CN108376260A - A kind of SVR tourism demand prediction techniques based on optimal subset optimization - Google Patents
A kind of SVR tourism demand prediction techniques based on optimal subset optimization Download PDFInfo
- Publication number
- CN108376260A CN108376260A CN201810092083.XA CN201810092083A CN108376260A CN 108376260 A CN108376260 A CN 108376260A CN 201810092083 A CN201810092083 A CN 201810092083A CN 108376260 A CN108376260 A CN 108376260A
- Authority
- CN
- China
- Prior art keywords
- data
- optimal
- parameter
- subset
- svr
- 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
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000005457 optimization Methods 0.000 title claims abstract description 21
- 238000002790 cross-validation Methods 0.000 claims abstract description 7
- 238000012098 association analyses Methods 0.000 claims abstract description 6
- 238000007405 data analysis Methods 0.000 claims abstract description 5
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 23
- 230000000694 effects Effects 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 230000004044 response Effects 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 235000013305 food Nutrition 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Mathematics (AREA)
- Operations Research (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of SVR tourism demand prediction techniques based on optimal subset optimization, including:1. including reflecting the exponential quantity of volumes of searches in n data analysis service respectively according to chronological classification acquisition tourist attractions scenic spot reception number data, search data and temperature record, the search data;2. using the relationship between the sight spot reception number data and (n+3) a index of passenger flow of Gray Association Analysis analysis of history;3. pair passenger flow data carries out regression fit, the penalty parameter c and kernel functional parameter g that cross validation selection returns are obtained, fitting predicted value Mse and prediction index R is obtained.The present invention takes continuous sequence of data segments to be realized by time response as input sequence of subsets to replace original historical data sequence, the dynamic input prediction model for waiting dimensions to fill vacancies in the proper order.So as to show that the influence factor of tourist flows is many aspects.
Description
【Technical field】
The present invention relates to tourism data processing technology fields, and in particular to a kind of SVR tourisms based on optimal subset optimization
Needing forecasting method.
【Background technology】
The development of tourism industry relies primarily on food and drink, retail, lodging, communications and transportation, the support of show business, so necessary
Develop universe tourism, payes attention to the development of tourist industry.It accurately predicts following tourism demand, establishes corresponding mathematical model, it is right
There is important role in formulating tourist industry decision plan.
Passenger flow forecast this problem in scenic spot was particularly important in recent years in China.Crowding phenomenon shows that the volume of the flow of passengers is short
The phase accuracy of prediction is a problem to be solved
Although traditional travel forecasting data come from the statistical report of tourism departments of government mostly, often
There is hysteresis quality, and situations such as shortage of data, sample is insufficient usually occurs in travel network publication statistical data, not so as to cause it
Effective prediction result can be reached.Therefore, new data source is found, is had more comprehensively, covering scope is wider, and quality is higher
Feature, to improve the accuracy of prediction.Because history passenger flow data sequence includes often certain trend and cycle information,
In addition to the passenger flow data of history, other exogenous variables, all such as max. daily temperature, Daily minimum temperature, temperature difference per day, day categorical variable
It may influence the precision of passenger flow estimation.Along with the development of information technology, occur some new prediction numbers on internet
According to --- search data, interactive data.Such as Baidu, Google, 360, microblogging, it is all proposed Baidu's index at present, Google becomes
Gesture, 360 indexes, micro- index.When tourist with internet when carrying out interactive, search engine platform is either used, or social
Media platform, mobile phone mobile platform etc., the data generated online can be all stored, and researcher can use search
Trend data, flow etc. are analyzed and passenger flow forecast situation.Wherein search engine data are used for reflecting one of tourist before going on a tour
The activity time and food and drink position that lodging information, scenic spot around a little ideas, such as the related scenic spot of search are held.And it is mutual
Networking search data can overcome the hysteresis quality of conventional statistics data, and basic history passenger flow data is tourist during visit
The true reflection of one of scenic spot and embodiment are used for reflecting some senses of tourist after trip by the data of social media interaction platform
By and opinion.
【Invention content】
The purpose of the present invention is to provide a kind of SVR tourism demand prediction techniques based on optimal subset optimization, by most
Excellent subset method and SVM regression prediction methods combine, and improve the accuracy of prediction.
