Tourism area passenger flow volume prediction method and device, electronic equipment and storage medium

Abstract

The invention provides a tourism area passenger flow volume prediction method and device, electronic equipment and a storage medium. The tourism area passenger flow volume prediction method comprises the steps of obtaining latest passenger flow data and multi-source data of a target tourism area; judging whether the target tourism area is a first type tourism area or a second type tourism area; if the target tourism area is the first type of tourism area, a passenger flow prediction model corresponding to the target tourism area predicts the passenger flow of a specific date according to the passenger flow data and the multi-source data; and determining a target passenger flow volume prediction model with the highest matching degree from a passenger flow volume prediction model pool corresponding to the first type of tourism area if the tourism area is the second type of tourism area, and predicting the passenger flow volume by the target passenger flow volume prediction model according to the passenger flow data and the multi-source data. According to the method, the problems of insufficient support of single passenger flow data and poor prediction effect caused by insufficient characterization capability of historical data in a specific time period of partial target areas can be optimized, and holiday and festival passenger flow prediction accuracy of different types of areas can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular, to a method, device, electronic device, and storage medium for predicting the passenger flow in a tourist area. Background Art

[0002] In recent years, with the change of people's consumption concepts, the tourism industry has become increasingly booming. Whether it is traditional or emerging business districts and scenic spots, they have received an endless stream of tourists during holidays. This booming tourism situation has also brought many problems. On the one hand, in terms of reception capacity, the reception capacities of some business districts and scenic spots have exceeded their own carrying capacities, resulting in insufficient local infrastructure preparation and reception capacity, and then causing commercial or safety problems such as tourist congestion and inability to check in. On the other hand, whether emerging cities need to make advance planning and layout before the next holiday also requires relevant population predictions to provide strong support for their decision-making. In this context, accurately predicting the number of tourists in business districts and scenic spots in advance is of crucial significance for doing relevant preparatory work.

[0003] However, traditional passenger flow prediction technologies have limitations in using algorithms or subjectively giving prediction conclusions when the data types are single, and the prediction effect is poor. Especially for the impact of temporary peaks, extreme events, or external environments, it is difficult to accurately predict the peak value of passenger flow during future holidays. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method, device, electronic device, and storage medium for predicting the passenger flow in a tourist area, which is used to improve the accuracy of passenger flow prediction for different types of tourist areas. To achieve the above purpose, the technical solutions adopted in the embodiments of the present invention are as follows: In the first aspect, the present invention provides a method for predicting the passenger flow in a tourist area, and the method includes: obtaining the latest passenger flow data and multi-source data of the target tourist area; determining whether the target tourist area is a first type of tourist area or a second type of tourist area; if it is a first type of tourist area, predicting the passenger flow on a specific date by the passenger flow prediction model corresponding to the target tourist area according to the passenger flow data and multi-source data; if it is a second type of tourist area, determining the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first type of tourist area, and predicting the passenger flow by the target passenger flow prediction model according to the passenger flow data and multi-source data.

[0005] In a second aspect, the present invention provides a device for predicting regional passenger flow, including: an acquisition module, a judgment module, and a prediction module; the acquisition module is used to obtain the latest passenger flow data and multi-source data of the target tourist area; the judgment module is used to judge whether the target tourist area is a first-type tourist area or a second-type tourist area; the prediction module is used to, if the judgment result of the judgment module is a first-type tourist area, predict the passenger flow on a specific date according to the passenger flow data and multi-source data by the passenger flow prediction model corresponding to the target tourist area; the prediction module is further used to, if the judgment result of the judgment module is a second-type tourist area, determine the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first-type tourist area, and predict the passenger flow according to the passenger flow data and multi-source data by the target passenger flow prediction model.

[0006] In a third aspect, the present invention provides an electronic device, including: a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the regional passenger flow prediction as described in any one of the foregoing embodiments.

[0007] In a fourth aspect, the present invention provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for predicting tourist area passenger flow as described in any one of the foregoing embodiments.

