CN109857983A - A kind of food and drink venue temperature analysis method towards super-large city - Google Patents
A kind of food and drink venue temperature analysis method towards super-large city Download PDFInfo
- Publication number
- CN109857983A CN109857983A CN201811631005.9A CN201811631005A CN109857983A CN 109857983 A CN109857983 A CN 109857983A CN 201811631005 A CN201811631005 A CN 201811631005A CN 109857983 A CN109857983 A CN 109857983A
- Authority
- CN
- China
- Prior art keywords
- data
- drink
- food
- factor
- super
- 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.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The food and drink venue temperature analysis method towards super-large city that the invention proposes a kind of, its step are as follows: selection target super-large city；Select data source；All sample datas of original analysis select primary categories factor in plurality of classes factor；Data detection and data transformation are carried out to the initial data of primary categories factor；Linear regression model (LRM) is established, linear regression model (LRM) is optimized, the regression model after obtaining tuning；The inverse transformation converted by data restores the regression model after tuning, obtains complementary quantitative relationship between the food and drink venue temperature and multiple types factor of target super-large city.The present invention, which demonstrates the comprehensive experience in dining room and population effect, influences maximum to the temperature in dining room, there is good reference significance for the addressing of super-large city Chinese Restaurant, operation mode, consumption upgrading etc., it can offer reference for urban area development plan and the optimization of food and drink pattern, to promote the benign sustainable development of super-large city catering industry.
Description
Technical field
The present invention relates to the technical field of restaurant temperature analysis more particularly to a kind of food and drink venue heat towards super-large city
Spend analysis method.
Background technique
With the fast transferring and economic advantages enrichment effect of population, in global range, the formation trend of super-large city is cured
Add obvious.With lasting economic growth and population migration, global super-large city quantity is gradually increasing, and influence is also related to
To the various aspects in city.Since food and drink consumption is the mankind's one of consumer behavior the most basic, population size in super-large city,
The elements such as Complex Flows, traffic condition, diversified consuming capacity will affect consumption mode and urban development.
Researchers are directed in development and a series of exploration had been unfolded in the catering industry of developed countries and regions.Ip
Et al. analyze Informatization Development situation of the Macao in recent years about catering trade competition, it is indicated that geospatial information, social data
And the importance of data mining.Mattera et al. pays close attention to the sme development situation of the country, Spain, studies which factor
Can power-assisted data mining, so that catering companies be helped comparatively fast to grow up.Comment is considered as consumer evaluation, differentiates dining room superiority and inferiority
A kind of important way, Zhai etc. utilizes online on-line evaluation data, including comment length, the comment time and comment it is readable come
The influence that research comment consumes food and drink.Meanwhile traffic factor seems also have an impact city retail business Price Impact,
In large size city, convenient traffic, such as subway and public bus network are likely to create new food and drink consumption business corridor.
Increasingly pay attention to today of cooking culture in people, consumer is ready to wait sometimes to go to taste a meal as long as a few hours
Do are cuisines in this case which kind of factor has caused so earnestly pursuit on earth? in other words, why some dining rooms by
It is more and more to welcome? the prior art does not provide specific analysis method also.
Summary of the invention
The technical issues of failing to disclose how the influence factor of analysis food and drink venue temperature for the prior art, the present invention mentions
A kind of food and drink venue temperature analysis method towards super-large city out, obtains and calculates using public comment net and Baidu map
To dining room consumption data sample, linear regression model (LRM) is established based on statistical method, analysis obtains what dining room consumption temperature influenced
Principal element.
In order to achieve the above object, the technical scheme of the present invention is realized as follows: a kind of food and drink towards super-large city
Venue temperature analysis method, its step are as follows:
Step 1: selection target super-large city: super-large city has the huge size of population, the economy of higher degree and text
Change horizontal and perfect work and life auxiliary facility；
Step 2: selection data source: potential impact food and drink venue temperature is a variety of in original definition target super-large city
Category factor selects suitable data source according to plurality of classes factor, obtains with plurality of classes factor corresponding sample number
According to；
Step 3: all sample datas of initial analysis select primary categories factor in plurality of classes factor, and select energy
The variable of dining room temperature is enough characterized as dependent variable, using the plurality of classes factor of potential impact temperature as independent variable；
Step 4: carrying out data detection to the initial data of step 3 treated primary categories factor and data convert,
Make transformed data close to stringent normal distribution；
Step 5: using step 4, treated that data establish linear regression model (LRM), optimizes to linear regression model (LRM),
Regression model after obtaining tuning；
Step 6: the inverse transformation converted by data in step 4 restores the regression model after tuning, obtains independent variable
With the relationship of dependent variable, complementary quantitative pass between the food and drink venue temperature and multiple types factor of target super-large city is obtained
System.
