CN106097015A - A kind of market prediction system and method - Google Patents
A kind of market prediction system and method Download PDFInfo
- Publication number
- CN106097015A CN106097015A CN201610442163.4A CN201610442163A CN106097015A CN 106097015 A CN106097015 A CN 106097015A CN 201610442163 A CN201610442163 A CN 201610442163A CN 106097015 A CN106097015 A CN 106097015A
- Authority
- CN
- China
- Prior art keywords
- business datum
- model
- data
- factor data
- factor
- 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
- 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
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of market prediction system and method, by collecting factor data and business datum, judge whether this factor data model corresponding with business datum exists, if having existed, then directly new business datum is predicted, if this factor data model corresponding with business datum does not exists, then the factor data collected and business datum are stored, and the factor data and business datum according to storage creates new model.The present invention can carry out the prediction of new business data by the existing model factor data to collecting and business datum, non-existent model can also be created, to ensure the perfect of model data, the market prediction system and method that the present invention provides can more adapt to the demand in market, the more business datum in Accurate Prediction market.
Description
Technical field
The present invention relates to a kind of market prediction system and method, particularly a kind of by market forecast model Accurate Prediction market
The system and method for demand.
Background technology
Along with the development of market economy, market development is entered into the Cost Competition epoch by the in-depth of price competition epoch,
Judgement in advance to market future trend, plays the most important effect in the operation management of enterprise, to a great extent,
Affect enterprise's cognitron meeting and the ability seized the opportunity.Currently, the market analysis for commodity is embodied in historical data more
Variation analysis and the event analysis aspect of influence factor, the forecast analysis for market future trend only rests on qualitative analysis
Category, lack using data as the quantitative analysis supported, say, that be also not set up perfect and can quantify
Market forecast model, cannot accurately predict the impact of commodity market demand various factors.
Summary of the invention
For the problems referred to above, the present invention provides a kind of market prediction system, and this market prediction system comprises: data collection mould
Block, data memory module, model memory module, the newly-built module of model and output module, wherein, data gathering module is used for collecting
Factor data and business datum；Data memory module is used for storing factor data and business datum；Model memory module is used for depositing
Storage model formation；The newly-built module of model is for carrying out newly-built to the model not found；Output module is for exporting the industry of prediction
Business data；Factor data and business datum that data gathering module is collected are stored in data memory module, if factor data pair
The business datum answered can find the model of correspondence in model memory module, then directly released new by the factor data collected
Business datum, if business datum corresponding to factor data does not finds the model of correspondence, then factor data in model memory module
After accumulating in a large number with business datum, by the model that the newly-built module creation of model is new, and it is stored in model memory module.
According to above-mentioned market prediction system, this market prediction system includes authentication module, for carrying out new established model
Checking.
According to above-mentioned market prediction system, data gathering module uses crawler technology to obtain data.
According to above-mentioned market prediction system, the acquisition of information data that data gathering module is manually entered according to personnel.
According to above-mentioned market prediction system, the newly-built module of model to model newly-built time, choose any two time point
Factor data and business datum, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)+...+an* (Efn-
Ecn)) * Dc carries out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n
Factor data is at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction
Business datum value, Dc is current service data value.
The present invention separately provides a kind of market prediction method, comprises the steps of:
S1: collect factor data and business datum；
S2: judge whether this factor data model corresponding with business datum exists, the most then carry out step S3, if it is not,
Then skip to step S4；
S3: business datum is predicted；
S4: storage factor data and business datum；
S5: create new model；
Collect factor data and business datum, it is judged that need whether the factor data of prediction and business datum exist corresponding mould
Type, if there is corresponding model, is then predicted business datum according to already present model, if there is not model, then storage is searched
The factor data of collection and business datum, and create new model according to the factor data stored and business datum.
According to a kind of above-mentioned market prediction method, after step S5, S6 the most in steps: new model is verified.
According to a kind of above-mentioned market prediction method, step S1 is collected factor data and business datum uses reptile skill
Art.
According to a kind of above-mentioned market prediction method, step S1 collects factor data and business datum is the most defeated by personnel
Enter acquisition of information.