[0008] The method, device, electronic device, and storage medium for predicting tourist area passenger flow provided by the embodiments of the present invention first obtain the latest passenger flow data and multi-source data of the target tourist area, and then judge whether the target research area is a first-type tourist area or a second-type tourist area. Since the corresponding passenger flow prediction model has been trained for the first-type tourist area, when the target tourist area is a first-type tourist area, the model can be directly called to predict the passenger flow on a specific date in combination with the passenger flow data and multi-source data, and the multi-source data is used to make up for the defects of single passenger flow data such as lack of support for external driving factors, irregularity of holidays, weak interpretability of extreme values and outliers, lack of instant response ability, limitations of periodic assumptions, cumulative effect of prediction inaccuracy, etc.; when the target tourist area is a second-type tourist area, the most matching model can be found from the various models of the first-type tourist area and migrated to the passenger flow prediction scenario of the target tourist area. Thus, the problem of insufficient support of single passenger flow data and poor prediction effect caused by insufficient representation ability of historical data in some target areas during a specific time period can be optimized, and the accuracy of predicting holiday passenger flow in different types of areas can be improved.

[0009] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following specifically gives preferred embodiments and cooperates with the attached drawings for detailed description as follows. Description of the Drawings

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following accompanying drawings only show certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related accompanying drawings can also be obtained based on these drawings.

[0011] Figure 1 It is a schematic flowchart of the tourism area passenger flow prediction method provided by the embodiments of the present invention; Figure 2 It is a schematic diagram of the tourism area passenger flow prediction process provided by the embodiments of the present invention; Figure 3 It is a functional module diagram of the tourism area passenger flow prediction device provided by the embodiments of the present invention; Figure 4 It is a structural block diagram of the electronic device provided by the embodiments of the present invention. Detailed Embodiments

[0012] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Usually, the components of the embodiments of the present invention described and shown in the accompanying drawings here can be arranged and designed in various different configurations.

[0013] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the present invention to be protected, but only represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0014] It should be noted that relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, the element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0015] Currently, when predicting the passenger flow of a certain tourist area, especially in business districts and scenic spots, two prediction methods are adopted. One is to use the number of people in the historical cycle of the tourist area as the prediction variable and use machine learning or deep learning methods for prediction; the other is to subjectively determine which day in the holiday period is likely to be the peak of passenger flow and the trend based on the historical experience of the fluctuation of historical holiday population data, so as to give the prediction result.

[0016] The inventor found during the research process that: First, the prediction of passenger flow that solely relies on population data fails to fully consider the influence of external driving factors, making it difficult to cope with the irregularity of holiday population flow, having a weak ability to explain extreme values and outliers, lacking sensitivity to immediate changes. At the same time, its prediction method based on periodic assumptions also has obvious limitations, which is prone to the accumulation of prediction errors. Second, in the process of modeling and prediction, the quality of historical data is the key factor determining the prediction effect. However, for emerging business districts and emerging scenic spots, due to the lack of historical population data of relevant holidays, it is almost impossible to effectively predict the passenger flow during holidays and thus unable to issue passenger flow warnings only based on the current population trend.

[0017] To solve the above problems, the embodiments of the present invention provide a method for predicting the passenger flow of a tourist area, which can make up for the defects of single population data and at the same time be applicable to the passenger flow prediction scenarios of various emerging business districts and scenic spots, effectively improving the accuracy of prediction.

[0018] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of the method for predicting the passenger flow of a tourist area provided by the embodiments of the present invention, including steps S101 to S104, described as follows: S101: Obtain the latest passenger flow data and multi-source data of the target tourist area; S102: Determine whether the target tourist area is a first-type tourist area or a second-type tourist area; S103: If it is a first-type tourist area, the passenger flow prediction model corresponding to the target tourist area predicts the passenger flow on a specific date according to the passenger flow data and multi-source data; S104: If it is a second-type tourist area, determine the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first-type tourist area, and the target passenger flow prediction model predicts the passenger flow according to the passenger flow data and multi-source data.