Dining room of the sample data on internet on disclosed food and drink consumption website, the correlation in dining room to be analyzed
Data are complete, and dining room to be analyzed is located at the typical position of super-large city.For example, the cuisines plate of the online Beijing area of public comment
Always there are the data in the governed dining room of data upper 2016 year, the position in dining room is 2 kilometers of models along two more bustling rings of Beijing
In enclosing.
Plurality of classes factor, which fully considers, in the step 2 may influence all kinds of of food and drink venue management in super-large city
Element, plurality of classes factor include dining room quality, eating surroundings, traffic condition, periphery consumption facility and or population group effect
It answers, the information in geographical location therein, public transport and landmark derives from Baidu map；The acquisition of the sample data
Method includes the method that network crawls, discloses api interface or third-party charging.
The primary categories factor of the sample data includes general comment number α, comprehensive star β, pre-capita consumption υ, nearest subway
Stop spacing is from τs, neighbouring public bus network number τb, periphery large sized commercial center number πcAnd large-scale residential quarter and university number πr,
Wherein, nearest subway stop spacing is from τsWith neighbouring public bus network number τbIt is category factor relevant to public transport convenience, it is comprehensive
Star β is the category factor of the comprehensive sexual experience in dining room, the number π of periphery large sized commercial centerc, large-scale residential quarter and university's number
πrAnd pre-capita consumption υ is category factor relevant to group's building-up effect；The general comment number α is dependent variable, comprehensive star
β, pre-capita consumption υ, nearest subway stop spacing are from τs, neighbouring public bus network number τb, periphery large sized commercial center number πcAnd it is large-scale
Residential quarter and university number πrFor independent variable.
The data detection utilizes the realization of Shapiro-Wilk Methods of Normality Test, Shapiro-Wilk test of normality
Method sample data carry out homogeneity test of variance, under 95% confidence level, inspection result if more than 0.05 sample data it
Between be not present notable difference, so that whether test samples data meet normal distribution.
The data transformation is handled using Box-Cox transformation, so that data are divided closer to stringent normal state after transformation
Cloth；The transformation rule of the Box-Cox transformation are as follows:
Wherein, y indicates that the raw value before Box-Cox transformation, y (λ) indicate raw value through the transformed number of Box-Cox
Value, the conversion parameter undetermined of λ representative sample data, so that dependent variable y (λ) meets:
Wherein, X indicates independent variable vector, β1For parameter vector, X and β1Be parameter to be estimated, ε indicate random error to
The residual error of amount, σ are error to standard deviation, InFor the unit matrix of n*n.
The linear regression model (LRM) is multiple linear regression model or generalized linear regression model, utilizes multiple linear regression
The regression model that model obtains are as follows: ylam=β0+β1x1+β2x2+β3x3+β4x4+β5x5+β6x6；
Wherein, ylam indicates dining room pouplarity, the general comment in sample data converted based on Box-Cox
Number α；x1It represents consumer and has merged the synthesis star that taste, environment and service are given a mark as dining room, in sample data
Comprehensive star β；x2Pre-capita consumption υ for the average consumption amount of money in dining room, in sample；x3For the dining room and nearest subway
The distance stood, the nearest subway stop spacing in sample data are from τs；x4For public bus network number, source within the scope of the 1km of dining room
Public bus network number τ near in sampleb；x5It indicates the number for the Large Residential District and university that 1km range in dining room is adjoined, come
Derived from the medium-and-large-sized residential quarter of sample data and university number πr；x6For the number at city-level commerce services center, from sample data
The number π of middle periphery large sized commercial centerc；Meanwhile (β0,β1,β2,β3,β4,β5,β6) it is unknown parameter.
The method that linear regression model (LRM) is optimized in the step 5 are as follows: 1. utilize Spearman or the side Pearson
Method judge each independent variable whether with dependent variable there are significant correlativities；2. if independent variable is with dependent variable, there are syntenies
Problem then reduces the synteny of independent variable and dependent variable using multiple ridge regression method；3. using method of gradual regression to initial line
Property regression model in the values of parameters calculated and analyzed, the degree of strength for influencing food and drink venue temperature is sentenced
It is disconnected, it will affect faint corresponding independent variable and rejected；4. obtaining final analysis model.
