CN107862555A - Forecasting system and method based on exponential smoothing - Google Patents

Forecasting system and method based on exponential smoothing Download PDF

Info

Publication number
CN107862555A
CN107862555A CN201711237765.7A CN201711237765A CN107862555A CN 107862555 A CN107862555 A CN 107862555A CN 201711237765 A CN201711237765 A CN 201711237765A CN 107862555 A CN107862555 A CN 107862555A
Authority
CN
China
Prior art keywords
mrow
msubsup
data
value
exponential smoothing
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
Application number
CN201711237765.7A
Other languages
Chinese (zh)
Inventor
钟吉英
赵亮
谭斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Sichuan Changhong Electric Co Ltd
Original Assignee
Sichuan Changhong Electric Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Sichuan Changhong Electric Co Ltd filed Critical Sichuan Changhong Electric Co Ltd
Priority to CN201711237765.7A priority Critical patent/CN107862555A/en
Publication of CN107862555A publication Critical patent/CN107862555A/en
Pending legal-status Critical Current

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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Finance (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • Game Theory and Decision Science (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of forecasting system and method based on exponential smoothing, and webpage is crawled from specified website, and the source code of each Webpage got is parsed, and obtains related information in webpage, and by the data write into Databasce of acquisition；Selected section historic sales data establishes Method for Sales Forecast model as training data, and for being predicted to the sales data of following a period of time.Model of the present invention is simple, easily realizes, effectively improves the accuracy and availability of prediction.