[0019] The solutions of steps S101 to S104 provided by the embodiments of the present invention first obtain the latest passenger flow data and multi-source data of the target tourist area, and then determine whether the target research area is a first-class tourist area or a second-class tourist area. Since the corresponding passenger flow prediction model has been trained for the first-class tourist area, when the target tourist area is a first-class tourist area, the model can be directly called to predict the passenger flow on a specific date by combining the passenger flow data and multi-source data. By integrating multi-source data, many defects in predicting with only single passenger flow data can be effectively made up for; when the target tourist area is a second-class tourist area, the most matching model can be found from the models of the first-class tourist areas and migrated to the passenger flow prediction scenario of the target tourist area. Thus, the problems of insufficient support of single passenger flow data and poor prediction effect caused by insufficient representation ability of historical data in some target areas during specific time periods can be optimized, and the prediction accuracy of holiday passenger flow in different types of areas can be improved.

[0020] Next, the embodiments of the present invention will introduce the above steps in detail and clearly.

[0021] In step S101, the target tourist area is a business district and scenic spots. For example, the target tourist area can be a shopping center, commercial street, shopping mall, traditional scenic spot, etc. The target tourist area can be set by data analysts according to actual research needs, and there is no limitation here. The passenger flow data refers to the passenger flow collected in the previous work. The multi-source data can include various data, namely regional climate data, urban rail transit ticketing data, urban air ticket ticketing data, related data of surrounding hotels, hot word data index of social platforms, and historical data of similar areas.

[0022] In the embodiments of the present invention, the regional climate data can include but is not limited to: temperature, moving average of air temperature, wind direction, wind force level, rainfall, rainfall probability, humidity, pollen concentration, etc.; the urban rail transit ticketing data can include but is not limited to: ticket issuing volume, ticket refund volume, sold-out rate, etc.; the urban air ticket ticketing data can include but is not limited to: price fluctuation index, ticket issuing volume, ticket refund volume, and sold-out rate, etc.; the related data of surrounding hotels refers to the related data of hotels within a certain distance centered on the target tourist area, including but not limited to: price fluctuation index, hotel occupancy rate, hotel reservation rate, etc.; the hot word data index of social platforms refers to a quantitative index that measures the popularity and attention of specific words or topics on social platforms after users frequently mention words or topics on social platforms within a certain period of time, including but not limited to: hot word search index, online platform, popularity quantification value of short video platforms, etc.; the historical data of similar areas, including passenger flow data and multi-source data, can be used as a reference input feature and can increase the generalization ability of the model to a certain extent.

[0023] It is understandable that the embodiments of the present invention can process data in a specific quantization manner according to the different characteristics of each type of data in multi-source data. For example, the pollen concentration index in regional climate data, the price fluctuation index in ticket data, and the platform hot word data index. Among them, the quantization method can be but is not limited to: moving average (weighted moving average, exponential moving average) index that highlights long-term trends, slope based on linear regression that reflects historical change patterns, and positive and cosine quantization periodic index that strengthens the capture of periodic changes in data, etc.

[0024] In step S101, the passenger flow data and multi-source data can be data of the most recent day or multiple days, and data analysts can flexibly set them according to actual needs. After the data collection is successful, in order to ensure the accuracy of the prediction results, these data can be preprocessed first.

[0025] Optionally, the preprocessing operations can include but are not limited to: data smoothing processing, outlier removal, data standardization, etc. Among them, the data standardization operation can cope with the sensitivity of different scales of the passenger flow prediction model used in the embodiments of the present invention, and contribute to the prediction stability of the model.

[0026] In addition, in order to adapt to the subsequent passenger flow prediction model, the embodiments of the present invention also pre-convert the obtained data into the form of a time series, and the data points in the series have a chronological order, and each data point is the passenger flow data and multi-source data.

[0027] Through the above implementation manners, the embodiments of the present invention can increase external driving factor support for the subsequent training of the passenger flow prediction model, enhance the interpretability of extreme values (if the passenger flow of a certain scenic area suddenly reaches an extremely high peak on a certain day, only the change in quantity may be seen from the passenger flow data itself, but the reason for the appearance of this peak cannot be explained), and weaken the continuous deviation of the prediction results caused by the cumulative effect of prediction inaccuracy.

[0028] Further, in step S102, the embodiments of the present invention first determine whether the target tourist area is a first type of tourist area or a second type of tourist area.

