Joint prediction method of urban tourism transportation demand based on attraction intensity

Abstract

The present invention discloses a joint prediction method for urban tourism traffic demand based on attraction intensity, including: (1) obtaining target area data; (2) dividing the target area into tourism-type, transportation-type and comprehensive-type communities; (3) defining tourism traffic and subdividing it into occurrence-type tourism traffic and attraction-type tourism traffic; (4) using the GM (1,1) grey prediction model to predict the total tourism demand in the target area; (5) using the factor analysis method to calculate the attraction intensity of scenic spots in the target area. (6) predicting the occurrence and attraction traffic volume of tourism in scenic spots. (7) predicting the tourism distribution traffic volume in communities. The present invention takes into account the differences in characteristics between tourism traffic and commuting traffic, and the impact of the location distribution of transportation, commerce, hotels, etc. on tourism traffic demand, introduces factor analysis to quantify the attraction intensity of scenic spots, proposes a joint prediction method for tourism traffic, and obtains the OD matrix of tourism traffic distribution, which provides a theoretical basis for tourism traffic planning.

Description

Technical Field

[0001] The invention relates to a method for predicting urban tourism traffic demand, and in particular to a method for jointly predicting urban tourism traffic demand based on attraction intensity. Background Art

[0002] Tourism transportation refers to the entire travel process completed with the help of specific transportation tools for the purpose of tourism. Scientific tourism transportation demand forecasting is the basis for reasonable tourism transportation planning, and can solve problems such as resource waste and traffic congestion caused by unreasonable planning. There are significant differences between the characteristics of tourism transportation and commuter transportation. Tourism transportation is strongly correlated with factors such as the scenic area itself and traffic impedance, but weakly correlated with land use and population and social factors. Therefore, traditional methods cannot be used to predict transportation demand.

[0003] The basis of tourism traffic demand forecasting is the total tourism demand forecasting. At present, there have been many studies on the total tourism demand forecasting in cities at home and abroad. The total tourism demand forecasting is the basis of tourism traffic demand forecasting. The tourism demand methods are divided into three categories: time series model, econometric model, and artificial intelligence-based model. The gray prediction model is a prediction model with simple principle, high prediction accuracy and less data requirements. For the problem of urban tourism traffic demand forecasting, many studies are based on the traditional four-stage prediction method, and an improved four-stage prediction method adapted to the characteristics of tourism traffic is proposed. Among them, tourism attraction is the most important influencing factor. Tourism attraction refers to the attraction of tourist destinations to tourists. It is a comprehensive reflection of factors such as the quality level, popularity, transportation conditions, and infrastructure of tourist resources. There are certain research results on the measurement of tourism attraction at home and abroad. Most of them start from the regional macro perspective, combine economic research methods, establish attraction measurement models, and study the overall attraction of provincial and city-level tourist destinations. However, there are relatively few studies on the attraction of tourist attractions. Traditional methods for studying tourism attraction rely on questionnaire surveys, hierarchical analysis methods, etc., which are subjective, time-consuming, labor-intensive, and insufficiently comprehensive in data acquisition. Therefore, further research is needed on the quantification of attraction. Summary of the invention

[0004] Purpose of the invention: In order to overcome the shortcomings of the prior art, the present invention provides a joint prediction method for urban tourism traffic demand based on attraction intensity. By collecting data from tourism websites, the factor analysis method is introduced to quantify the attraction intensity of scenic spots, and the tourism distribution traffic volume is predicted based on the attraction intensity. The present invention can take into account the characteristics of tourism traffic demand, effectively predict urban tourism traffic demand, and lay the foundation for tourism traffic planning.

[0005] Technical solution: A joint prediction method for urban tourism traffic demand based on attraction intensity, which is used to jointly predict urban tourism occurrence, attraction traffic volume and distribution traffic volume.

[0006] The method of the present invention comprises the following steps:

[0007] (1) Obtain data on the target area, including: scenic area characteristics data, location data, travel itinerary data, and total tourism demand data over the years;

[0008] (2) Based on the scenic area characteristic data and location data of the target area, combined with the administrative area and natural resource boundaries, the target area is divided into a number of different types of sub-areas, wherein the sub-area types are divided into three categories: tourism type, transportation type and comprehensive type;

[0009] (3) define tourism transportation and its types;

[0010] (4) Use the GM(1,1) grey forecasting model to predict the total tourism demand in the target area;

[0011] (5) Use factor analysis to calculate the attraction intensity of scenic spots in the target area;

[0012] (6) Forecast tourism occurrence and attracted traffic volume based on scenic spots;

[0013] (7) Predict tourism distribution traffic volume based on the community.