Description

Forecasting system and method based on exponential smoothing
Technical field
The present invention relates to sales forecast technical field, and in particular to a kind of forecasting system and method based on exponential smoothing.
Background technology
One of central task of enterprise marketing plan is exactly sales forecast.Sales forecast is in following special time, entirely The estimation of the sales volume and consumption sum of portion's product or specific products.Sales forecast be take into full account following various influences because On the basis of element, with reference to the sale actual achievement of this enterprise, practicable sales target is proposed by certain analysis method.Sales volume The effect of prediction has at 3 points：First, enterprise can be helped effectively to arrange production, commodity stocks is reduced；Second, transport can be improved Management；Third, information content higher price and promotion decisions can be made.
The content of the invention
Instant invention overcomes the deficiencies in the prior art, there is provided a kind of forecasting system and method based on exponential smoothing, it is intended to The comment data after commodity is bought on e-commerce website using user, carries out Sales Volume of Commodity modeling, and then to following one section The sales volume situation of time commodity is estimated.
In view of the above mentioned problem of prior art, according to one side disclosed by the invention, the present invention uses following technology Scheme：
A kind of forecasting system based on exponential smoothing, including：
Data acquisition module, for crawling webpage from specified website, and to each Webpage got Source code is parsed, and obtains related information in webpage, and by the data write into Databasce of acquisition；
Method for Sales Forecast module, for selected section historic sales data as training data, Method for Sales Forecast model is established, with And for being predicted to the sales data of following a period of time.
In order to which the present invention is better achieved, further technical scheme is：
According to one embodiment of the invention, the training process of the Method for Sales Forecast module includes：
1) initial value S is set0 (1)、S0 (2), it is assumed that training sample number is Num；
As Num ＞ 15, the value of first sample is taken to make initial value, then
When Num≤15, the average of preceding 3 samples is taken to make initial value, then
2) interval 0.2 takes a value as α values from 0.1~1；
3) to each training sample, its secondary exponential forecasting value is calculated：
Calculate the mean absolute error under the α：
The 2) step is returned to, is calculated under different αError；
Corresponding α when taking Error minimums,Training terminates；Wherein,For last Training sample once, secondary exponential forecasting value.
According to another embodiment of the invention, the prediction deterministic process of the Method for Sales Forecast module includes：
Test process need to use the α that training process obtains,
Once, secondary index discriminant function is：
It is then the predicted value of t；
Due to when seeking an index, initial data x can be usedt-1Value, and in the period predicted, without initial data； The initial data of t is arranged to the predicted value of t and the average of the original value at (t-T) moment, so meets data Periodically, formula is：
The present invention can also be：
A kind of Forecasting Methodology based on exponential smoothing, it is characterised in that including：
Step 1：Data are carried out using web crawlers to the merchandise news and comment information of electric business website to crawl；
Step 2：Model training is predicted using historic sales data, obtains model parameter；
Step 3：Using the model obtained in step 2, the sales volume of following a period of time is predicted.
According to another embodiment of the invention, business is crawled from webpage using breadth-first pattern in the step 1 Product information.
According to another embodiment of the invention, the merchandise news includes comment on commodity data.
According to another embodiment of the invention, in the step 2, training sample is selected using year as the cycle, utilizes two Secondary exponential smoothing carries out model training.
Compared with prior art, one of beneficial effects of the present invention are：
A kind of forecasting system and method based on exponential smoothing of the present invention, its model is simple, easily realizes, effective lifting The accuracy and availability of prediction.
Brief description of the drawings