[0029] In the embodiments of the present invention, the first type of tourist area refers to a tourist area where the historical passenger flow data shows a periodic change pattern, such as traditional business districts and scenic spots with a certain number of years. The passenger flow in these areas usually shows regular fluctuations with specific time periods such as seasons, holidays, and weekends. For example, some famous historical and cultural scenic spots will receive a large number of tourists during the tourist peak season (such as summer or holidays), and will be relatively deserted during the off-season; the passenger flow in traditional business districts will also increase significantly during holidays. This periodic change pattern can be identified and verified through time series analysis of historical passenger flow data.

[0030] Then, the second type of tourist area refers to a tourist area where the historical passenger flow data does not show an obvious periodic change pattern, such as some emerging business districts and scenic spots with a short history. These areas may not have formed a stable tourist flow pattern due to their short development time, or their passenger flow is affected by a combination of various factors, making it difficult to identify a clear periodic pattern. For example, some emerging online celebrity check-in spots may suddenly become popular due to social media promotion, but the fluctuations in their passenger flow may be more affected by non-periodic factors such as online popularity and temporary events, rather than fixed seasons or holidays.

[0031] Among them, the historical passenger flow data showing a periodic change pattern means that when statistically analyzing the historical passenger flow data, it can be observed that the passenger flow shows a regular rising and falling pattern over time, and this pattern repeats in multiple time periods. This periodic change pattern can be quantitatively evaluated by calculating the periodic correlation coefficient, performing Fourier transform, or applying other time series analysis methods.

[0032] Therefore, for step S102, the judgment method given in the embodiments of the present invention may include steps a1 to a2: Step a1: Statistically analyze the passenger flow data to determine whether there is a periodic change pattern; Step a2: If it exists, determine that the target tourist area is the first type of tourist area; otherwise, it is the second type of tourist area.

[0033] In the embodiments of the present invention, combined with the periodic change pattern presented by the historical passenger flow data of the first type of tourist area, the embodiments of the present invention can pre-train a machine learning model for predicting the passenger flow for the first type of tourist area to improve the efficiency and accuracy of passenger flow prediction in the first type of tourist area. The following introduces the training process of the passenger flow prediction model.

[0034] For each first type of tourist area, the model can be trained in the following manner: Step b1: Obtain the historical passenger flow data and historical multi-source data of each first type of tourist area within a preset historical time period.

[0035] Among them, the historical time period can be flexibly set by data analysts. For example, obtain the historical passenger flow and historical multi-source data in the past three years, or obtain all the accumulated historical passenger flow data and historical source data, including historical data on weekdays, weekends, holidays, etc. For all historical passenger flows, they can also be aggregated according to different time granularities. For example, if the time granularity is weekly or monthly, the historical passenger flow in a week or a month can be aggregated, which can reduce the amount of data and improve the prediction speed.

[0036] In the embodiments of the present invention, for each historical passenger flow volume, it is also possible to determine whether the corresponding historical timestamp is a holiday. If it is a holiday, a holiday flag is configured for the historical passenger flow volume at this historical timestamp. It is also possible to further determine the type of holiday, and use a historical passenger flow volume and its corresponding holiday flag and holiday type as a piece of passenger flow data. In addition, in order to adapt to the passenger flow volume prediction model, all data can be converted into time series data in chronological order of history. The data points in this sequence are arranged in chronological order, and each data point has a clear historical timestamp. The information contained in each data point includes passenger flow volume, multi-source data indicators, holiday flag, and holiday type.

[0037] Step b2: Construct an input time series and a label series according to the historical passenger flow data and historical multi-source data according to a preset time window and prediction step length; Step b3: Train the initial machine learning model according to the input time series and the label series until the convergence condition is reached to obtain a passenger flow volume prediction model.

[0038] The machine learning model used in the embodiments of the present invention is a Long Short-Term Memory (LSTM) algorithm model. Therefore, it is necessary to first convert the historical passenger flow data and historical multi-source data into the format required for model training. Before the format conversion, the historical data obtained in step b1 can be preprocessed, such as data cleaning, data normalization, etc., and then an input time series is constructed.