In order to achieve the above objectives, the present invention adopts the following technical scheme that:
A kind of SVR tourism demand prediction techniques based on optimal subset optimization, include the following steps:
(1) described according to chronological classification acquisition tourist attractions scenic spot reception number data, search data and temperature record
It includes reflecting that the exponential quantity of volumes of searches, n are positive integer in n data analysis service respectively to search for data;The temperature record
Including day lowest temperature, the day highest temperature and the temperature difference;
(2) between the sight spot reception number data and (n+3) a index of passenger flow that use Gray Association Analysis analysis of history
Relationship, obtain gray relation grades ρ0i(m);(n+3) a index includes n exponential quantity, the day highest temperature, day lowest temperature and day temperature
Difference,
(3) regression fit is carried out to passenger flow data using optimal training subset method and the SVM regression model established, obtained
The penalty parameter c and kernel functional parameter g that cross validation selection returns are trained SVM fitting predictions according to the optimal parameter, are intended
Close predicted value Mse and prediction index R.
Preferably:Use in step (2) method of the degree of association of Gray Association Analysis acquisition (n+3) a index for:
(2.1) the initial matrix x of (n+3) a index of correlation is establishedi, and carry out initialization and obtain transformation matrix xi'
xi=(xi(1),xi(2),…xi(m),…) (1)
Wherein, xi(m) original value of the i indexs at the m days is indicated;
(2.2) difference sequence Δ0i(m) it calculates as follows
Δ0i(m)=| xi 0(m)-xi′(m)|,Δ0i(m)=(Δ0i(1),Δ0i(2),…Δ0i(m),…) (3)
(2.3) incidence coefficient ε is obtained0i(m), then gray relation grades ρ0i(m) calculation formula such as formula (5)
In formula (1), δ is defined as resolution ratio, δ ∈ (0,1).
Further:δ values are 0.5.
Preferably:The regression model of SVM foundation is to the method for passenger flow data progress regression fit in step (3):
(3.1) data prediction;Training set and test set are normalized, the normalized of use is as follows:
In formula, x, y ∈ Rn,xmin=min (x), xmax=max (x), normalized effect are arrived initial data is regular
[0,1] in range;
(3.2) it uses cross-validation method to obtain penalty parameter c and kernel functional parameter g, c and g is allowed to take in a certain range
Value obtains the training verification under this group for determining c and g using training set as raw data set, final so that training is accurate
Highest that group of c and g of rate is as optimal parameter;
(3.3) training and prediction;Using the optimal parameter obtained in step (3.2), SVM is trained and is returned is pre-
It surveys, it is final to obtain mean square error Mse and coefficient R.
Preferably:Optimal training subset method is in the step (3):
(3-1) intercepts continuous sequence of data segments as input sequence of subsets to replace original historical data sequence;
Build dynamic majorized subset's gray model:For the first time, it removes in tourist attraction history reception number original series
X(0)(1), x is filled vacancies in the proper order(0)(n+1), one group of new dynamic sequence (x is constituted(0)(2),x(0)(3),…x(0)(n+1))
And so on, it keeps sequence length constant, is built into dynamic majorized subset's gray model;
(3-2) determines the number of optimal input subset;
Intercept continuous data segment (x(0)(n-l),…x(0)(n-1),x(0)(n)), l ∈ (4,5 ... n-1) are as input
Collect sequence to replace original history reception number sequence (x(0)(1),x(0)(2),x(0)(3)…x(0)(n)), according to step (3)
The maximum value of the minimum value and coefficient R of the mean square error Mse of acquisition determines the number l of optimal input subset;
(3-3) chooses l optimal input subsets and carries out the final prediction result of prediction acquisition.
Preferably:Temperature record data are collected in weather net.
Preferably:It is that metering is classified according to year, the moon, week or day with number of days when according to chronological classification.
Preferably:Reflect respectively in n data analysis service volumes of searches exponential quantity be Baidu's index, Google trend,
At least one of 360 indexes and micro- index.
Preferably:In step (3), the regression model established using optimal training subset method and SVM to passenger flow data into
Row regression fit, when obtaining the optimal parameter c and g that cross validation selection returns;
Using training set as raw data set, obtained under each group of penalty parameter c and kernel functional parameter g using K-CV methods
The accuracy rate of training set fitting finally takes so that training set verification accuracy rate fitting effect highest parameter c and g is as best ginseng
Number;If multigroup parameter c and g corresponds to the equal highest of effect, the first group of parameter c and g searched is chosen as best parameter.