Beneficial effects of the present invention: choosing the food and drink place along 2 ring of Beijing in 1 kilometer range is goal in research, first
Sample is carried out based on public comment net and Baidu map to crawl and calculate, and by analyzing the temperature characteristic in dining room, establishes influence
The Factor system of dining room temperature, including comprehensive score, average price, away from the nearest subway station shortest distance, 1 kilometer city-level business
Multiple elements such as service centre's number；The multiple regression point for influencing dining room temperature is established by significance test, Box-Cox transformation
Model is analysed, and utilizes method of gradual regression Optimized model；Operational analysis is carried out according to this 200 dining room effective samples.The result shows that:
(1) person sponging on an aristocrat values the comprehensive impression including taste, environment and service of dining room offer very much, these belong to catering trade
In core the most element；(2) population effect has importance for catering industry, also illustrates dining room addressing from another angle
Place has positive influence to its temperature；(3) in super-large city public transport, the dining factors such as level of consumption for dining room
Temperature does not have direct influence.The present invention demonstrates within the scope of super-large city central area, the comprehensive experience in dining room and group
Bulk effect influences the temperature in dining room maximum, it is proposed that downtown integrated planning will fully consider the composition of food and drink consuming capacity
And the promotion of operation management ability.The present invention has the addressing of super-large city Chinese Restaurant, operation mode, consumption upgrading etc.
Good reference significance can offer reference for urban area development plan and the optimization of food and drink pattern, to promote super-large city
The benign sustainable development of catering industry.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is flow chart of the invention.
Fig. 2 is the schematic diagram of the derived region of sample of the present invention data.
Fig. 3 is the scatter diagram of independent variable of the present invention.
Fig. 4 is the residual value histogram of the regression model after present invention optimization.
Specific embodiment
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 description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under that premise of not paying creative labor
Embodiment shall fall within the protection scope of the present invention.
As shown in Figure 1, a kind of food and drink venue temperature analysis method towards super-large city, its step are as follows:
Step 1: selection target super-large city: super-large city has the huge size of population, the economy of higher degree and text
Change horizontal and perfect work and life auxiliary facility.
In order to analyze the food and drink consumer behavior in super-large city, especially obtain about influence dining room pouplarity because
Element needs that all related datas are defined and are described.China covers north there are four actual super-large cities
Capital, Shanghai, Guangzhou and Shenzhen, wherein Beijing is incontrovertible has comprehensive leading superiority, either politics, economy still
Cultural layer.According to the data of Beijing statistics bureau, by 2017, for Beijing permanent resident population close to 21,700,000, the amount of consumption reached 912
Hundred million yuan (wherein only having counted the enterprise that annual turnover is more than 2,000,000), rate of increase reaches 9.6% year by year.It is total due to having
Huge and rich and varied food and drink current condition of consumption is measured, the present invention is using Beijing as ideal research object.
Step 2: selection data source: potential impact food and drink venue temperature is a variety of in original definition target super-large city
Category factor selects suitable data source according to plurality of classes factor, obtains with plurality of classes factor corresponding sample number
According to.
There are data can within the sample data 2016 on the cuisines plate of the online Beijing area of public comment always
The data in the dining room followed, the position in dining room is in 2 kilometer range along two more bustling rings of Beijing.
With internet in the continuous promotion of social every field importance, the food and drink consumer behavior of consumer is gradually
Be embodied on internet, relevant evaluation has become very valuable data analysis foundation.Based on this, the present invention will
Current more mainstream, influential consumer website -- public comment net (http://www.dianping.com/) conduct
Goal in research obtains relevant food and drink consumption data.
The super-large city mature as one, business activity the most active and perfect infrastructure all collect in Beijing
In in inner city, therefore along two rings for selecting Beijing more bustling herein, 2 kilometer ranges are as survey region, such as Fig. 2 institute
Show, to collect food and drink relevant information, it is contemplated that the diversification of China's type of diet includes meal as much as possible in sample data
Eat type.By the cuisines plate of public comment Beijing area, gets and cover a variety of food and drink types, amount to 200 dining rooms
Consumption information.It may be noted that having been selected straight since in January, 2016 since part businessman is too short in website on-line time
There is the relevant information in governed 200 restaurants of data always in December, 2016.In addition, about geographical location, public transport,
The information such as mark building derive from Baidu map, and are calculated.
Step 3: all sample datas of initial analysis select primary categories factor in plurality of classes factor, and select energy
The variable of dining room temperature is enough characterized as dependent variable, using the plurality of classes factor of potential impact temperature as independent variable.