[0039] For the LSTM algorithm model, the converted time series can be divided according to a preset time window, and each time window contains a fixed number of time steps. For example, assume there is the following price time series data: [100.5, 101.2, 100.8, 102.0, 101.5, 103.0, 102.8, 104.0, 103.5, 105.0]. If the time window is set to 3, then the constructed input time series is: [[100.5, 101.2, 100.8], [101.2, 100.8, 102.0], [100.8, 102.0, 101.5], [102.0, 101.5, 103.0], [101.5, 103.0, 102.8], [103.0, 102.8, 104.0], [102.8, 104.0, 103.5]]. It can be seen that each input time series contains price data for 3 days.

[0040] In order to enable the model to predict the passenger flow during the training process, a label sequence can be constructed for each input time series according to the prediction step length. The label is the value of one or more future time steps of the input time series, which specifically depends on the set prediction step length. Continuing with the above input time series as an example, for [100.5, 101.2, 100.8], two days later, taking the input sequence [100.5, 101.2, 100.8] as an example, its corresponding label sequence is the price data corresponding to the 4th and 5th in the price time series data, [102.0, 101.5].

[0041] The initial LSTM algorithm model is trained with the constructed input time series and label sequence. After the model training is completed and the prediction step length to be predicted is input, the model can output the passenger flow prediction value corresponding to the number of days of the required step length. During the training process, the convergence condition can be that the number of iterations reaches the preset number, or the loss function of the model converges, etc.

[0042] Through the above method, the embodiments of the present invention use the historical passenger flow data of traditional business district scenic spots combined with multi-source data as the eigenvalue for model training. The trained LSTM algorithm model can effectively realize the prediction of the passenger flow of traditional business district scenic spots during holidays.

[0043] In step S103, if the target tourist area is a first-class tourist area, the passenger flow prediction model corresponding to the target tourist area predicts the passenger flow on a specific date according to the passenger flow data and multi-source data.

[0044] When using the passenger flow prediction model for prediction, the latest passenger flow data and multi-source data can be used to construct an input time series according to the conversion method of historical data during the training process, and input it into the trained passenger flow prediction model, so as to realize the accurate prediction of the passenger flow. For example, if a data analyst wants to predict the passenger flow on a certain holiday, the prediction step length can be set according to the specific situation of the holiday and used as a parameter to input into the model. The model will output the corresponding passenger flow prediction according to the set prediction step length, so as to obtain the passenger flow prediction result of the holiday.

[0045] Furthermore, the embodiments of the present invention can also cache the passenger flow prediction models corresponding to each first-class tourist area in the model pool, which can facilitate the quick call of the model to predict the passenger flow of the first-class tourist area, and at the same time provide a model basis for the second-class tourist area.

[0046] To improve the accuracy and adaptability of the model, for each passenger flow prediction model in the model pool, the model can also be updated according to the updated passenger flow data and multi-source data corresponding to the first type of tourist area, so that the model can better adapt to the dynamic changes of the tourist area and ensure that it can still maintain a high prediction accuracy when facing new data and situations. For the second type of tourist area, such as some emerging business district scenic spots, since their historical passenger flow data does not have a periodic change pattern and lacks historical reference data, it is impossible to establish a passenger flow prediction model for them in advance. For this reason, the embodiment of the present invention provides a new idea, that is, model migration, which refers to migrating the passenger flow prediction model of the first type of tourist area to the second type of tourist area for passenger flow prediction, that is, performing step S104 to determine the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first type of tourist area.

[0047] To accurately find the passenger flow prediction model with the highest matching degree, the embodiment of the present invention provides the following implementation manners, including step c1 to step c3: Step c1: Obtain the latest passenger flow data and multi-source data of each first type of tourist area; It can be understood that these data of the first type of tourist area and the data of the target tourist area in step S101 belong to the same period. For example, in step S101, the data obtained is other data in the past until today, then step c1 also collects data within this time period.

[0048] Step c2: Determine the weight of each data index in the multi-source data of the first type of tourist area, the correlation degree and similarity degree between the same data indexes of the target tourist area and the first type of tourist area.

[0049] In the embodiment of the present invention, the different data indexes included in the multi-source data have been introduced before. Therefore, for each data index, the positive correlation degree between the time series data of each data index and the corresponding time series data of the passenger flow data can be used as the weight corresponding to each data index. It can be understood as follows: for each data index, first organize it into a time series in chronological order. Similarly, the passenger flow data also needs to be organized into a time series, and then calculate the positive correlation degree of the linear correlation of these two time series.