, below will be to embodiment for clearer explanation present specification embodiment or technical scheme of the prior art Or the required accompanying drawing used is briefly described in the description of prior art, it should be apparent that, drawings in the following description are only It is the reference to some embodiments in present specification, for those skilled in the art, is not paying creative work In the case of, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is a Method for Sales Forecast schematic diagram being enumerated according to the scheme of one embodiment of the invention, and the 1-7 months is go through in figure History sales volume, the 8-12 months are prediction sales volume.
Embodiment
The present invention is described in further detail with reference to embodiment, but the implementation of the present invention is not limited to this.
A kind of forecasting system based on exponential smoothing, it buys the comment after commodity using user on e-commerce website Data, Sales Volume of Commodity modeling is carried out, and then the sales volume situation of following a period of time commodity is estimated.The technology bag provided Contain：Crawler technology, forecast analysis technology.The main functional modules included below for the present invention：
Data acquisition module：Data are carried out to the merchandise news and comment information of the electric business websites such as Jingdone district using web crawlers Crawl.
Method for Sales Forecast models：According to historic sales data, double smoothing forecast model is established, to following a period of time Sales data be predicted.
Specifically, data acquisition module：
The main flow of data acquisition is as follows：(1) since specified website (starting website), with the pattern of breadth-first Crawl webpage；(2) for the webpage that each gets, its page source code is parsed, obtains letter related in webpage Breath, such as：User comment information etc.；(3) database is write data into.
Method for Sales Forecast models：
Selected section historic sales data establishes Method for Sales Forecast model, to the pin of following a period of time as training data Data are sold to be predicted.(here, carry out Method for Sales Forecast by the use of comment on commodity data volume as offtake.)
Key step is as follows：
Input：
train：Historic sales data, the sales volume at each time point in continuous time section is stored, in chronological sequence order Preserve；
N：Need the time point number predicted；
T：The cycle of training data
Output：Predicted value
(A) training process
1) initial value S0 (1), S0 (2) are set.Assuming that training sample number is Num.
As Num ＞ 15, the value of first sample is taken to make initial value.Then
When Num≤15, the average of preceding 3 samples is taken to make initial value.Then
2) value is taken as α values from 0.1~1 interval 0.2；
3) to each training sample, its secondary exponential forecasting value is calculated：
Calculate the mean absolute error under the α：
The 2nd step is returned to, is calculated under different αError。
Corresponding α when taking Error minimums,Training terminates.Wherein,For last Training sample once, secondary exponential forecasting value.
(B) deterministic process is predicted
Test process need to use the α that training process obtains,
Once secondary index discriminant function is：
It is then the predicted value of t.
Due to when seeking an index, initial data x can be usedt-1Value.And in the period predicted, it is no original number According to.Therefore, the initial data of t is arranged to the predicted value of t and the average of the original value at (t-T) moment by us, So meet the periodicity of data.Formula is as follows：
" one embodiment " for being spoken of in this manual, " another embodiment ", " embodiment ", etc., refer to tying Specific features, structure or the feature for closing embodiment description are included at least one embodiment of the application generality description In.It is not necessarily to refer to same embodiment that statement of the same race, which occur, in multiple places in the description.Appoint furthermore, it is understood that combining When one embodiment describes a specific features, structure or feature, what is advocated is this to realize with reference to other embodiment Feature, structure or feature are also fallen within the scope of the present invention.
Although reference be made herein to invention has been described for multiple explanatory embodiments of the invention, however, it is to be understood that Those skilled in the art can be designed that a lot of other modifications and embodiment, and these modifications and embodiment will fall in this Shen Please be within disclosed spirit and spirit.More specifically, can be to master in the range of disclosure and claim The building block and/or layout for inscribing composite configuration carry out a variety of variations and modifications.Except what is carried out to building block and/or layout Outside variations and modifications, to those skilled in the art, other purposes also will be apparent.