According to a kind of above-mentioned market prediction method, step S5 creates new model, chooses the factor of any two time point
Data and business datum, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)+...+an* (Efn-Ecn)) * Dc
Carry out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n
Factor data is at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction
Business datum value, Dc is current service data value.
The present invention provides a kind of market prediction system and method, by collecting factor data and business datum, it is judged that this because of
Whether the prime number evidence model corresponding with business datum exists, if having existed, is then directly predicted new business datum, if
This factor data model corresponding with business datum does not exists, then the factor data collected and business datum are stored,
And the factor data and business datum according to storage creates new model.The present invention can be by existing model to collecting
Factor data and business datum carry out the prediction of new business data, it is also possible to create non-existent model, to ensure mould
Type data perfect, the market prediction system and method that the present invention provides can more adapt to the demand in market, more accurately pre-
Survey the business datum in market.
Accompanying drawing explanation
Fig. 1 is market prediction system schematic of the present invention；
Fig. 2 is market prediction method flow diagram of the present invention.
Detailed description of the invention
For making the purpose of the present invention, feature and function thereof are had further understanding, embodiment is hereby coordinated to describe in detail such as
Under.
First the several nouns related in the present patent application are explained, hereinafter described business datum, refer to
The data of reaction merchandise sales situation, such as trade name, sales volume and stock etc., business datum is divided into current service data and prediction
Business datum；Following factor data, refers to affect the factor of business datum, such as weather, population, sales promotion information etc., factor
Data are divided into current factor data and anticipated factor data；Following model, refers to the computing between factor data and business datum
Relation.Specifically, factor data can affect the business datum of commodity, and factor data can pass through model prediction business datum,
Substantial amounts of factor data accumulates with business datum, can create new model.
The present invention provides a kind of market prediction system, and seeing Fig. 1, Fig. 1 is market prediction system schematic of the present invention, this city
Field prediction system comprises: data gathering module 1, data memory module 2, model memory module 3, the newly-built module of model 4 and output
Module 5, wherein, data gathering module 1 is used for collecting factor data and business datum；Data memory module 2 is used for storing factor
Data and business datum；Model memory module 3 is used for storing model formation；The newly-built module of model 4 is for the mould not found
Type carries out newly-built；Output module 5 is for exporting the business datum of prediction.
Data gathering module 1 collected factor data and business datum, in model memory module 3, there is this because of prime number
According to the model with business datum, then according to factor data and existing business datum and corresponding model, business number can be predicted
According to, it was predicted that business datum exports via output module 5 and shows user；Data gathering module 1 collected factor data and industry
, in model memory module 3, there is not the model of this factor data and business datum in business data, then data gathering module 1 is collected
To factor data and business datum be stored in data memory module 2, when data memory module 2 stores substantial amounts of factor
After data and business datum, substantial amounts of factor data and business datum are sent to the newly-built module of model 4, by the newly-built module of model
4 create new models, and the new model of establishment are stored in model memory module 3, input again next time this new established model corresponding because of
Prime number evidence and business datum, can directly be predicted business according to new established model；Output module 5 is for prediction business number
According to exporting, can be shown with chart, such as cake chart, line diagram, thermodynamic chart or radar map etc..The present invention compares existing
Some market prediction systems, can more accurately be predicted marketing business datum.
According to above-mentioned market prediction system, this market prediction system includes authentication module 6, for carrying out new established model
Checking.Owing to new established model is to create based on substantial amounts of factor data and business datum, possess certain theoretical basis, but
It is that, in order to preferably adapt to the demand of marketing business datum prediction, the present invention adds checking mould after the newly-built module of model 4
Block 6, newly created model is verified by authentication module 6, by the prediction business datum of newly-built model prediction and market reality
Commodity business datum is compared, if difference is less than certain threshold value after Yan Zheng, then new established model stores model storage mould
In block 3, if difference is more than certain threshold value after Yan Zheng, then continued to collect factor data and business datum by data gathering module 1,
New model is re-created by the newly-built module of model 4.
According to above-mentioned market prediction system, data gathering module 1 uses crawler technology to obtain data, by system via climbing
Worm technology and third party's interface obtain factor data and business datum, such as weather conditions data, demographic data, merchandise sales number
According to etc..