[0050] For the target tourist area and the first type of tourist area, extract the same type of data indexes from their multi-source data for correlation degree and similarity degree calculation. For example, calculate the correlation degree and similarity degree between the temperature time series of the target tourist area and the first type of tourist area within the past 7 days.

[0051] In the embodiments of the present invention, the relevance can, but is not limited to, using the Pearson correlation coefficient or other correlation quantification methods. The similarity is calculated using the Dynamic Time Warping (DTW) algorithm, that is, the distance between the time series of the same data metrics of the target tourist area and the first type of tourist area. Assuming that the time series of temperature data of the target tourist area and the first type of tourist area are denoted as A and B, then the similarity can be expressed as where D[n][m] is the last element in the distance matrix D, representing the DTW distance between sequences A and B; where n and m are the lengths of sequences A and B respectively; each element D[i][j] of the matrix D represents the distance between the first i elements of sequence A and the first j elements of sequence B.

[0052] Step c3: Determine the scores of the first type of tourist areas according to the weights, relevance, and similarity of each data metric; In the embodiments of the present invention, the scores satisfy the following relationship:

[0053] where, and respectively represent the weight, relevance, and similarity corresponding to the i-th data metric; n represents the type of data metrics.

[0054] In the above implementation, the reason for comprehensively using relevance and DTW similarity is that correlation can be used to measure the strength and direction of the linear relationship between pairs, and different dimensions can also be considered, while DTW similarity focuses on the overall shape and dynamic changes of the time series. Combining the scores of the two can provide a more comprehensive evaluation perspective.

[0055] Step c4: Use the passenger flow prediction model of the first type of tourist area with the highest score as the target passenger flow prediction model.

[0056] In the embodiments of the present invention, for the target passenger flow prediction model with the highest matching degree, the model can be directly loaded for prediction, or the model can be fine-tuned and retrained. Among them, during the fine-tuning process, a new fully connected layer can be added to the model structure, or partial frozen LSTM layers can be fine-tuned (fine-tuning, the underlying convolutional layers can be frozen and the upper layers can be fine-tuned). This process helps the performance of the new model on the data of the second type of tourist area. Although the second type of tourist area does not have historical passenger flow data showing obvious periodic change patterns, due to the strong similarity between the multi-source feature data and the first type of tourist area, adjustments are made based on the knowledge learned by the model before, gradually adapting to the data pattern of the second type of tourist area.

[0057] In the application stage, it is first necessary to load the trained target passenger flow prediction model. Subsequently, based on the latest passenger flow data and multi-source data obtained in step S101, an input time series is constructed. At the same time, the prediction step is adjusted according to the set specific date (such as a holiday), and the constructed input time series is input into the target passenger flow prediction model. Through this process, the target passenger flow prediction model outputs the corresponding passenger flow prediction results according to the specified prediction step. Here, the specific date can be a certain date specified by the data analyst, especially a time period with special passenger flow patterns, such as holidays or other important event days.

[0058] For an overall understanding of the above-mentioned tourist area passenger flow prediction process, please refer to Figure 2 , Figure 2 which is the schematic diagram of the tourist area passenger flow prediction process provided by the embodiment of the present invention. Combining Figure 2 with the above-mentioned implementation manners, it can be summarized that: First, the embodiment of the present invention provides an idea of training an LSTM model by combining passenger flow data and multi-source data, which can accurately complete the passenger flow prediction task of business district scenic spots on specific dates by using the trained model. By introducing multi-source data, this method effectively makes up for many defects of single signaling population data, such as lack of support for external driving factors, irregularity of holidays, weak interpretability of extreme values and outliers, lack of instant response ability, limitations of periodic assumptions, and cumulative effects of prediction inaccuracies. Second, the embodiment of the present invention provides a model migration idea for emerging business district scenic spots, which can solve the problem that it is difficult to predict emerging business district scenic spots due to lack of historical data. In summary, the embodiment of the present invention not only realizes the passenger flow prediction of business districts and scenic spots during the holiday period, but also significantly improves the prediction accuracy.