Claims (7)

  1. A kind of 1. forecasting system based on exponential smoothing, it is characterised in that including：
    Data acquisition module, for crawling webpage, and the source generation to each Webpage got from specified website Code is parsed, and obtains related information in webpage, and by the data write into Databasce of acquisition；
    Method for Sales Forecast module, for selected section historic sales data as training data, establish Method for Sales Forecast model, Yi Jiyong It is predicted in the sales data to following a period of time.
  2. 2. the forecasting system according to claim 1 based on exponential smoothing, it is characterised in that the Method for Sales Forecast module Training process includes：
    1) initial value S is set0 (1)、S0 (2), it is assumed that training sample number is Num；
    As Num ＞ 15, the value of first sample is taken to make initial value, then
    When Num≤15, the average of preceding 3 samples is taken to make initial value, then
    <mrow> <msubsup> <mi>S</mi> <mn>0</mn> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <msubsup> <mi>S</mi> <mn>0</mn> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <mfrac> <mrow> <mi>t</mi> <mi>r</mi> <mi>a</mi> <mi>i</mi> <mi>n</mi> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> <mo>+</mo> <mi>t</mi> <mi>r</mi> <mi>a</mi> <mi>n</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>+</mo> <mi>t</mi> <mi>r</mi> <mi>a</mi> <mi>i</mi> <mi>n</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </mrow> <mn>3</mn> </mfrac> </mrow>
    2) interval 0.2 takes a value as α values from 0.1~1；
    3) to each training sample, its secondary exponential forecasting value is calculated：
    <mrow> <msubsup> <mi>S</mi> <mi>t</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <msub> <mi>&amp;alpha;x</mi> <mi>L</mi> </msub> <mo>+</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>&amp;alpha;</mi> <mo>)</mo> </mrow> <msubsup> <mi>S</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> </mrow>
    <mrow> <msubsup> <mi>S</mi> <mi>t</mi> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> <mo>=</mo> <msubsup> <mi>&amp;alpha;S</mi> <mi>t</mi> <mrow> <mo>(</mo> <mn>1</mn> <mo>)</mo> </mrow> </msubsup> <mo>+</mo> <mrow> <mo>(</mo> <mn>1</mn> <mo>-</mo> <mi>&amp;alpha;</mi> <mo>)</mo> </mrow> <msubsup> <mi>S</mi> <mrow> <mi>t</mi> <mo>-</mo> <mn>1</mn> </mrow> <mrow> <mo>(</mo> <mn>2</mn> <mo>)</mo> </mrow> </msubsup> </mrow>
    Calculate the mean absolute error under the α：
    <mrow> <mi>E</mi> <mi>r</mi> <mi>r</mi> <mi>o</mi> <mi>r</mi> <mo>=</mo> <mfrac> <mn>1</mn> <mrow> <mi>N</mi> <mi>u</mi> <mi>m</mi> </mrow> </mfrac> <munderover> <mo>&amp;Sigma;</mo> <mrow> <mi>t</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <mi>N</mi> <mi>u</mi> <mi>m</mi> </mrow> </munderover> <mrow> <mo>|</mo> <mrow> <msub> <mi>x</mi> <mi>t</mi> </msub> <mo>-</mo> <msubsup> <mi>S</mi> <mi>t</mi> <mn>2</mn> </msubsup> </mrow> <mo>|</mo> </mrow> </mrow>
    The 2) step is returned to, is calculated under different αError；
    Corresponding α when taking Error minimums,Training terminates；Wherein,Trained for last Sample once, secondary exponential forecasting value.
  3. 3. the forecasting system according to claim 1 based on exponential smoothing, it is characterised in that the Method for Sales Forecast module Prediction deterministic process includes：
    Test process need to use the α that training process obtains,
    Once, secondary index discriminant function is：
    It is then the predicted value of t；
    Due to when seeking an index, initial data x can be usedt-1Value, and in the period predicted, without initial data；By t The initial data at moment is arranged to the predicted value of t and the average of the original value at (t-T) moment, so meets the cycle of data Property, formula is：
    <mrow> <msub> <mi>x</mi> <mi>t</mi> </msub> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>S</mi> <mi>t</mi> <mn>2</mn> </msubsup> <mo>+</mo> <msub> <mi>x</mi> <mrow> <mi>t</mi> <mo>-</mo> <mi>T</mi> </mrow> </msub> </mrow> <mn>2</mn> </mfrac> <mo>.</mo> </mrow>
  4. A kind of 4. Forecasting Methodology based on exponential smoothing for realizing system as claimed in claim 1, it is characterised in that including：
    Step 1：Data are carried out using web crawlers to the merchandise news and comment information of electric business website to crawl；
    Step 2：Model training is predicted using historic sales data, obtains model parameter；
    Step 3：Using the model obtained in step 2, the sales volume of following a period of time is predicted.
  5. 5. the Forecasting Methodology according to claim 4 based on exponential smoothing, it is characterised in that width is utilized in the step 1 Mode of priority crawls merchandise news from webpage.
  6. 6. the Forecasting Methodology according to claim 5 based on exponential smoothing, it is characterised in that the merchandise news includes business Product comment data.
  7. 7. the Forecasting Methodology according to claim 4 based on exponential smoothing, it is characterised in that in the step 2, select instruction Practice sample using year as the cycle, model training is carried out using Secondary Exponential Smoothing Method.
CN201711237765.7A 2017-11-30 2017-11-30 Forecasting system and method based on exponential smoothing Pending CN107862555A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711237765.7A CN107862555A (en) 2017-11-30 2017-11-30 Forecasting system and method based on exponential smoothing