It addition, the acquisition of information data that data gathering module 1 is manually entered also dependent on personnel, such as some business datum
Need by being manually entered acquisition, such as, the sales data of a certain history node, the promotion approach etc. of a certain history node.
According to above-mentioned market prediction system, the newly-built module of model 4 to model newly-built time, choose any two time point
Factor data and business datum, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)+...+an* (Efn-
Ecn)) * Dc carries out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n
Factor data is at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction
Business datum value, Dc is current service data value.
For example, first factor data is weather conditions, then Ec1 be first time point chosen weather because of
Element value, Ef1 is the weather conditions value of second time point chosen, and a1 is that weather conditions data are on shared by business datum impact
Weight, second factor data are a certain regional population's data, then Ec2 is this regional population of first time point chosen
Data, Ef2 is these regional population's data of second time point chosen, and a2 is that these regional population's data are to shared by business datum
Weight, by that analogy, by current service data value and all factor data respectively first time point value, second time
Between point value and this factor data to shared by business datum weight substitute into formula, then can predict business datum Df.
In above-mentioned factor data, can be numerical value corresponding to some factor, it is also possible to be amount corresponding to a certain class factor
Temperature value in weather conditions when integrated value after change, such as Ec1 can be first time points, it is also possible to be first time
Functional value after the quantization of temperature, wind-force, humidity etc. in weather conditions during point, factor data value is not done concrete limit by the present invention
Fixed, different expression waies can be selected according to the actual requirements.
The present invention separately provides a kind of market prediction method, and Fig. 2 is market prediction method flow diagram of the present invention, comprises following step
Rapid:
S1: collect factor data and business datum；
S2: judge whether this factor data model corresponding with business datum exists, the most then carry out step S3, if it is not,
Then skip to step S4；
S3: business datum is predicted；
S4: storage factor data and business datum；
S5: create new model；
First collecting factor data and business datum, business datum now is current service data, it is judged that collected
Whether factor data and this commodity current service data exist the model of correspondence, if there is corresponding model, then by factor data,
Business datum according to already present model to prediction business datum be predicted, it was predicted that after prediction business datum can be with chart
Mode shows user, chart to include cake chart, line diagram, thermodynamic chart or radar map etc., if there is not model, then to collection
Factor data and current service data store, after storing substantial amounts of factor data and current service data, according to storage
Factor data and business datum create new model, next time input factor data corresponding to this new established model and business datum again,
Directly according to new established model, business can be predicted.Comparing existing market prediction system, the present invention can be the most right
Marketing business datum is predicted.
According to a kind of above-mentioned market prediction method, after step S5, S6 the most in steps: new model is verified.By
It is to create based on substantial amounts of factor data and business datum in new established model, possesses certain theoretical basis, but, in order to
Preferably adapting to the demand of marketing business datum prediction, the present invention adds step S6 after step S5, and step S6 is to new mould
Type is verified, commodity business datum actual with market for the prediction business datum of newly-built model prediction is compared, if
After checking, difference is less than certain threshold value, then stored by new established model, if difference is more than certain threshold value after Yan Zheng, then continues
Collect factor data and current service data, re-create new model.
According to a kind of above-mentioned market prediction method, step S1 is collected factor data and business datum uses reptile skill
Art, is obtained factor data and business datum, such as weather conditions data, population by system via crawler technology and third party's interface
Data, article sales data etc..
It addition, collection factor data and business datum are manually entered acquisition of information by personnel in step S1, such as some industry
Business data need by being manually entered acquisition, such as, and the sales data of a certain history node, the promotion approach of a certain history node
Deng.
According to a kind of above-mentioned market prediction method, step S5 creates new model, chooses the factor of any two time point
Data and business datum, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)+...+an* (Efn-Ecn)) * Dc
Carry out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n
Factor data is at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction
Business datum value, Dc is current service data value.
For example, first factor data is weather conditions, then Ec1 be first time point chosen weather because of
Element value, Ef1 is the weather conditions value of second time point chosen, and a1 is that weather conditions data are on shared by business datum impact
Weight, second factor data are a certain regional population's data, then Ec2 is this regional population of first time point chosen
Data, Ef2 is these regional population's data of second time point chosen, and a2 is that these regional population's data are to shared by business datum
Weight, by that analogy, by current service data value and all factor data respectively first time point value, second time
Between point value and this factor data to shared by business datum weight substitute into formula, then can predict business datum Df.
In above-mentioned factor data, can be numerical value corresponding to some factor, it is also possible to be amount corresponding to a certain class factor
Temperature value in weather conditions when integrated value after change, such as Ec1 can be first time points, it is also possible to be first time
Functional value after the quantization of temperature, wind-force, humidity etc. in weather conditions during point, factor data value is not done concrete limit by the present invention
Fixed, different expression waies can be selected according to the actual requirements.
The present invention provides a kind of market prediction system and method, by collecting factor data and business datum, it is judged that this because of
Whether the prime number evidence model corresponding with business datum exists, if having existed, is then directly predicted new business datum, if
This factor data model corresponding with business datum does not exists, then the factor data collected and business datum are stored,
And the factor data and business datum according to storage creates new model.The present invention can be by existing model to collecting
Factor data and business datum carry out the prediction of new business data, it is also possible to create non-existent model, to ensure mould
Type data perfect, the market prediction system and method that the present invention provides can more adapt to the demand in market, more accurately pre-
Survey the business datum in market.
The present invention is been described by by above-mentioned related embodiment, but above-described embodiment is only the example implementing the present invention.
It must be noted that, the embodiment disclosed is not limiting as the scope of the present invention, without departing from the spirit and scope of the present invention
The change made and retouching, all belong to the scope of patent protection of the present invention.
Claims (10)
1. a market prediction system, it is characterised in that this market prediction system comprises:
Data gathering module, is used for collecting factor data and business datum；
Data memory module, is used for storing factor data and business datum；
Model memory module, is used for storing model formation；
The newly-built module of model, for carrying out newly-built to the model not found；
Output module, for exporting the business datum of prediction；
Wherein, factor data and business datum that data gathering module is collected are stored in data memory module, if factor data
Corresponding business datum can find the model of correspondence in model memory module, then directly released new by the factor data collected
Business datum, if business datum corresponding to factor data does not finds the model of correspondence in model memory module, then because of prime number
After accumulating in a large number with business datum, by the model that the newly-built module creation of model is new, and it is stored in model memory module.
Market prediction system the most according to claim 1, it is characterised in that this market prediction system includes authentication module,
For new established model is verified.
Market prediction system the most according to claim 1, it is characterised in that data gathering module uses crawler technology to obtain
Data.
Market prediction system the most according to claim 1, it is characterised in that data gathering module is manually entered according to personnel
Acquisition of information data.
Market prediction system the most according to claim 1, it is characterised in that the newly-built module of model to model newly-built time, choosing
Take factor data and the business datum of any two time point, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)
+ ...+an* (Efn-Ecn)) * Dc carries out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n factor
Data are at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction business
Data value, Dc is current service data value.
6. a market prediction method, it is characterised in that this market prediction method comprises the steps of:
S1: collect factor data and business datum；
S2: judge whether this factor data model corresponding with business datum exists, the most then carry out step S3, if it is not, then jump
To step S4；
S3: business datum is predicted；
S4: storage factor data and business datum；
S5: create new model；
Collect factor data and business datum, it is judged that need whether the factor data of prediction and business datum exist corresponding model,
If there is corresponding model, being then predicted business datum according to already present model, if there is not model, then storage is collected
Factor data and business datum, and create new model according to the factor data stored and business datum.
A kind of market prediction method the most according to claim 6, it is characterised in that after step S5, S6 the most in steps: right
New model is verified.
A kind of market prediction method the most according to claim 6, it is characterised in that collect factor data and industry in step S1
Business data acquisition crawler technology.
A kind of market prediction method the most according to claim 6, it is characterised in that collect factor data and industry in step S1
Business data are manually entered acquisition of information by personnel.
A kind of market prediction method the most according to claim 6, it is characterised in that step S5 creates new model, chooses and appoints
The factor data of two time points of meaning and business datum, according to formula Df=(1+a1* (Ef1-Ec1)+a2* (Ef2-Ec2)
+ ...+an* (Efn-Ecn)) * Dc carries out newly-built；
Wherein a1, a2 ... an is the weight that business datum is affected by n factor data, Ec1, Ec2 ... Ecn is n factor
Data are at the value of first time point, Ef1, Ef2 ... Efn is n the factor data value at the second time point, and Df is prediction business
Data value, Dc is current service data value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610442163.4A CN106097015A (en) | 2016-06-20 | 2016-06-20 | A kind of market prediction system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610442163.4A CN106097015A (en) | 2016-06-20 | 2016-06-20 | A kind of market prediction system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106097015A true CN106097015A (en) | 2016-11-09 |
Family
ID=57236976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610442163.4A Pending CN106097015A (en) | 2016-06-20 | 2016-06-20 | A kind of market prediction system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106097015A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767836A (en) * | 2018-12-29 | 2019-05-17 | 上海亲看慧智能科技有限公司 | A kind of medical diagnosis artificial intelligence system, device and its self-teaching method |
| CN110335092A (en) * | 2019-07-15 | 2019-10-15 | 联想(北京)有限公司 | A kind of data processing method, device and calculate equipment |
| CN112381560A (en) * | 2020-10-23 | 2021-02-19 | 东北石油大学 | Shared equipment product market prediction system and method |
-
2016
- 2016-06-20 CN CN201610442163.4A patent/CN106097015A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109767836A (en) * | 2018-12-29 | 2019-05-17 | 上海亲看慧智能科技有限公司 | A kind of medical diagnosis artificial intelligence system, device and its self-teaching method |
| CN110335092A (en) * | 2019-07-15 | 2019-10-15 | 联想(北京)有限公司 | A kind of data processing method, device and calculate equipment |
| CN112381560A (en) * | 2020-10-23 | 2021-02-19 | 东北石油大学 | Shared equipment product market prediction system and method |
| CN112381560B (en) * | 2020-10-23 | 2022-10-21 | 东北石油大学 | Shared equipment product market prediction system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Thowsen | A dynamic, nonstationary inventory problem for a price/quantity setting firm | |
| CN102654909B (en) | SLA pro-jobs level establishing method and system | |
| US9726502B2 (en) | Route planner for transportation systems | |
| US7730022B2 (en) | Data processing system and method for supply chain management | |
| CN109360413B (en) | Vehicle scheduling method and system | |
| CN107798341A (en) | User view Forecasting Methodology, electronic equipment and computer-readable recording medium | |
| US11551078B2 (en) | Data processing apparatus and data processing method | |
| CN106504534B (en) | A kind of method, apparatus and user equipment for predicting road conditions | |
| CN112182421B (en) | Potential customer mining method and device and electronic equipment | |
| CN105096152A (en) | Commodity popularity-based operation execution method and device | |
| CN106097015A (en) | A kind of market prediction system and method | |
| CN111476441A (en) | Load prediction method for electric vehicle charging equipment and related device | |
| CN110083633A (en) | Public affairs based on big data analysis estimate company evaluation method, apparatus, terminal and medium | |
| US20130060712A1 (en) | Bulk Distribution Method | |
| CN106127316A (en) | Automated vehicle maintenance prediction system | |
| Chao et al. | Optimal policy for a periodic-review inventory system under a supply capacity contract | |
| CN113762926B (en) | Intelligent supervision system and intelligent supervision method | |
| CN110986992A (en) | Navigation method and device for unmanned vending vehicle, electronic equipment and storage medium | |
| CN106779245B (en) | Event-based civil aviation demand prediction method and device | |
| Schwarz et al. | On the interactions between routing and inventory-management policies in a one-warehouse n-retailer distribution system | |
| CN107844496A (en) | Statistical information output intent and device | |
| CN113962656A (en) | Power grid data asset management method, system, equipment and storage medium | |
| CN109360041A (en) | Merchant display method and device, electronic equipment and storage medium | |
| CN115016902A (en) | Industrial process digital management system and method | |
| Tarapata et al. | Machine learning algorithms for the problem of optimizing the distribution of parcels in time-dependent networks: the case study |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |
Application publication date: 20161109 |
|
| RJ01 | Rejection of invention patent application after publication |