[0014] Furthermore, in step (1):

[0015] The characteristic data of scenic spots include the name of the scenic spot, longitude and latitude, ticket price, scenic spot rating, and number of scenic spot reviews;

[0016] Location data refers to the POI data related to tourism resources (including rail stations, commercial centers, hotels, homestays, etc.). POI data is the basic data reflecting the spatial distribution form. The attributes contained in each POI data include POI name, longitude and latitude, address information, and category.

[0017] Travel itinerary data includes the departure and destination;

[0018] The total tourism demand data over the years include the year and the total tourism demand.

[0019] Furthermore, in step (3):

[0020] A tourist transportation trip is defined as a single transportation trip with at least one departure or destination located in a tourist-type community;

[0021] Tourism transportation is divided into two categories: occurrence-type tourism transportation and attraction-type tourism transportation. Among them, occurrence-type tourism transportation includes three categories: tourism-type→tourism-type, tourism-type→transportation-type and tourism-type→comprehensive. "Tourism-type→tourism-type" refers to tourism transportation whose departure and destination are both tourism-type communities, "tourism-type→transportation-type" refers to tourism transportation whose departure place is a tourism-type community and whose destination is a transportation-type community, and "tourism-type→comprehensive-type" refers to tourism transportation whose departure place is a tourism-type community and whose destination is a comprehensive community. Attraction-type tourism transportation includes two categories: transportation-type→tourism-type and comprehensive→tourism-type. "Transportation-type→tourism-type" refers to tourism transportation whose departure place is a transportation-type community and whose destination is a tourism-type community, and "comprehensive-type→tourism-type" refers to tourism transportation whose departure place is a comprehensive community and whose destination is a tourism-type community.

[0022] Furthermore, the specific steps of step (4) are as follows:

[0023] (4.1) The total tourism demand in the target area is expressed as the series X (0 )={x (0) (1),x (0) (2),...,x (0) (n)}, calculate the cumulative sequence X (1) ={x (1) (1),...,x (1) (n)}, where x (1) The calculation formula of (t) is:

[0024]

[0025] In the formula, x (0) (t) represents the total tourism demand of the target area in year t, x (0) (i) represents the total tourism demand of the target area in year i, and n represents the total number of years;

[0026] (4.2) for X (1) Set up the linear differential equation:

[0027]

[0028] In the formula, a and u are the parameters to be calculated;

[0029] (4.3) Calculate a and u by the least squares method:

[0030]

[0031] In the formula, Y=(x (0) (2),x (0) (3),…,x (0) (n))T ,

[0032]

[0033] (4.4) According to a and u, we can get the solution of the linear differential equation:

[0034]

[0035] (4.5) Yes Do cumulative subtraction to calculate the total tourism demand x of the target area in year t (0) The predicted value T of (t) is calculated as follows:

[0036]

[0037] In the formula,

[0038] Furthermore, the specific steps of step (5) are as follows:

[0039] (5.1) Select indicators for factor analysis from the scenic spot characteristic data: use ticket price as the cost indicator, scenic spot rating as the rating indicator, the number of scenic spot reviews as the review indicator, and the longitude and latitude spherical distance from the scenic spot to the nearest rail station as the transportation indicator;

[0040] (5.2) Perform factor analysis on the indicators selected in step (5.1), extract common factors using principal component analysis, perform orthogonal rotation transformation using the maximum variance method, and select the cumulative variance contribution rate Reach 80% of the first M common factors and get the mth common factor about the i-th index x i The score coefficient β mi , and then calculate the comprehensive factor score F of the scenic spot, the formula is as follows:

[0041]

[0042] In the formula, F m is the score of the scenic spot on the mth common factor, γ m is the variance contribution rate of the mth common factor, γ i is the variance contribution rate of the i-th common factor;

[0043] (5.3) Calculate the attraction intensity w of the scenic spot:

[0044]

[0045] In the formula, F min is the minimum value of the comprehensive factor scores of all scenic spots; F max It is the maximum value of the comprehensive factor scores of all scenic spots.