[0059] In order to execute the corresponding steps in the above embodiments and various possible manners, an implementation manner of a tourist area passenger flow prediction device is given below. Please refer to Figure 3 , Figure 3 which is the functional module diagram of the tourist area passenger flow prediction device provided by the embodiment of the present invention. It should be noted that the basic principle and the technical effects generated by the tourist area passenger flow prediction device provided in this embodiment are the same as those in the above embodiments. For the sake of brief description, for the parts not mentioned in this embodiment, reference can be made to the corresponding content in the above embodiments. The tourist area passenger flow prediction device 30 includes: an acquisition module 301, a judgment module 302, and a prediction module 303.

[0060] The acquisition module 301 is used to obtain the latest passenger flow data and multi-source data of the target tourist area; The judgment module 302 is used to judge whether the target tourist area is a first-class tourist area or a second-class tourist area; A prediction module 303, configured to, if the judgment result of the judgment module 302 is a first type of tourist area, predict the passenger flow of a specific date according to the passenger flow data and multi-source data by using the passenger flow prediction model corresponding to the target tourist area; The prediction module 303 is further configured to, if the judgment result of the judgment module 302 is a second type of tourist area, determine the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first type of tourist area, and predict the passenger flow according to the passenger flow data and multi-source data by using the target passenger flow prediction model.

[0061] It can be understood that the obtaining module 301, the judgment module 302, and the prediction module 303 can cooperate in each step of the embodiments of the present invention to achieve corresponding technical effects. Details are not described herein again.

[0062] It should be noted that the division of modules in the above embodiments of the present application is illustrative, and is only a logical function division. In actual implementation, there may be other division methods. In addition, in each embodiment of the present application, each functional unit may be integrated into one processing unit, may exist separately physically, or two or more units may be integrated into one unit. The above integrated units may be implemented in the form of hardware or in the form of software functional units.

[0063] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing an electronic device (which may be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods in the various embodiments of the present application. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program executable code.

[0064] The embodiments of the present invention further provide an electronic device. Please refer to Figure 4 , Figure 4 which is a structural block diagram of the electronic device provided by the embodiments of the present invention, including: a memory 401, a processor 402, and a communication interface 403. The memory 401, the processor 402, and the communication interface 403 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these components may be electrically connected to each other through one or more communication buses or signal lines.

[0065] Optionally, the bus can be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0066] In an embodiment of the present invention, the processor 402 can be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field-programmable gate array, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware processor, or can be executed and completed by a combination of hardware and software modules in the processor. The software module can be located in the memory 401, and the processor 402 reads the program instructions in the memory 401 and combines its hardware to complete the steps of the above method.

[0067] In an embodiment of the present invention, the memory 401 can be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), etc., and can also be a volatile memory, such as RAM. The memory can also be any other medium that can be used to carry or store the desired program executable code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiments of the present invention can also be a circuit or any other device capable of implementing a storage function, for storing instructions and / or data.

[0068] The memory 401 can be used to store software programs and modules, such as the instructions / modules of the tourist area passenger flow prediction device 30 provided in the embodiments of the present invention, and can be stored in the memory 401 in the form of software or firmware, or can be solidified in the operating system (OS) of the electronic device 40. The processor 402 executes the software programs and modules stored in the memory 401, thereby performing various functional applications and data processing. The communication interface 403 can be used for signaling or data communication with other node devices.

[0069] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the devices and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0070] It can be understood that Figure 4 The structure shown is only schematic, and the electronic device 40 may further include more or fewer components than those shown in Figure 4 or have a different configuration from that shown in Figure 4 shown. Figure 4 Each component shown can be implemented by hardware, software, or a combination thereof.

[0071] Based on the above embodiments, the present application also provides a storage medium. A computer program is stored in the computer storage medium. When the computer program is executed by a computer, the computer is caused to execute the tourism area passenger flow prediction method provided by the above embodiments.

[0072] Based on the above embodiments, the embodiments of the present invention also provide a computer program. When the computer program runs on a computer, the computer is caused to execute the tourism area passenger flow prediction method provided by the above embodiments.

[0073] Based on the above embodiments, the embodiments of the present invention also provide a chip. The chip is used to read the computer program stored in the memory and is used to execute the tourism area passenger flow prediction method provided by the above embodiments.