The SVR tourism demand prediction techniques based on optimal subset optimization of the present invention, from Tourism Bureau official website and both at home and abroad
Major search platform gathered data, analyzing influence index is to the influence degree of passenger flow, in conjunction with optimal subset method and SVM regression forecastings
Method solves the optimization problem of input, compared with existing technology, the invention has the advantages that:
1) the advantages of introducing grey correlation analysis is to find out tourist flows amount relevant with which passenger flow index, and shows
These indexs produce the volume of the flow of passengers influence of much degree, so as to show that the influence factor of tourist flows is many aspects
's.
2) comprehensive optimal subset method and SVM regression prediction methods, are concentrated mainly on the optimization to list entries, from theory
Angle says that the quality of prediction result be unable to do without the selection of input information.The advantages of optimal subset method be abandon some it is out-of-date when
Between state, the advantages of SVM regression prediction methods is to concentrate on the amendment to later stage prediction result and ignore to the excellent of list entries
Change.It is said from theoretical angle, the quality of the output of prediction result, be unable to do without selection and the later stage prediction result of input information
Continuous amendment.Therefore the advantages of comprehensive optimal subset method and SVM regression prediction methods, is to use before having considered prediction
Optimization of the optimal subset method to list entries, while prediction result is corrected using the SVM regression prediction method later stages.
3) dynamic inputted takes continuous sequence of data segments as input sequence of subsets to replace original history number
It is realized by time response according to sequence, the dynamic input prediction model for waiting dimensions to fill vacancies in the proper order.
Further, from largely coming from, domestic and international search platform, (such as Baidu refers to social platform gathered data search data
Number, Google trend, 360 indexes and micro- index), it is better than traditional statistical data.
【Description of the drawings】
Fig. 1 is a SVR tourism demand prediction technique frame diagram optimized based on optimal subset;
Fig. 2 is SVR regression forecasting flow charts;
【Specific implementation mode】
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
His embodiment, shall fall within the protection scope of the present invention.
Embodiment 1
By taking Xi'an museum day passenger flow data as an example, a kind of SVR tourism demand prediction techniques based on optimal subset optimization
It is as follows:
As shown in Figure 1:The practical application that domestic tourism demand is predicted in conjunction with the embodiments --- by taking the museum of Xi'an as an example, base
It is specifically realized by following steps in the SVR tourism demands prediction technique of optimal subset optimization:
(1) acquisition tourist attractions scenic spot reception number data hereinafter referred to as basic data, searches for data index containing Baidu
(http://index.baidu.com/), Google trend (https://www.google.co.jp/trends/), 360 refer to
Number (http://index.so.com/), micro- exponent data (http://data.weibo.com/index) four kinds, temperature record
It is collected in weather net (www.tianqi.com), including day lowest temperature, the day highest temperature and the temperature difference.As shown in table 1:
1 Xi'an museum passenger flow sample data of table
(2) relationship between the sight spot passenger flow data and 7 indexs of passenger flow of Gray Association Analysis analysis of history, tool are chosen
7 index of body is:Baidu's index, Google Trends, 360 indexes, micro- index, the day highest temperature, day lowest temperature, temperature difference per day.Ash is obtained to close
Connection degree ρ 0i (m);
1. data information and pretreatment
For the Accurate Prediction Xi'an museum daily volume of the flow of passengers in scenic spot, including 2013/1/1-2015/12/31 amounts to 1095
Item records.Specific targets include 7:Baidu's index, Google Trends, 360 indexes, micro- index, the day highest temperature, day lowest temperature, day
The temperature difference.Following supplementary explanation is carried out for daily passenger flow and 7 indexs:Day volume of the flow of passengers data, due to 4A grades of scenic spots Xi'an natural science
Institute closes shop whenever Tuesday, amounts to 52 weeks within 2013, rejects null value on every Tuesdays.For special holidays, such as Spring Festival in 2013 it
Adjacent Tuesday does not close shop afterwards, and similar data amount includes New Year's Day 2013/1/1, the Spring Festival movable (junior one to 15), 2013/4/
30,2013/6/11 Dragon Boat Festival, 2013/10/1 National Day seven-day holiday, New Year's Eve, i.e., reject within 2013 data 48,2014 years 50
Item, 2015 50, remaining 946, micro- index Start Date is 2013/3/1, in order to be consistent with other indexs, passenger flow,
52 records before deleting 2013/3/1,360 indexes are 0 in 2013/3/1 and 20,13/,3/2 two day index, therefore are met
The record of standard amounts to 892.