Plurality of classes factor, which fully considers, in the step 2 may influence all kinds of of food and drink venue management in super-large city
Element, plurality of classes factor include dining room quality, eating surroundings, traffic condition, periphery consumption facility and or population group effect
It answers, the information in geographical location therein, public transport and landmark derives from Baidu map；The method of the original analysis
It is crawled including network, the method for open api interface or third-party charging.
The primary categories factor of the sample data includes general comment number α, comprehensive star β, pre-capita consumption υ, nearest subway
Stop spacing is from τs, neighbouring public bus network number τb, periphery large sized commercial center number πcAnd large-scale residential quarter and university number πr,
Nearest subway stop spacing is from τsWith neighbouring public bus network number τbIt is category factor relevant to public transport convenience, comprehensive star β
It is the category factor of the comprehensive sexual experience in dining room, the number π of periphery large sized commercial centerc, large-scale residential quarter and university number πrAnd
Pre-capita consumption υ is category factor relevant to group's building-up effect；As shown in table 1, all sample numbers for crawling and being calculated
According to including 7 classifications, has with the analysis of the temperature in dining room and directly contact.It may be noted that number of reviews α can be more appropriate
Reflect the concerned degree in a dining room.It is undeniable to be, even if one seems unremarkable coffee chop house, it is also possible to attract
Numerous beans vermicelli, regardless of hierarchy of consumption.Therefore, general comment number α is dependent variable, comprehensive star β, pre-capita consumption υ, nearest subway
Stop spacing is from τs, neighbouring public bus network number τb, periphery large sized commercial center number πcAnd large-scale residential quarter and university number πrFor
Independent variable.
The classification of 1. consumption information of table
Step 4: carrying out data detection to the initial data of step 3 treated primary categories factor and data convert,
Make transformed data close to stringent normal distribution.
The data detection utilizes the realization of Shapiro-Wilk Methods of Normality Test, Shapiro-Wilk test of normality
Method is to carry out homogeneity test of variance to sample data, and under 95% confidence level, inspection result is if more than 0.05 sample number
Notable difference is not present between, so that whether test samples data meet normal distribution.By Shapiro-Wilk normality
It examines, it is found that the dependent variable data in sample are right avertence state, in order to meet normal distribution requirement, converted using Box-Cox to it
It is operated, so that data are closer to stringent normal distribution after transformation, thus the foundation as subsequent analysis.
The data transformation is handled using Box-Cox transformation, so that data are divided closer to stringent normal state after transformation
Cloth；The transformation rule of the Box-Cox transformation are as follows:
Wherein, y (λ) indicates the dependent variable after conversion, and y indicates the raw value of Box-Cox transformation antecedents,
In, y indicates that the raw value before Box-Cox transformation, y (λ) indicate raw value through the transformed numerical value of Box-Cox, λ representative sample
The conversion parameter undetermined of notebook data, so that dependent variable y (λ) meets:
Wherein, X indicates independent variable vector, β1For parameter vector, X and β1Be parameter to be estimated, ε indicate random error to
The residual error of amount, σ are error to standard deviation, InFor the unit matrix of n*n.
For the sample space containing 200 sample datas, conversion undetermined is calculated using maximum likelihood estimate (MLE) and is joined
The value of number λ, the results are shown in Table 2, it is known that the more excellent value range of conversion parameter λ undetermined is optimal between [0.0855,0.2597]
Estimated value is 0.17.Succinct in order to make to calculate, the present invention takes 0.2 to be used as λ value.
2. dependent variable Box-Cox transformation coefficient of table
Step 5: using step 4, treated that data establish linear regression model (LRM), optimizes to linear regression model (LRM),
Regression model after obtaining tuning.
The linear regression model (LRM) is multiple linear regression model or generalized linear regression model.By tentative calculation, using more
First linear regression model (LRM) can preferably explain influence of all kinds of factors to food and drink consumption temperature.Utilize multiple linear regression model
Obtained regression model are as follows: ylam=β0+β1x1+β2x2+β3x3+β4x4+β5x5+β6x6,
Wherein, ylam indicates dining room pouplarity, the general comment in sample data converted based on Box-Cox
Number α；x1It represents consumer and has merged the synthesis star that taste, environment and service are given a mark as dining room, in sample data
Comprehensive star β；x2Pre-capita consumption υ for the average consumption amount of money in dining room, in sample；x3For the dining room and nearest subway
The distance stood, the nearest subway stop spacing in sample data are from τs；x4For public bus network number, source within the scope of the 1km of dining room
Public bus network number τ near in sampleb；x5It indicates the number for the Large Residential District and university that 1km range in dining room is adjoined, come
Derived from the medium-and-large-sized residential quarter of sample data and university number πr；x6For the number at city-level commerce services center, from sample data
The number π of middle periphery large sized commercial centerc；Meanwhile (β0,β1,β2,β3,β4,β5,β6) it is unknown parameter.