[0046] Furthermore, the step (5.1) further includes standardizing the index and passing the KMO test and Bartlett's test.

[0047] Furthermore, in step (6), the attraction traffic volume of the scenic spot is equal to the occurrence traffic volume, then:

[0048]

[0049] In the formula, a n is the traffic volume attracted by the nth scenic spot in the target area; g n is the traffic volume of the nth scenic spot in the target area; w n is the attraction intensity of the nth scenic spot in the target area; T is the predicted value of the total tourism demand in the target area in the tth year; N is the number of scenic spots in the target area.

[0050] Furthermore, the step (7) specifically comprises the following steps:

[0051] (7.1) The amount of traffic generated / attracted by the scenic spots in each tourist community is added up to obtain the amount of tourism generated / attracted by each tourist community; the attraction intensity of the scenic spots in each tourist community is added up to obtain the attraction intensity of each tourist community. The calculation formula is as follows:

[0052]

[0053]

[0054] In the formula, A t is the traffic volume attracted by the t-th tourist community in the target area; G t is the traffic volume of the t-th tourist community in the target area; W t is the attraction intensity of the t-th tourist community in the target area; w ts is the attraction intensity of the sth scenic spot in the tth tourist community; S is the number of scenic spots in the tth tourist community in the target area.

[0055] (7.2) Calculate the probability of occurrence and attraction type tourism transportation trips, where the probability of occurrence type tourism transportation trip p g l = ... It is equal to the ratio of the h-th type of attraction tourism distribution traffic volume to the total attraction tourism distribution traffic volume in the target area; l = 1 means "tourism type → tourism type", l = 2 means "tourism type → transportation type", l = 3 means "tourism type → comprehensive type"; h = 1 means "transportation type → tourism type", h = 2 means "comprehensive type → tourism type";

[0056] (7.3) Calculate the traffic volume of occurrence tourism distribution:

[0057]

[0058] In the formula, is the traffic volume of the first type of tourism distribution from cell t1 to cell t2 in the target area, is the traffic volume of cell t1 in the target area; is the attraction intensity of cell t2 in the target area; is the traffic impedance function in the form of a power function, γ is a parameter (its value is generally between 0.6 and 3.5), is the spherical distance in longitude and latitude between the centroids of cells t1 and t2 in the target area; is the sum of the number of track edges connected to the track stations in cell t2 within the target area; is the number of POIs in the target area within the community t2; M1, M2, and M3 are the number of tourist communities, transportation communities, and comprehensive communities in the target area, respectively. is the attraction intensity of the tourist community m1 in the target area; is the spherical distance in longitude and latitude between the centroids of the cell t1 and the tourist-type cell m1 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and traffic-type cell m2 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and comprehensive cell m3 in the target area; is the sum of the number of track edges connected to the track stations in the transportation-type community m2 in the target area, is the number of POIs in the comprehensive community m3 in the target area;

[0059] (7.4) Calculate the traffic volume of attraction tourism distribution:

[0060]

[0061] In the formula, is the h-th type of attractive tourism distribution traffic volume from cell t1 to cell t2 in the target area, Attract traffic to community j;

[0062] (7.5)Thus, the OD matrix of tourism traffic distribution is obtained.

[0063] Furthermore, the probability of each type of occurrence-type tourist transportation trip and the probability of attraction-type tourist transportation trip in step (7.2) satisfy the following constraints:

[0064]

[0065]

[0066] In the formula, p g 1 represents the probability of occurrence of tourism transportation trip of type “tourism type → tourism type”, p g 2 represents the probability of occurrence of tourism transportation trip of type “tourism type → transportation type”, p g 3 represents the probability of occurrence of tourism transportation trip of type "tourism type → comprehensive type"); p a 1 represents the probability of "transportation type → tourism type" type of attraction type tourism transportation trip, p a 2 It indicates the probability of “comprehensive → tourism” attraction-type tourism transportation trip.

[0067] Furthermore, in the step (7.2), the departure place and destination in the travel itinerary data are marked as tourism type, transportation type or comprehensive type respectively, and the tourism traffic volume of each type of tourism transportation is obtained by classification and summary, and then the probability of occurrence type tourism transportation and the probability of attraction type tourism transportation are calculated.