2. establishing the initial matrix x of 7 indexs of correlationi, and initialized, obtain transformation matrix xi'
xi=(xi(1),xi(2),…xi(m),…) (1)
Wherein, xi(k) original value of the i indexs at the m days is indicated.
3. difference sequence Δ0i(m) it calculates as follows
Δ0i(m)=| xi 0(m)-xi′(m)|,Δ0i(m)=(Δ0i(1),Δ0i(2),…Δ0i(m),…) (3)
4. obtaining incidence coefficient ε0i(m), then gray relation grades ρ0i(m) calculation formula such as formula (5)
In formula (1), δ is defined as resolution ratio, δ ∈ (0,1), and general value is 0.5, in order to it is aobvious to enhance difference
Work property.The degree of association of 7 influence indexs is as shown in table 2.
The gray relation grades of 27 influence indexs of table
(3) regression fit modeling is carried out to passenger flow data using optimal training subset method and the SVM regression model established,
The best penalty parameter c and kernel functional parameter g that cross validation selection returns are obtained, trains SVM fittings pre- according to the optimal parameter
It surveys, obtains fitting predicted value and prediction index Mse and R, as shown in Figure 2.
The above is the preferred embodiment of the present invention, passes through above description content, the related work of the art
Personnel can carry out various improvement and replacement under the premise of without departing from the technology of the present invention principle, these improve and replace
It should be regarded as protection scope of the present invention.
Claims (9)
1. a kind of SVR tourism demand prediction techniques based on optimal subset optimization, it is characterised in that include the following steps:
(1) according to chronological classification acquisition tourist attractions scenic spot reception number data, search data and temperature record, the search
Data include reflecting that the exponential quantity of volumes of searches, n are positive integer in n data analysis service respectively;The temperature record includes
Day lowest temperature, the day highest temperature and the temperature difference;
(2) pass between the sight spot reception number data and (n+3) a index of passenger flow of Gray Association Analysis analysis of history is used
System obtains gray relation grades ρ0i(m);(n+3) a index includes n exponential quantity, the day highest temperature, day lowest temperature and temperature difference per day;
(3) regression fit is carried out to passenger flow data using optimal training subset method and the SVM regression model established, is intersected
The penalty parameter c and kernel functional parameter g that verification selection returns train SVM fitting predictions according to the optimal parameter, it is pre- to obtain fitting
Measured value Mse and prediction index R.
2. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
Use in step (2) method of the degree of association of Gray Association Analysis acquisition (n+3) a index for:
(2.1) the initial matrix x of (n+3) a index of correlation is establishedi, and carry out initialization and obtain transformation matrix xi'
xi=(xi(1),xi(2),…xi(m),…) (1)
Wherein, xi(m) original value of the i indexs at the m days is indicated;
(2.2) difference sequence Δ0i(m) it calculates as follows
(2.3) incidence coefficient ε is obtained0i(m), then gray relation grades ρ0i(m) calculation formula such as formula (5)
In formula (1), δ is defined as resolution ratio, δ ∈ (0,1).
3. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as claimed in claim 2, it is characterised in that:δ
Value is 0.5.
4. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
The regression model of SVM foundation is to the method for passenger flow data progress regression fit in step (3):
(3.1) data prediction;Training set and test set are normalized, the normalized of use is as follows:
In formula, x, y ∈ Rn,xmin=min (x), xmax=max (x), normalized effect are that initial data is regular to [0,1] model
In enclosing;
(3.2) it uses cross-validation method to obtain penalty parameter c and kernel functional parameter g, allows c and g values in a certain range, it is right
The training verification under this group is obtained using training set as raw data set in determining c and g, it is final so that training accuracy rate highest
That group of c and g as optimal parameter;
(3.3) training and prediction;Using the optimal parameter obtained in step (3.2), SVM is trained and regression forecasting, most
Mean square error Mse and coefficient R are obtained eventually.
5. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
Optimal training subset method is in the step (3):
(3-1) intercepts continuous sequence of data segments as input sequence of subsets to replace original historical data sequence;
Build dynamic majorized subset's gray model:For the first time, the x in tourist attraction history reception number original series is removed(0)
(1), x is filled vacancies in the proper order(0)(n+1), one group of new dynamic sequence (x is constituted(0)(2),x(0)(3),…x(0)(n+1))
And so on, it keeps sequence length constant, is built into dynamic majorized subset's gray model;
(3-2) determines the number of optimal input subset;
Intercept continuous data segment (x(0)(n-l),…x(0)(n-1),x(0)(n)), l ∈ (4,5 ... n-1) are as input subset sequence
It arranges to replace original history reception number sequence (x(0)(1),x(0)(2),x(0)(3)…x(0)(n)) it, is obtained according to step (3)
Mean square error Mse minimum value and coefficient R maximum value, determine it is optimal input subset number l;
(3-3) chooses l optimal input subsets and carries out the final prediction result of prediction acquisition.
6. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
Temperature record data are collected in weather net.
7. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
It is that metering is classified according to year, the moon, week or day with number of days when according to chronological classification.
8. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:n
Reflect that the exponential quantity of volumes of searches is in Baidu's index, Google trend, 360 indexes and micro- index in a data analysis service respectively
At least one.
9. a kind of SVR tourism demand prediction techniques based on optimal subset optimization as described in claim 1, it is characterised in that:
In step (3), regression fit is carried out to passenger flow data using optimal training subset method and the SVM regression model established, is obtained
When the optimal parameter c and g that cross validation selection returns;
Using training set as raw data set, training under each group of penalty parameter c and kernel functional parameter g is obtained using K-CV methods
The accuracy rate for collecting fitting finally takes so that training set verifies the highest parameter c and g of accuracy rate fitting effect as optimal parameter;
If multigroup parameter c and g corresponds to the equal highest of effect, the first group of parameter c and g searched is chosen as best parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810092083.XA CN108376260A (en) | 2018-01-30 | 2018-01-30 | A kind of SVR tourism demand prediction techniques based on optimal subset optimization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810092083.XA CN108376260A (en) | 2018-01-30 | 2018-01-30 | A kind of SVR tourism demand prediction techniques based on optimal subset optimization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108376260A true CN108376260A (en) | 2018-08-07 |
Family
ID=63017056
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810092083.XA Pending CN108376260A (en) | 2018-01-30 | 2018-01-30 | A kind of SVR tourism demand prediction techniques based on optimal subset optimization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108376260A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110163408A (en) * | 2019-04-03 | 2019-08-23 | 浙江工业大学 | A kind of travelling stimulation strategy efficiency analysis method based on Source market attributive character |
CN111126736A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术(北京)有限公司 | Enterprise passenger flow determining method and device, server and storage medium |
CN114091768A (en) * | 2021-11-25 | 2022-02-25 | 安徽大学 | STL (Standard template library) and LSTM (local Scale TM) with attention mechanism based tourism demand prediction method |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104951845A (en) * | 2015-05-29 | 2015-09-30 | 华南理工大学 | Tourist attraction saturation pre-warning method with function of passenger flow volume fitting on basis of network attention |
CN105404942A (en) * | 2015-12-07 | 2016-03-16 | 山东科技大学 | Optimal weight combination prediction model of tourism data |
CN106779247A (en) * | 2017-01-11 | 2017-05-31 | 陕西师范大学 | The Forecasting Methodology of the tourism demand based on Information Entropy Combinatorial Optimization |
CN106960250A (en) * | 2017-03-03 | 2017-07-18 | 陕西师范大学 | A kind of method of dynamic prediction tourist attractions passenger flow |
CN107145962A (en) * | 2017-04-01 | 2017-09-08 | 上海诺悦智能科技有限公司 | A kind of sight spot domestic visitors forecasting system |
CN107316103A (en) * | 2017-06-04 | 2017-11-03 | 东南大学 | Sight spot dynamic mass Forecasting Methodology based on echo state network |
-
2018
