeling and be based on time demand rule and demand dimension, the day of order is ordered
Forms data, all order datas, year order data, order laterally ratio modeled.
Loop iteration valve optimizes order error amount, obtains order network model by making network by iteration undated parameter.
Neural metwork training is by obtaining the output valve of network middle layer, the thresholding parameter value of intermediate node, and according to
Label true value does error correction to the predicted value of the training network, thus further to the parameters weighting parameter of network node into
Row updates.
According to actual implementation process: data acquisition and system modelling, the data obtained from backstage carry out initial data
The original order data of input data of the accumulation process as network, online course training are stored under mat formatted file, this article
Matrix N is 48 row, 5 column matrix in part, and first is classified as certain course quantity on order, 2-5 column be respectively vocational need, the market characteristics,
Selling price fluctuation, 4 influence factor data in graduation recruitment season.
Netinit, initializes the weighting parameter and thresholding parameter value of the structure grey neural network first, and by net
The learning rate of network is set as 0.001, and network architecture parameters are generated by preset random seed point.
Grey neural network training, setting loop iteration number are 200 times, calculate separately every layer of output of network middle layer
Value, is arranged the thresholding parameter value of intermediate node, and do error correction according to predicted value of the label true value to the training network,
To be further updated to the parameters weighting parameter of network node.
Grey neural network prediction, after network training reaches preset value the number of iterations, using trained model to elder generation
Preceding test data carries out online network courses order demand prediction, and setting cycle-index is 8 times, and final test mean error is
5.9%, it is constantly reduced and with test result by model error when training it is found that the invention can be to the order of online training business
Prediction reaches extraordinary effect, and cycle of training is few, saves time cost, has good generalization ability to data, adapts to
Small sample prediction, catches the market demand to have directive function business event in time.
Example given above is to realize the present invention preferably example, and the present invention is not limited to the above embodiments.This field
Technical staff's technical solution according to the present invention technical characteristic any nonessential addition, the replacement made, belong to this
The protection scope of invention.
Claims (5)
1. a kind of network courses order demand prediction technique based on grey neural network, which comprises the steps of:
S1: it is carried out according to network courses order data over the years and order demand dimensionality exponent using data modeling, the net over the years
Network order is carried out from four demand dimensions using data modeling respectively according to time series;
S2: it is fitted to obtain network courses order demand over the years and refer to according to application data modeling and grey neural network modeling
Number;
S3: initializing the weighting parameter and threshold parameter of grey neural network, sets learning rate, by default random
Idea generates network architecture parameters;
S4: grey neural network training is carried out by cyclic iterative method;
S5: order demand predictive information is obtained by training;
S6: predicting order by trained model, obtain difference, tests into new data.
2. a kind of network courses order demand prediction technique based on grey neural network according to claim 1, special
Sign is: the demand dimension is vocational need, the market characteristics, selling price fluctuation and graduation recruitment season four dimensions.
3. a kind of network courses order demand prediction technique based on grey neural network according to claim 1, special
Sign is: the application data modeling mainly passes through time-sequencing method and is based on time demand rule and demand dimension, to order
Day order data, all order datas, year order data, laterally ratio is modeled order.
4. a kind of network courses order demand prediction technique based on grey neural network according to claim 1, special
Sign is: the loop iteration valve optimizes order error amount, obtains order network mould by making network by iteration undated parameter
Type.
5. a kind of network courses order demand prediction technique based on grey neural network according to claim 4, special
Sign is: the neural metwork training is by obtaining the output valve of network middle layer, the thresholding parameter value of intermediate node, and root
Error correction is done according to predicted value of the label true value to the training network, thus further to the parameters weighting parameter of network node
It is updated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910373767.1A CN110110927A (en) | 2019-05-07 | 2019-05-07 | A kind of network courses order demand prediction technique based on grey neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910373767.1A CN110110927A (en) | 2019-05-07 | 2019-05-07 | A kind of network courses order demand prediction technique based on grey neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110110927A true CN110110927A (en) | 2019-08-09 |
Family
ID=67488442
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910373767.1A Pending CN110110927A (en) | 2019-05-07 | 2019-05-07 | A kind of network courses order demand prediction technique based on grey neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110110927A (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107808212A (en) * | 2017-10-09 | 2018-03-16 | 南京邮电大学 | Solar energy collecting power forecasting method based on grey neural network |
CN108230043A (en) * | 2018-01-31 | 2018-06-29 | 安庆师范大学 | A kind of product area pricing method based on Grey Neural Network Model |
-
2019
- 2019-05-07 CN CN201910373767.1A patent/CN110110927A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107808212A (en) * | 2017-10-09 | 2018-03-16 | 南京邮电大学 | Solar energy collecting power forecasting method based on grey neural network |
CN108230043A (en) * | 2018-01-31 | 2018-06-29 | 安庆师范大学 | A kind of product area pricing method based on Grey Neural Network Model |
Non-Patent Citations (2)
Title |
---|
石磊: "灰色神网络对海歆团洗衣机订单需求预测分析", 《阜阳职业技术学院学报》 * |
黄劲潮: "灰色神经网络在空调订单预报中的应用", 《宜宾学院学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
O'Neill et al. | The effect of urbanization on energy use in India and China in the iPETS model | |
CN111563706A (en) | Multivariable logistics freight volume prediction method based on LSTM network | |
CN109214948A (en) | A kind of method and apparatus of electric system heat load prediction | |
CN107622329A (en) | The Methods of electric load forecasting of Memory Neural Networks in short-term is grown based on Multiple Time Scales | |
Phyo et al. | Electricity load forecasting in Thailand using deep learning models | |
CN107563705A (en) | Household electrical appliances product safety stock and the system and method ordered goods again are analyzed using big data | |
CN113537600B (en) | Medium-long-term precipitation prediction modeling method for whole-process coupling machine learning | |
CN109118013A (en) | A kind of management data prediction technique, readable storage medium storing program for executing and forecasting system neural network based | |
CN108710905B (en) | Spare part quantity prediction method and system based on multi-model combination | |
Boudon et al. | V-Mango: a functional–structural model of mango tree growth, development and fruit production | |
CN108596242A (en) | Power grid meteorology load forecasting method based on wavelet neural network and support vector machines | |
CN114492191A (en) | Heat station equipment residual life evaluation method based on DBN-SVR | |
Mitra | A white paper on scenario generation for stochastic programming | |
CN113722997A (en) | New well dynamic yield prediction method based on static oil and gas field data | |
Hänsel et al. | Intertemporal distribution, sufficiency, and the social cost of carbon | |
CN115759415A (en) | Power consumption demand prediction method based on LSTM-SVR | |
CN112132356A (en) | Stock price prediction method based on space-time diagram attention mechanism | |
CN115660725A (en) | Method for depicting multi-dimensional energy user portrait | |
Tang | The dynamic demand for capital and labor: Evidence from Chinese industrial firms | |
Li et al. | An Integrated Artificial Neural Network-based Precipitation Revision Model. | |
CN114581141A (en) | Short-term load prediction method based on feature selection and LSSVR | |
Afrah et al. | The Utilization of Deep Learning in Forecasting The Inflation Rate of Education Costs in Malang | |
CN110110927A (en) | A kind of network courses order demand prediction technique based on grey neural network | |
Li et al. | Prediction of china’s housing price based on a novel grey seasonal model | |
Nyandwi et al. | Neural network approach to short and long term load forecasting using weather conditioning |
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: 20190809 |