[0068] Beneficial effect: Compared with the prior art, the significant effect of the present invention is that, by collecting Internet tourism data, the present invention is efficient, convenient and covers a wide range, thus avoiding the shortcomings of traditional methods such as long time consumption, large workload and incomplete data acquisition; the factor analysis method is introduced to quantitatively calculate the attraction intensity of scenic spots, thus avoiding the problem of certain subjectivity in traditional methods; the attraction intensity of scenic spots is used to calculate the occurrence and attraction traffic volume of scenic spots, and then the occurrence-type and attraction-type tourism distribution traffic volume is distributed and calculated to obtain the tourism traffic distribution OD matrix, thus laying the foundation for tourism traffic planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 is a flow chart of the steps of the present invention;

[0070] Figure 2 This is the tourism transportation classification diagram of the present invention. DETAILED DESCRIPTION

[0071] In order to explain the technical solution disclosed in the present invention in detail, further elaboration is given below in conjunction with the accompanying drawings and specific implementation cases.

[0072] In one embodiment, Figure 1 As shown, a joint prediction method for urban tourism traffic demand based on attraction intensity is provided, and the method is used to jointly predict the occurrence of urban tourism and the attracted traffic volume and the distributed traffic volume, including the following steps:

[0073] (1) Obtain data on the target area, including: scenic area characteristics data, location data, travel itinerary data, and total tourism demand data over the years;

[0074] (2) Based on the scenic area characteristic data and location data of the target area, combined with the administrative area and natural resource boundaries, the target area is divided into a number of different types of sub-areas, wherein the sub-area types are divided into three categories: tourism type, transportation type and comprehensive type;

[0075] (3) define tourism transportation and its types;

[0076] (4) Use the GM(1,1) grey forecasting model to predict the total tourism demand in the target area;

[0077] (5) Use factor analysis to calculate the attraction intensity of scenic spots in the target area;

[0078] (6) Forecast tourism occurrence and attracted traffic volume based on scenic spots;

[0079] (7) Predict tourism distribution traffic volume based on the community.

[0080] In one embodiment, the data in step (1) has the following characteristics:

[0081] (1.1) Scenic spot characteristic data, each data includes five indicators: scenic spot name, longitude and latitude, ticket price, scenic spot rating, and number of scenic spot reviews;

[0082] (1.2) Location data refers to the POI data related to tourism resources (including rail stations, commercial centers, hotels, homestays, etc.). POI data is the basic data reflecting the spatial distribution pattern. The attributes of each POI data include POI name, longitude and latitude, address information, and category.

[0083] (1.3) Travel itinerary data, each data includes two indicators: departure place and destination;

[0084] (1.4) Data on total tourism demand over the years, including two indicators: year and total tourism demand.

[0085] In one embodiment, the specific steps of step (3) are as follows:

[0086] (3.1) Definition of tourism transportation

[0087] Tourism transportation is divided into three levels: regional transportation, urban transportation and internal transportation of scenic spots. The research object of this invention is urban transportation. Urban tourism transportation is greatly affected by individual preferences of travelers and characteristics of scenic spots, so it is more discrete. If the study is based on the complete travel chain, the calculation is difficult and the data volume is large. Therefore, the tourism travel chain is decomposed into single trips between communities, that is, the tourism transportation trip in this invention is defined as: a single transportation trip with at least one of the departure or destination distributed in a tourist community.

[0088] (3.2) Tourism and transportation classification

[0089] Tourism transportation can be divided into occurrence-type tourism transportation and attraction-type tourism transportation. The characteristic of occurrence-type tourism transportation is that the departure place is a tourism-type community, and the characteristic of attraction-type tourism transportation is that the destination is a tourism-type community. For the sake of convenience, single transportation trips where both the departure place and the destination are tourism-type communities are classified as occurrence-type tourism transportation. Therefore, there are three types of occurrence-type tourism transportation, namely tourism-type→tourism-type (indicating tourism transportation trips where both the departure place and the destination are tourism-type communities), tourism-type→transportation-type (indicating tourism transportation trips where the departure place is a tourism-type community and the destination is a transportation-type community), and tourism-type→comprehensive-type (indicating tourism transportation trips where the departure place is a tourism-type community and the destination is a comprehensive-type community); there are two types of attraction-type tourism transportation, namely transportation-type→tourism-type (tourism transportation trips where the departure place is a transportation-type community and the destination is a tourism-type community), and comprehensive-type→tourism-type (tourism transportation trips where the departure place is a comprehensive-type community and the destination is a tourism-type community), such as Figure 2 shown.

[0090] In one embodiment, the specific steps of step (4) are as follows:

[0091] (4.1) The total tourism demand in the target area can be expressed as the series X (0) ={x (0) (1),x (0) (2),...,x (0) (n)}, calculate the cumulative sequence X (1) ={x (1) (1),...,x (1) (n)}, where element x (1) The calculation formula of (t) is:

[0092]

[0093] In the formula, x(0) (t) represents the total tourism demand of the target area in year t, x (0) (i) represents the total tourism demand of the target area in the i-th year, n represents the total number of years. In formula 1, x (0) (k) represents the total tourism demand data for the kth year.

[0094] (4.2) for X (1) Establish a linear differential equation, where a and u are the parameters to be calculated.

[0095]

[0096] (4.3) Calculate the parameters a,u by the least squares method:

[0097]

[0098] In the formula, Y=(x (0) (2),x (0) (3),…,x (0) (n)) T ,

[0099]

[0100] (4.4) According to the parameters a and u, we can get the solution of the linear differential equation:

[0101]

[0102] (4.5) Yes Do cumulative subtraction to calculate the total tourism demand x of the target area in year t (0) The predicted value T of (t) is calculated as follows:

[0103]

[0104] In one embodiment, the attraction intensity of a scenic spot is a quantitative index that characterizes the attraction of a scenic spot to tourists, and the attraction intensity of a scenic spot is quantified using a factor analysis method. The specific steps of step (5) are as follows:

[0105] (5.1) Use the characteristic data of the scenic spot to select k indicators for factor analysis: use the ticket price as the value of the cost indicator; use the scenic spot score as the value of the score indicator; use the number of scenic spot reviews as the value of the review indicator; calculate the longitude and latitude spherical distance from the scenic spot to the nearest rail station as the value of the transportation indicator. After the above indicators are standardized, k indicators x1, x2, ..., x3 can be selected through the KMO test (requires KMO value> 0.5) and Bartlett's test (requires Bartlett's test P value< 0.05). k .

[0106] (5.2) Calculate the comprehensive factor score F of the scenic spot: Perform factor analysis on the k indicators selected in step (5.1), extract factors using principal component analysis, perform orthogonal rotation transformation using the maximum variance method, and select the cumulative variance contribution rate Reach 80% of the first M common factors and get the mth common factor about the i-th index x i The score coefficient β mi , and then calculate the comprehensive factor score of the scenic spot.

[0107] The comprehensive factor score is the specific value of each factor in each scenic spot, so the comprehensive factor score F is:

[0108]

[0109] In the formula, F m is the score of the scenic spot on the mth common factor, γ m is the variance contribution rate of the mth common factor, γ i is the variance contribution rate of the ith common factor.

[0110] (5.3) Calculate the attraction intensity w of the scenic spot, w is distributed in the interval [0.1, 0.9], and the calculation formula is:

[0111]

[0112] In the formula, F min is the minimum value of the comprehensive factor scores of all scenic spots; F max It is the maximum value of the comprehensive factor scores of all scenic spots.

[0113] In one embodiment, a scenic spot attracts one traffic trip, and tourists will inevitably make one traffic trip after leaving the scenic spot, so the traffic volume attracted by each scenic spot is equal to the traffic volume generated. The predicted value T of the total tourism demand of the target area in the tth year is allocated based on the proportion of the attraction intensity of the scenic spot in the total attraction intensity of all scenic spots in the target area (this value is obtained from step (4)). The step (6) is specifically as follows:

[0114]

[0115] In the formula, a n is the traffic volume attracted by the nth scenic spot in the target area; g n is the traffic volume of the nth scenic spot in the target area; w n is the attraction intensity of the nth scenic spot in the target area; T is the predicted value of the total tourism demand in the target area in the tth year; N is the number of scenic spots in the target area.

[0116] In one embodiment, the step (7) specifically comprises the following steps:

[0117] (7.1) The amount of traffic generated (attracted) by the scenic spots in each tourist community is added up to obtain the amount of tourism generated (attracted) traffic in each tourist community; the attraction intensity of the scenic spots in each tourist community is added up to obtain the attraction intensity of each tourist community. The calculation formula is as follows:

[0118]

[0119]

[0120] In the formula, A t is the traffic volume attracted by the t-th tourist community in the target area; G t is the traffic volume of the t-th tourist community in the target area; W t is the attraction intensity of the t-th tourist community in the target area; w ts is the attraction intensity of the sth scenic spot in the tth tourist community; S is the number of scenic spots in the tth tourist community in the target area.

[0121] (7.2) Calculate the probability of occurrence and attraction type tourist traffic. The probability of occurrence type tourist traffic of the first type is equal to the proportion of this type of tourist traffic volume to the total occurrence type tourist traffic volume, expressed as p g l = (l = 1 means "tourism type → tourism type", l = 2 means "tourism type → transportation type", l = 3 means "tourism type → comprehensive type"); the probability of the hth type of attraction tourism traffic is equal to the proportion of this type of tourism traffic to the total attraction tourism traffic, expressed as Indicates (h=1 means "transportation type → tourism type", h=2 means "comprehensive type → tourism type"), as shown in Table 1.

[0122] Table 1

[0123]

[0124] Using the travel itinerary data (step (1.3)), the departure and destination of each trip data are marked as tourism type, transportation type or comprehensive type, and the tourism traffic volume of each type of tourism transportation trip can be classified and summarized to calculate p g l and where p g l and The following constraints are met:

[0125]

[0126]

[0127] (7.3) Calculate the occurrence type tourism distribution traffic volume. Convert the occurrence and attraction traffic volume into the OD volume between the communities, that is, the OD matrix, where O represents the origin and D represents the destination.

[0128]

[0129] In the formula, is the traffic volume of the first type of tourism distribution from cell t1 to cell t2 in the target area, is the traffic volume of cell t1 in the target area; is the attraction intensity of cell t2 in the target area; is the traffic impedance function in the form of a power function, γ is a parameter (its value is generally between 0.6 and 3.5), is the spherical distance in longitude and latitude between the centroids of cells t1 and t2 in the target area; is the sum of the number of track edges connected to the track stations in cell t2 within the target area; is the number of POIs in the target area within the community t2; M1, M2, and M3 are the number of tourist communities, transportation communities, and comprehensive communities in the target area, respectively. is the attraction intensity of the tourist community m1 in the target area; is the spherical distance in longitude and latitude between the centroids of the cell t1 and the tourist-type cell m1 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and traffic-type cell m2 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and comprehensive cell m3 in the target area; is the sum of the number of track edges connected to the track stations in the transportation-type community m2 in the target area, is the number of POIs in the comprehensive community m3 in the target area;

[0130] (7.4) Calculate the traffic volume of attraction tourism distribution.

[0131]

[0132] In the formula, is the h-th type of attractive tourism distribution traffic volume from cell t1 to cell t2 in the target area, Attract traffic to community j;

[0133] (7.5) After all the above steps, the OD matrix of tourism traffic distribution can be obtained, and the joint prediction of tourism traffic demand is completed.

[0134] For the convenience of explanation, it is assumed that there are 2 cells of each type, and the OD matrix is ​​shown in Table 2.

[0135] Table 2

[0136]

[0137] The present invention takes into account the differences in characteristics between tourist transportation and commuter transportation, the impact of the location distribution of transportation, commerce, hotels, etc. on tourist transportation demand, introduces factor analysis to quantify the attraction intensity of scenic spots, and proposes a tourist transportation joint prediction method to obtain the OD matrix of tourist transportation distribution, which provides a theoretical basis for tourist transportation planning.