Then implement linear fit and analysis, obtained each independent variable coefficient.As shown in table 3, which obtains
The coefficient of determination be 0.2171, p value 8.538e-10.The coefficient of determination and p value are by calling linear regression method to obtain
, the coefficient of determination for estimate regression equation fitting in ylam to independent variable x1-6Covariance Plots effect amount number；P value is body
The standard of existing significance test, be generally with p < 0.05 it is significant, p < 0.01 is highly significant, therefore this example highly significant.
3. initial linear fitting coefficient of table
The method that linear regression model (LRM) is optimized in the step 5 are as follows: 1. utilize Spearman or the side Pearson
Method judge each independent variable whether with dependent variable there are significant correlativities；2. if independent variable is with dependent variable, there are syntenies
Problem then reduces the synteny of independent variable and dependent variable using multiple ridge regression method；3. using method of gradual regression to initial line
Property regression model in the values of parameters calculated and analyzed, the degree of strength for influencing food and drink venue temperature is sentenced
It is disconnected, it will affect faint corresponding independent variable and rejected；4. obtaining final analysis model.
In order to judge that present invention introduces variance inflation factor (VIF) with the presence or absence of Problems of Multiple Synteny between independent variable
Detected, call the obtained data of VIF method in R language as shown in table 4, as the result is shown all dependent variables be all satisfied 0 <
VIF<10.In addition, further implementing to judge by the scatter plot between constructed variable, parameter is that above-mentioned a few class independents variable are joined
Number calls Plot method as shown in figure 3, there is no Problems of Multiple Synteny to show the linear regression model (LRM).
Table 4.VIF testing result
Subsequent to reaching more preferably fitting degree, independent variable x is carried out1-x6With the correlation analysis of dependent variable ylam, by table 5
It is found that independent variable x3And x4Do not have correlativity between dependent variable ylam, because the p value of the two is noticeably greater than 0.05, i.e.,
In the case where confidence level is 95%, the original hypothesis with correlativity can not be set up.
The analysis of 5. correlation between variables of table
Based on this, initial linear regression model (LRM) is carried out using method of gradual regression (Stepwise regression) excellent
Change, wherein eliminating uncorrelated independent variable from model using gradient (Bidirectional Elimination) method is compared.Through
It crosses and compares, obtain ylam~x1+x5For best model, in such cases, it is fitted that the results are shown in Table 6 to it, wherein can
Certainly coefficient is 0.2248, and p value 4.722e-12 is superior to initial linear regression model.
Independent variable coefficient after table 6. optimizes
As a result, by the independent variable parameter after optimization, the present invention obtains the regression model after tuning, as shown in formula (4),
Its residual error meets the normal distribution that average value is 0.00, standard deviation is 3.7744, and residual error histogram is as shown in Figure 4.
Ylam=-3.64965+3.38566x1+0.30572x5 (4)
Step 6: the inverse transformation converted by data in step 4 restores the regression model after tuning, obtains independent variable
With the relationship of dependent variable, complementary quantitative pass between the food and drink venue temperature and multiple types factor of target super-large city is obtained
System.
Due to having carried out Box-Cox transformation before returning, formula (4) cannot intuitively show commercial residential buildings project sale price
Primitive relation between lattice and each factor, thus according to Box-Cox mapping mode by formula (4) restore, obtain independent variable with
The correlativity of dependent variable is as follows:
Y=(0.27+0.677x1+0.0061x5)5 (5)
Shown in linear fit model such as formula (5) after Box-Cox transformation reduction, it follows that influencing food and drink comment
In every factor of number, consumer is the synthesis star x that dining room provides1Corresponding absolute coefficient is maximum, shows some meal
The comprehensive enjoyment that the Room assigns a person sponging on an aristocrat is mostly important, it has merged taste, environment and service, can influence the dining room most significantly
Temperature.Secondly, the element being affected is the number of large-scale residential quarter and university, in fact, this is also dining room position attribution
One reflection.Element explanation, for the importance of catering industry, more inlet flow rates are likely to bring more population effect
A person sponging on an aristocrat consume.