- 2018-01-30 CN CN201810092083.XA patent/CN108376260A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104951845A (en) * | 2015-05-29 | 2015-09-30 | 华南理工大学 | Tourist attraction saturation pre-warning method with function of passenger flow volume fitting on basis of network attention |
CN105404942A (en) * | 2015-12-07 | 2016-03-16 | 山东科技大学 | Optimal weight combination prediction model of tourism data |
CN106779247A (en) * | 2017-01-11 | 2017-05-31 | 陕西师范大学 | The Forecasting Methodology of the tourism demand based on Information Entropy Combinatorial Optimization |
CN106960250A (en) * | 2017-03-03 | 2017-07-18 | 陕西师范大学 | A kind of method of dynamic prediction tourist attractions passenger flow |
CN107145962A (en) * | 2017-04-01 | 2017-09-08 | 上海诺悦智能科技有限公司 | A kind of sight spot domestic visitors forecasting system |
CN107316103A (en) * | 2017-06-04 | 2017-11-03 | 东南大学 | Sight spot dynamic mass Forecasting Methodology based on echo state network |
Non-Patent Citations (3)
Title |
---|
HONG WEI-CHIANG等: "《SVR with hybrid chaotic genetic algorithms for tourism demand forecasting》", 《APPLIED SOFT COMPUTING》 * |
李瑶 等: "《基于改进的灰色模型的旅游需求预测研究》", 《计算机科学》 * |
王小川 等: "《MATLAB神经网络43个案例分析》", 31 August 2013 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111126736A (en) * | 2018-11-01 | 2020-05-08 | 百度在线网络技术(北京)有限公司 | Enterprise passenger flow determining method and device, server and storage medium |
CN111126736B (en) * | 2018-11-01 | 2023-09-01 | 百度在线网络技术(北京)有限公司 | Enterprise passenger flow volume determining method and device, server and storage medium |
CN110163408A (en) * | 2019-04-03 | 2019-08-23 | 浙江工业大学 | A kind of travelling stimulation strategy efficiency analysis method based on Source market attributive character |
CN110163408B (en) * | 2019-04-03 | 2021-10-29 | 浙江工业大学 | Travel stimulation strategy effectiveness analysis method based on passenger source and place attribute characteristics |
CN114091768A (en) * | 2021-11-25 | 2022-02-25 | 安徽大学 | STL (Standard template library) and LSTM (local Scale TM) with attention mechanism based tourism demand prediction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103577413B (en) | Search result ordering method and system, search results ranking optimization method and system | |
CN103577888B (en) | Method for optimization of product design schemes | |
WO2017166370A1 (en) | Method for delineating metropolitan area based on regional inter-city flow intensity measuring model | |
CN109783738A (en) | A kind of double extreme learning machine mixing collaborative filtering recommending methods based on more similarities | |
CN108376260A (en) | A kind of SVR tourism demand prediction techniques based on optimal subset optimization | |
CN102194006B (en) | Search system and method capable of gathering personalized features of group | |
CN104881472A (en) | Combined recommendation method of traveling scenic spots based on network data collection | |
CN110135587A (en) | It is hesitated based on section and obscures the Multiple Attribute Group Decision of more granularity decision rough sets | |
CN105447633A (en) | Scientific research institution integration evaluation method and system thereof | |
CN107609938B (en) | Service recommendation method based on user qualitative and quantitative preference | |
CN108415928A (en) | A kind of book recommendation method and system based on weighted blend k- nearest neighbor algorithms | |
CN115344767A (en) | Supplier evaluation method based on network data | |
Zheng | Research on the competitiveness of China’s leisure sports industry based on statistical method | |
CN114743010A (en) | Ultrahigh voltage power transmission line point cloud data semantic segmentation method based on deep learning | |
CN102982035A (en) | Search ordering method and search ordering system for community users | |
CN104778205B (en) | A kind of mobile application sequence and clustering method based on Heterogeneous Information network | |
JP2018534674A (en) | A global network system for creating real-time global company rankings based on globally acquired data | |
Yang et al. | Optimization of tourism information analysis system based on big data algorithm | |
Li et al. | Short and long term tourism demand forecasting based on Baidu search engine data | |
CN112488438A (en) | Industrial enterprise comprehensive benefit evaluation method and system combining credit information | |
Hu | Research on Dynamic Mechanism and Management Mode of Sports Tourism Development in China | |
Li et al. | Research on Big Data System Based on Cultural Tourism in Dongguan | |
Lv | Analysis and optimization strategy of travel hotel website reservation behavior based on collaborative filtering | |
CN111191826A (en) | Load prediction method based on cosine similarity classification | |
Zheng et al. | Network Monitoring and Early Warning System of Economic Benefit Decline in Ecological, Agricultural and Pastoral Areas. |
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: 20180807 |
|
RJ01 | Rejection of invention patent application after publication |