[0138] The above description is only a specific implementation of the present invention, but the protection scope of the present invention is not limited thereto. Any person familiar with the technology can understand and think of any changes or substitutions within the technical scope disclosed by the present invention, which should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A joint prediction method for urban tourism traffic demand based on attraction intensity, characterized in that: The method comprises the following steps: (1) Obtain data on the target area, including: scenic area characteristics data, location data, travel itinerary data, and total tourism demand data over the years; (2) Based on the scenic area characteristic data and location data of the target area, combined with the administrative area and natural resource boundaries, the target area is divided into a number of different types of sub-areas, wherein the sub-area types are divided into three categories: tourism type, transportation type and comprehensive type; (3) define tourism transportation and its types; (4) Use the GM(1,1) grey forecasting model to predict the total tourism demand in the target area; (5) Use factor analysis to calculate the attraction intensity of scenic spots in the target area; (6) Taking scenic spots as units, predict the occurrence of tourism and the volume of traffic attracted; (7) Predict tourism distribution traffic volume based on the community; The specific steps of step (4) are as follows: (4.1) The total tourism demand in the target area is expressed as the series X (0) ={x (0) (1),x (0) (2),...,x (0) (n)}, calculate the cumulative sequence X (1) ={x (1) (1),...,x (1) (n)}, where x (1) The calculation formula of (t) is: In the formula, x (0) (t) represents the total tourism demand of the target area in year t, x (0) (i) represents the total tourism demand of the target area in year i, and n represents the total number of years; (4.2) for X (1) Set up the linear differential equation: In the formula, a and u are the parameters to be calculated; (4.3) Calculate a and u by the least squares method: In the formula, Y=(x (0) (2),x (0) (3),…,x (0) (n)) T , (4.4) According to a and u, we can get the solution of the linear differential equation: (4.5) Yes Do cumulative subtraction to calculate the total tourism demand x of the target area in year t (0) The predicted value T of (t) is calculated as follows: In the formula, 2. A joint prediction method for urban tourism traffic demand based on attraction intensity as claimed in claim 1, characterized in that: In the step (1): The characteristic data of scenic spots include the name of the scenic spot, longitude and latitude, ticket price, scenic spot rating, and number of scenic spot reviews; Location data refers to POI data related to tourism resources, including POI name, longitude and latitude, address information, and category; Travel itinerary data includes the departure and destination; The total tourism demand data over the years include the year and the total tourism demand.

3. The method for joint prediction of urban tourism traffic demand based on attraction intensity as claimed in claim 1, characterized in that: In the step (3): A tourist transportation trip is defined as a single transportation trip with at least one departure or destination located in a tourist-type community; Tourism transportation is divided into two categories: occurrence-type tourism transportation and attraction-type tourism transportation. Among them, occurrence-type tourism transportation includes three categories: tourism-type→tourism-type, tourism-type→transportation-type and tourism-type→comprehensive. "Tourism-type→tourism-type" means tourism transportation with both the departure and destination being tourism-type communities, "tourism-type→transportation-type" means tourism transportation with the departure place being a tourism-type community and the destination being a transportation-type community, and "tourism-type→comprehensive-type" means tourism transportation with the departure place being a tourism-type community and the destination being a comprehensive community; attraction-type tourism transportation includes two categories: transportation-type→tourism-type and comprehensive→tourism-type. "Transportation-type→tourism-type" means tourism transportation with the departure place being a transportation-type community and the destination being a tourism-type community, and "comprehensive→tourism-type" means tourism transportation with the departure place being a comprehensive community and the destination being a tourism-type community.

4. The method for joint prediction of urban tourism traffic demand based on attraction intensity as claimed in claim 1, characterized in that: The specific steps of step (5) are as follows: (5.1) Select indicators for factor analysis from the scenic spot characteristic data: use ticket price as the cost indicator, scenic spot rating as the rating indicator, the number of scenic spot reviews as the review indicator, and the longitude and latitude spherical distance from the scenic spot to the nearest rail station as the transportation indicator; (5.2) Perform factor analysis on the indicators selected in step (5.1), extract common factors using principal component analysis, perform orthogonal rotation transformation using the maximum variance method, and select the cumulative variance contribution rate Reach 80% of the first M common factors and get the mth common factor about the i-th index x i The score coefficient β mi , and then calculate the comprehensive factor score F of the scenic spot, the formula is as follows: In the formula, F m is the score of the scenic spot on the mth common factor, γ m is the variance contribution rate of the mth common factor, γ i is the variance contribution rate of the i-th common factor; (5.3) Calculate the attraction intensity w of the scenic spot: In the formula, F min is the minimum value of the comprehensive factor scores of all scenic spots; F max It is the maximum value of the comprehensive factor scores of all scenic spots.

5. A joint prediction method for urban tourism traffic demand based on attraction intensity as claimed in claim 4, characterized in that: The step (5.1) further includes standardizing the index and passing the KMO test and Bartlett's test.