Expection imagination with before is runed counter to, public transport factor x3And x4Seem to consume dining room comment sum without shadow
Ring, analyzed from result, be that usually there is the fixed operation period due to public transport, it is contemplated that time for eating meals it is uncertain because
Element, consumer may not necessarily be had dinner by the way of public transport, therefore the superiority and inferiority of public transport condition does not have dining room temperature
It significantly affects.Meanwhile pre-capita consumption situation be also it is unexpected, substantially dependent variable is not had an impact, it is contemplated that Beijing compared with
The high level of consumption and food resources extremely abundant, a persons sponging on an aristocrat many times can more focus on the quality of food and drink, rather than valence
Lattice.
Beijing is as well-known and typical super-large city, the hotel industry development high mature of inner city, present invention benefit
Dining room consumption data sample is obtained and be calculated with public comment net and Baidu map, and linear return is established based on statistical method
Return model, the dining room consumption temperature influence factor along 2 ring of Beijing is analyzed, is concluded that (1) analyzes result table
A bright person sponging on an aristocrat values the comprehensive impression including taste, environment and service of dining room offer very much, these belong to catering trade
In core the most element；(2) analysis result illustrates population effect for the importance of catering industry, also from another angle
Illustrate that dining room addressing place has positive influence to its temperature；(3) opposite, analysis is the results show that public in super-large city
The factors such as traffic, dining level of consumption do not have direct influence for dining room temperature altogether.To sum up, the present invention is for super-large city
Addressing, operation mode, consumption upgrading of Chinese Restaurant etc. have good reference significance, can be urban area development plan and meal
Drink pattern optimization is offered reference, to promote the benign sustainable development of super-large city catering industry.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (8)
1. a kind of food and drink venue temperature analysis method towards super-large city, which is characterized in that its step are as follows:
Step 1: selection target super-large city: super-large city has the huge size of population, the economy of higher degree and educational level
Flat and perfect work and life auxiliary facility；
Step 2: selection data source: the plurality of classes of potential impact food and drink venue temperature in original definition target super-large city
Factor selects suitable data source according to plurality of classes factor, obtains with plurality of classes factor corresponding sample data；
Step 3: all sample datas of initial analysis select primary categories factor in plurality of classes factor, and selecting being capable of table
The variable of dining room temperature is levied as dependent variable, using the plurality of classes factor of potential impact temperature as independent variable；
Step 4: data detection is carried out to the initial data of step 3 treated primary categories factor and data convert, makes to become
Data after changing are close to stringent normal distribution；
Step 5: using step 4, treated that data establish linear regression model (LRM), optimizes to linear regression model (LRM), obtains
Regression model after tuning；
Step 6: by step 4 data convert inverse transformation by after tuning regression model restore, obtain independent variable and because
The relationship of variable obtains complementary quantitative relationship between the food and drink venue temperature and multiple types factor of target super-large city.
2. the food and drink venue temperature analysis method according to claim 1 towards super-large city, which is characterized in that the sample
Dining room of the notebook data on internet on disclosed food and drink consumption website, the related data in dining room to be analyzed is complete, wait divide
Analysis dining room is located at the typical position of super-large city.
3. the food and drink venue temperature analysis method according to claim 1 or 2 towards super-large city, which is characterized in that institute
It states plurality of classes factor in step 2 and fully considers all kinds of elements that may influence food and drink venue management in super-large city, multiple types
Other factor include dining room quality, eating surroundings, traffic condition, periphery consumption facility and or population group's effect, geography therein
The information of position, public transport and landmark derives from Baidu map；The acquisition methods of the sample data include network
It crawls, the method for open api interface or third-party charging.
4. the food and drink venue temperature analysis method according to claim 3 towards super-large city, which is characterized in that the sample
The primary categories factor of notebook data includes general comment number α, comprehensive star β, pre-capita consumption υ, nearest subway stop spacing from τs, it is neighbouring
Public bus network number τb, periphery large sized commercial center number πcAnd large-scale residential quarter and university number πr, wherein nearest subway station
Distance τsWith neighbouring public bus network number τbIt is category factor relevant to public transport convenience, comprehensive star β is that dining room is comprehensive
The category factor of sexual experience, the number π of periphery large sized commercial centerc, large-scale residential quarter and university number πrAnd pre-capita consumption υ
It is category factor relevant to group's building-up effect；The general comment number α is dependent variable, comprehensive star β, pre-capita consumption υ, recently
Subway station distance τs, neighbouring public bus network number τb, periphery large sized commercial center number πcAnd large-scale residential quarter and university's number
πrFor independent variable.