[0074] The embodiments of the present invention also provide a computer program product, including instructions. When the instructions run on a computer, the computer is caused to execute the tourism area passenger flow prediction method provided by the above embodiments.

[0075] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by instructions. These instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0076] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions in the processFigure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0077] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0078] The above is only a specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for predicting passenger flow in a tourist area, characterized in that: The method comprises: Obtain the latest passenger flow data and multi-source data in the target tourist area; Determining whether the target tourist area is a first-category tourist area or a second-category tourist area; If it is a first type of tourist area, the passenger flow prediction model corresponding to the target tourist area predicts the passenger flow on a specific date based on the passenger flow data and multi-source data; If it is a second-category tourist area, the target passenger flow prediction model with the highest matching degree is determined from the passenger flow prediction model pool corresponding to the first-category tourist area, and the target passenger flow prediction model predicts the passenger flow based on the passenger flow data and multi-source data.

2. The method for predicting passenger flow in a tourist area according to claim 1, characterized in that: Determining a target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first type of tourist area includes: Obtain the latest passenger flow data and multi-source data for each of the first-category tourist areas; Determine the weight of each data indicator in the multi-source data of the first type of tourist area, and the correlation and similarity between the same data indicators of the target tourist area and the first type of tourist area; Determine the score of the first type of tourist area according to the weight of each of the data indicators, the relevance and the similarity; The passenger flow prediction model of the first type of tourist area with the highest score is used as the target passenger flow prediction model.

3. The method for predicting passenger flow in a tourist area according to claim 2, characterized in that: The score satisfies the following relationship: in, and They respectively represent the weight, relevance and similarity corresponding to the i-th data indicator; n represents the type of data indicator.

4. The method for predicting passenger flow in a tourist area according to claim 2, characterized in that: Determine the weight of each data indicator in the multi-source data of the first type of tourist area, including: The degree of positive correlation between the time series data of each of the data indicators and the time series data corresponding to the passenger flow data is used as the weight corresponding to each of the data indicators.

5. The method for predicting passenger flow in a tourist area according to any one of claims 1 to 4, characterized in that: Determining whether the target tourist area is a first-category tourist area or a second-category tourist area includes: Conduct statistical analysis on the historical passenger flow data of the target tourist area to determine whether there is a periodic change pattern; If so, the target tourist area is determined to be the first type of tourist area; otherwise, it is determined to be the second type of tourist area.

6. A tourist area passenger flow prediction device, characterized in that: include: Acquisition module, judgment module and prediction module; The acquisition module is used to obtain the latest passenger flow data and multi-source data of the target tourist area; The judging module is used to judge whether the target tourist area is a first-category tourist area or a second-category tourist area; The prediction module is used to predict the passenger flow on a specific date based on the passenger flow data and multi-source data by using the passenger flow prediction model corresponding to the target tourist area if the judgment result of the judgment module is the first type of tourist area; The prediction module is also used to determine the target passenger flow prediction model with the highest matching degree from the passenger flow prediction model pool corresponding to the first type of tourist area if the judgment result of the judgment module is the second type of tourist area, and the target passenger flow prediction model predicts the passenger flow based on the passenger flow data and multi-source data.

7. The tourist area passenger flow prediction device according to claim 6, characterized in that: The prediction module is specifically used for: Obtain the latest passenger flow data and multi-source data for each of the first-category tourist areas; Determine the weight of each data indicator in the multi-source data of the first type of tourist area, and the correlation and similarity between the same data indicators of the target tourist area and the first type of tourist area; Determine the score of the first type of tourist area according to the weight of each of the data indicators, the relevance and the similarity; The passenger flow prediction model of the first type of tourist area with the highest score is used as the target passenger flow prediction model.

8. The tourist area passenger flow prediction device according to claim 7, characterized in that: The score satisfies the following relationship: in, and They respectively represent the weight, relevance and similarity corresponding to the i-th data indicator; n represents the type of data indicator.

9. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, the method for predicting passenger flow in a tourist area as claimed in any one of claims 1 to 5 is implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for predicting passenger flow in a tourist area as described in any one of claims 1 to 5 is implemented.

Images