6. The method for joint prediction of urban tourism traffic demand based on attraction intensity as claimed in claim 1, characterized in that: In step (6), the attraction traffic volume of the scenic spot is equal to the occurrence traffic volume, then: In the formula, a n is the traffic volume attracted by the nth scenic spot in the target area; g n is the traffic volume of the nth scenic spot in the target area; w n is the attraction intensity of the nth scenic spot in the target area; T is the predicted value of the total tourism demand in the target area in year t; N is the number of scenic spots in the target area.

7. The method for joint prediction of urban tourism traffic demand based on attraction intensity as claimed in claim 3, characterized in that: The step (7) specifically comprises the following steps: (7.1) The amount of traffic generated / attracted by the scenic spots in each tourist community is added up to obtain the amount of tourism generated / attracted by each tourist community; the attraction intensity of the scenic spots in each tourist community is added up to obtain the attraction intensity of each tourist community. The calculation formula is as follows: In the formula, A t is the traffic volume attracted by the t-th tourist community in the target area; G t is the traffic volume of the t-th tourist community in the target area; W t is the attraction intensity of the t-th tourist community in the target area; w ts is the attraction intensity of the sth scenic spot in the tth tourist community; S is the number of scenic spots in the tth tourist community in the target area; (7.2) Calculate the probability of occurrence and attraction type tourism transportation trips, where the probability of occurrence type tourism transportation trip p g l = ... It is equal to the ratio of the h-th type of attraction tourism distribution traffic volume to the total attraction tourism distribution traffic volume in the target area; l = 1 means "tourism type → tourism type", l = 2 means "tourism type → transportation type", l = 3 means "tourism type → comprehensive type"; h = 1 means "transportation type → tourism type", h = 2 means "comprehensive type → tourism type"; (7.3) Calculate the traffic volume of occurrence tourism distribution: In the formula, is the traffic volume of the first type of tourism distribution from cell t1 to cell t2 in the target area, is the traffic volume of cell t1 in the target area; is the attraction intensity of cell t2 in the target area; is the traffic impedance function in the form of a power function, γ is a parameter, is the spherical distance in longitude and latitude between the centroids of cells t1 and t2 in the target area; is the sum of the number of track edges connected to the track stations in cell t2 within the target area; is the number of POIs in the target area within the community t2; M1, M2, and M3 are the number of tourist communities, transportation communities, and comprehensive communities in the target area, respectively. is the attraction intensity of the tourist community m1 in the target area; is the spherical distance in longitude and latitude between the centroids of the cell t1 and the tourist-type cell m1 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and traffic-type cell m2 in the target area; is the spherical distance in longitude and latitude between the centroids of cell t1 and comprehensive cell m3 in the target area; is the sum of the number of track edges connected to the track stations in the transportation-type community m2 in the target area, is the number of POIs in the comprehensive community m3 in the target area; (7.4) Calculate the traffic volume of attraction tourism distribution: In the formula, is the h-th type of attractive tourism distribution traffic volume from cell t1 to cell t2 in the target area, Attract traffic to community j; (7.5)Thus, the OD matrix of tourism traffic distribution is obtained.

8. A joint prediction method for urban tourism traffic demand based on attraction intensity as claimed in claim 7, characterized in that: In step (7.2), the probability of each type of occurrence-type tourist transportation trip and the probability of attraction-type tourist transportation trip satisfy the following constraints: In the formula, p g 1 represents the probability of occurrence of tourism transportation trip of type "tourism type → tourism type", p g 2 represents the probability of occurrence of tourism transportation trip of type "tourism type → transportation type", p g 3 represents the probability of occurrence of tourism transportation trip of type "tourism type → comprehensive type"); p a 1 represents the probability of "transportation type → tourism type" type of attraction tourism transportation trip, p a 2 It indicates the probability of "comprehensive → tourism" attraction-type tourism transportation trip.

9. The method for joint prediction of urban tourism traffic demand based on attraction intensity as claimed in claim 7, characterized in that: In the step (7.2), the departure place and destination in the travel itinerary data are marked as tourism type, transportation type or comprehensive type respectively, and the tourism traffic volume of each type of tourism transportation is obtained by classification and summary, and then the probability of occurrence type tourism transportation and the probability of attraction type tourism transportation are calculated.

Images