5. the food and drink venue temperature analysis method according to claim 1 towards super-large city, which is characterized in that the number
According to examine using Shapiro-Wilk Methods of Normality Test realize, Shapiro-Wilk Methods of Normality Test sample data into
Row homogeneity test of variance, under 95% confidence level, there is no obvious poor if more than between 0.05 sample data for inspection result
It is different, so that whether test samples data meet normal distribution.
6. the food and drink venue temperature analysis method according to claim 4 or 5 towards super-large city, which is characterized in that institute
It states data transformation to be handled using Box-Cox transformation, so that data are closer to stringent normal distribution after transformation；The Box-
The transformation rule of Cox transformation are as follows:
Wherein, y indicates that the raw value before Box-Cox transformation, y (λ) indicate raw value through the transformed numerical value of Box-Cox, λ
The conversion parameter undetermined of representative sample data, so that dependent variable y (λ) meets:
Wherein, X indicates independent variable vector, β1For parameter vector, X and β1It is parameter to be estimated, ε indicates random error vector
Residual error, σ are error to standard deviation, InFor the unit matrix of n*n.
7. the food and drink venue temperature analysis method according to claim 6 towards super-large city, which is characterized in that the line
Property regression model be multiple linear regression model or generalized linear regression model, the recurrence obtained using multiple linear regression model
Model are as follows: ylam=β0+β1x1+β2x2+β3x3+β4x4+β5x5+β6x6；
Wherein, ylam indicates the dining room pouplarity converted based on Box-Cox, the general comment number α in sample data；
x1It represents consumer and has merged the synthesis star that taste, environment and service are given a mark as dining room, the synthesis in sample data
Star β；x2Pre-capita consumption υ for the average consumption amount of money in dining room, in sample；x3For the dining room and nearest subway station
Distance, the nearest subway stop spacing in sample data are from τs；x4For public bus network number within the scope of the 1km of dining room, derive from sample
Public bus network number τ near in thisb；x5It indicates the number for the Large Residential District and university that 1km range in dining room is adjoined, derive from
The medium-and-large-sized residential quarter of sample data and university number πr；x6Number for city-level commerce services center, the week in sample data
The number π of side large sized commercial centerc；Meanwhile (β0,β1,β2,β3,β4,β5,β6) it is unknown parameter.
8. the food and drink venue temperature analysis method according to claim 6 towards super-large city, which is characterized in that the step
The method that linear regression model (LRM) is optimized in rapid five are as follows: 1. judged using Spearman or Pearson method each from change
Whether with dependent variable, there are significant correlativities for amount；2. if independent variable and dependent variable have synteny, using more
First Ridge Regression Modeling Method reduces the synteny of independent variable and dependent variable；3. using method of gradual regression to each in initial linear regression model
The value of item parameter is calculated and is analyzed, and the degree of strength for influencing food and drink venue temperature judges, will affect faint
Corresponding independent variable is rejected；4. obtaining final analysis model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811631005.9A CN109857983B (en) | 2018-12-29 | 2018-12-29 | Catering venue heat degree analysis method for super-large cities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811631005.9A CN109857983B (en) | 2018-12-29 | 2018-12-29 | Catering venue heat degree analysis method for super-large cities |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109857983A true CN109857983A (en) | 2019-06-07 |
| CN109857983B CN109857983B (en) | 2022-09-30 |
Family
ID=66893159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811631005.9A Active CN109857983B (en) | 2018-12-29 | 2018-12-29 | Catering venue heat degree analysis method for super-large cities |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109857983B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783213A (en) * | 2020-07-09 | 2020-10-16 | 沈阳建筑大学 | Cavity ventilation design optimization method and system for extra-large railway station waiting hall |
| CN112434262A (en) * | 2020-11-22 | 2021-03-02 | 同济大学 | Waterfront public space activity influence factor identification method and terminal |
| CN113379269A (en) * | 2021-06-21 | 2021-09-10 | 华南理工大学 | Urban business function zoning method, device and medium for multi-factor spatial clustering |
| CN113743540A (en) * | 2021-11-04 | 2021-12-03 | 华能（天津）煤气化发电有限公司 | Coal quality melting point prediction method based on multi-model fusion Stacking algorithm |