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711237765.7A CN107862555A (en) 2017-11-30 2017-11-30 Forecasting system and method based on exponential smoothing

Publications (1)

Publication Number Publication Date
CN107862555A true CN107862555A (en) 2018-03-30

Family

ID=61704181

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711237765.7A Pending CN107862555A (en) 2017-11-30 2017-11-30 Forecasting system and method based on exponential smoothing

Country Status (1)

Country Link
CN (1) CN107862555A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509030A (en) * 2018-11-15 2019-03-22 北京旷视科技有限公司 Method for Sales Forecast method and its training method of model, device and electronic system
CN111275514A (en) * 2020-01-07 2020-06-12 载信软件(上海)有限公司 Intelligent purchasing method and system, storage medium and electronic device
CN111327464A (en) * 2020-02-12 2020-06-23 安超云软件有限公司 Method and system for determining virtual router
CN115423223A (en) * 2022-11-04 2022-12-02 山东捷瑞数字科技股份有限公司 Industrial internet production prediction device and method for machine manufacturing enterprises

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103617459A (en) * 2013-12-06 2014-03-05 李敬泉 Commodity demand information prediction method under multiple influence factors
CN104820938A (en) * 2015-05-15 2015-08-05 南京大学 Optimal ordering period prediction method for seasonal and periodic goods
CN105869019A (en) * 2016-03-31 2016-08-17 金蝶软件（中国）有限公司 Method and apparatus for predicting goods price
CN107146015A (en) * 2017-05-02 2017-09-08 联想(北京)有限公司 Multivariate Time Series Forecasting Methodology and system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103617459A (en) * 2013-12-06 2014-03-05 李敬泉 Commodity demand information prediction method under multiple influence factors
CN104820938A (en) * 2015-05-15 2015-08-05 南京大学 Optimal ordering period prediction method for seasonal and periodic goods
CN105869019A (en) * 2016-03-31 2016-08-17 金蝶软件（中国）有限公司 Method and apparatus for predicting goods price
CN107146015A (en) * 2017-05-02 2017-09-08 联想(北京)有限公司 Multivariate Time Series Forecasting Methodology and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
孙静娟: "《经济预测 理论·方法·评价》", 31 May 1999, 中国经济出版社 *
王桂红 等: "一种提高农产品市场价格预测精度的改进算法", 《浙江农业学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509030A (en) * 2018-11-15 2019-03-22 北京旷视科技有限公司 Method for Sales Forecast method and its training method of model, device and electronic system
CN111275514A (en) * 2020-01-07 2020-06-12 载信软件(上海)有限公司 Intelligent purchasing method and system, storage medium and electronic device
CN111327464A (en) * 2020-02-12 2020-06-23 安超云软件有限公司 Method and system for determining virtual router
CN115423223A (en) * 2022-11-04 2022-12-02 山东捷瑞数字科技股份有限公司 Industrial internet production prediction device and method for machine manufacturing enterprises

Similar Documents

Publication Publication Date Title
CN107862555A (en) Forecasting system and method based on exponential smoothing
Sun et al. SFFS-PC-NN optimized by genetic algorithm for dynamic prediction of financial distress with longitudinal data streams
Bardwell et al. Most recent changepoint detection in panel data
CN101329683A (en) Recommendation system and method
Alwan et al. Sustainable and resilient e-commerce under COVID-19 pandemic: a hybrid grey decision-making approach
Chang et al. Retracted: ergonomic techniques for a mobile e‐invoice system: operational requirements of an information management system
CN109903099A (en) Model building method and system for score in predicting
Zhang et al. Real-time shop-floor production performance analysis method for the internet of manufacturing things
Salutina et al. Transformation of business technologies into digital platforms and evaluation of the effectiveness of their application
Okwu et al. Artificial neural network model for cost optimization in a dual-source multi-destination outbound system
Kaoud et al. Design and optimization of the dual-channel closed loop supply chain with e-commerce
He et al. Application of unascertained number for the integration of carbon footprint in conceptual design
CN106651427A (en) Data association method based on user behavior
Fosgerau et al. Bikeability and the induced demand for cycling
Mancini et al. Responsible materials management for a resource-efficient and low-carbon society
Anil et al. Performance analysis of deep learning architectures for recommendation systems
Wen et al. Green smart manufacturing: energy-efficient robotic job shop scheduling models
Mulligan Computationally networked urbanism and advanced sustainability analytics in internet of things-enabled smart city governance
Tahiduzzaman et al. Big data and its impact on digitized supply chain management
Hong et al. Two-stage double bootstrap data envelopment analysis for efficiency evaluation of shared-bicycle stations in urban cities
Yuan et al. Can online user behavior improve the performance of sales prediction in E-commerce?
Zhou et al. Extreme value modeling of coincident lane load effects for multi-lane factors of bridges using peaks-over-threshold method
CN106682151A (en) Personalized educational-resource recommendation method and system
Xu Risk factor analysis combined with deep learning in the risk assessment of overseas investment of enterprises
Azadeh et al. An adaptive neural network-fuzzy linear regression approach for improved car ownership estimation and forecasting in complex and uncertain environments: the case of Iran

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

Application publication date: 20180330

RJ01 Rejection of invention patent application after publication