| CN114048910A (en) * | 2021-11-17 | 2022-02-15 | 东南大学 | Community public greenbelt group layout optimization method based on big data |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106776868A (en) * | 2016-11-29 | 2017-05-31 | 浙江工业大学 | A kind of restaurant score in predicting method based on multiple linear regression model |
| US20180018704A1 (en) * | 2016-07-15 | 2018-01-18 | NXT-ID, Inc. | Preference-Driven Advertising Systems and Methods |
-
2018
- 2018-12-29 CN CN201811631005.9A patent/CN109857983B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180018704A1 (en) * | 2016-07-15 | 2018-01-18 | NXT-ID, Inc. | Preference-Driven Advertising Systems and Methods |
| CN106776868A (en) * | 2016-11-29 | 2017-05-31 | 浙江工业大学 | A kind of restaurant score in predicting method based on multiple linear regression model |
Non-Patent Citations (1)
| Title |
|---|
| 夏令军等: "中国地级城市餐饮业分布格局及影响因素――基于"大众点评网"数据的实证研究", 《经济地理》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111783213A (en) * | 2020-07-09 | 2020-10-16 | 沈阳建筑大学 | Cavity ventilation design optimization method and system for extra-large railway station waiting hall |
| CN111783213B (en) * | 2020-07-09 | 2023-10-24 | 沈阳建筑大学 | Cavity ventilation design optimization method and system for oversized train station waiting hall |
| CN112434262A (en) * | 2020-11-22 | 2021-03-02 | 同济大学 | Waterfront public space activity influence factor identification method and terminal |
| CN113379269A (en) * | 2021-06-21 | 2021-09-10 | 华南理工大学 | Urban business function zoning method, device and medium for multi-factor spatial clustering |
| CN113379269B (en) * | 2021-06-21 | 2023-08-18 | 华南理工大学 | Urban business function partitioning method, device and medium for multi-factor spatial clustering |
| CN113743540A (en) * | 2021-11-04 | 2021-12-03 | 华能（天津）煤气化发电有限公司 | Coal quality melting point prediction method based on multi-model fusion Stacking algorithm |
| CN114048910A (en) * | 2021-11-17 | 2022-02-15 | 东南大学 | Community public greenbelt group layout optimization method based on big data |
| CN114048910B (en) * | 2021-11-17 | 2024-03-29 | 东南大学 | Community public green land group layout optimization method based on big data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109857983B (en) | 2022-09-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109857983A (en) | A kind of food and drink venue temperature analysis method towards super-large city | |
| Xu et al. | Difference of urban development in China from the perspective of passenger transport around Spring Festival | |
| CN105183870B (en) | A kind of urban function region detection method and system using microblogging location information | |
| Park et al. | Does distance matter? Examining the distance effect on tourists’ multi-attraction travel behaviors | |
| Truchet et al. | Spatial influence of attractions on tourism development | |
| CN105224681B (en) | Customer requirement retrieval method and system based on family's place of working context environmental | |
| Parroco et al. | Multi-destination trips and tourism statistics: Empirical evidences in Sicily | |
| Cao et al. | City-level determinants of private car ownership in China | |
| CN109214952A (en) | A kind of traveller's decision model construction method | |
| CN106022573A (en) | Multi-dimensional hotel grading method | |
| CN113204719B (en) | Urban house rent valuation method based on position information superposition and deep neural network | |
| Zhang et al. | Mining spatial patterns of food culture in China using restaurant POI data | |
| Wang et al. | Spatial distribution pattern of the customer count and satisfaction of commercial facilities based on social network review data in Beijing, China | |
| Hu et al. | Using Wi-Fi probe and location data to analyze the human distribution characteristics of green spaces: A case study of the Yanfu Greenland Park, China | |
| CN106021290A (en) | Method for social network association excavation based on multi-scale geographic information | |
| Gao et al. | A multi-scale comparison of tourism attraction networks across China | |
| CN113239316A (en) | City business facility layout rationality evaluation method and system considering crowd consumption characteristics | |
| Borges et al. | Hybrid approach to representative building archetypes development for urban models–A case study in Andorra | |
| Zhen et al. | Spatial distribution characteristics of residents’ emotions based on Sina Weibo big data: A case study of Nanjing | |
| Wang et al. | Inter‐city association pattern recognition by constructing cultural semantic similarity network | |
| Dewilde et al. | Housing and poverty: the'missing link' | |
| Shi et al. | Do similar social groups have similar mobility in a city? Social areas and mobility in Shenzhen, China | |
| Liu et al. | An analysis on the spatiotemporal behavior of inbound tourists in Jiaodong Peninsula based on Flickr geotagged photos | |
| Zhao et al. | Depicting urban multi-scale tourist activity spaces using digital footprints for smart destinations | |
| Wu et al. | Urban functional area recognition based on unbalanced clustering |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |