Digital platform passenger behavior analysis and personalized service pushing method

Abstract

The invention relates to a digital platform passenger behavior analysis and personalized service pushing method, which comprises the following steps that: L1, a passenger enters a digital platform, data information of a historical travel track of the passenger is acquired, image data information of the passenger is acquired in real time based on a built-in camera of the platform, and the image data information of the passenger is acquired; the method comprises the following steps: acquiring data information of a position coordinate sequence of a passenger in real time based on a built-in infrared sensor of a platform, acquiring data information of a pressure change time sequence of the platform in real time based on a built-in pressure sensor of the platform, constructing a skeleton key point detection model based on a dynamic learning factor, and representing a limb action feature matrix of the passenger; and obtaining data information of the limb movement feature matrix of the passenger. According to the invention, not only can information such as behaviors, tracks and demands of passengers and pedestrians at the station be accurately collected and analyzed, but also services such as vehicle congestion degree informing and personalized riding suggestion pushing can be provided in advance, so that the operation efficiency, the service quality and the passenger travel experience of the unmanned bus are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of driverless buses, and in particular to a method for analyzing passenger behavior at a digital platform and pushing personalized services. Background Art

[0002] With the rapid development of technology, intelligent buses, as an important part of intelligent transportation, remain a major part of urban transportation. However, there are still many deficiencies in the current bus system in terms of passenger services. Traditional bus stops have a single function and mainly provide simple route information displays for passengers. They cannot effectively analyze passenger behavior and are also difficult to provide personalized services according to individual passenger needs.

[0003] Therefore, under the existing bus operation mode, buses cannot accurately grasp information such as passengers' travel habits and waiting preferences. For example, behavioral characteristics such as passengers' waiting time at the stop, whether they carry luggage, and whether they have accompanying persons are not fully utilized. This leads to a lack of pertinence in bus operation scheduling, and buses cannot reasonably arrange vehicle resources according to actual passenger needs, thus affecting the passenger travel experience and reducing the overall operation efficiency of the bus system.

[0004] At the same time, due to the lack of a personalized service push mechanism, passengers cannot obtain information related to their own needs when taking the bus, such as surrounding scene recommendations, real-time traffic condition reminders, bus transfer suggestions, etc. This makes the bus system have a large gap from modern passengers' expectations in terms of service diversity and user stickiness.

[0005] In the prior art, Chinese Patent (Application No.: 202411972313.3, Publication No.: CN 119848349A) discloses a method and system for processing data of an intelligent bus waiting pavilion based on digital analysis, including collecting seat usage information and air-conditioning usage status in the target vehicle, calculating temperature distribution information and crowding degree information of different seats and sending them to all waiting pavilions for data sharing; receiving a request instruction from the target passenger, identifying the target passenger's historical riding records and current riding demand information, predicting riding preference behavior and the expected distance from the desired seat to the disembarkation point; constructing a seat recommendation model based on the temperature distribution information, crowding degree information, riding preference behavior and expected distance, generating a target seat for recommendation, and when the target passenger determines to use the target seat, updating the data and synchronizing it to all waiting pavilions to continue recommending seats for the next passenger. In this solution, the data accuracy depends on devices: the collection of data such as seat usage information and air-conditioning usage status depends on the accuracy of in-vehicle devices. If the devices malfunction or have insufficient accuracy, it may lead to incorrect calculation of temperature distribution and crowding degree information, thereby affecting the accuracy of seat recommendation.

[0006] Limitations of passenger behavior prediction: Although passenger preference behaviors are predicted based on historical riding records and current demands, passenger travel demands may change temporarily due to special circumstances. Predictions based only on partial historical and current information are difficult to cover all complex and changing actual situations, resulting in deviations between the recommended seats and the actual demands of passengers. System complexity and maintenance costs: It involves complex operations such as data interaction between vehicles and multiple waiting pavilions, various information analyses, and model construction. The construction and maintenance of the system require high costs, including equipment maintenance, data management, and investment in technical personnel, etc.

[0007] In the prior art, a platform passenger detection system and method for a driverless bus are disclosed in a Chinese patent (application number: 201810938571.8, publication number: CN 108830264A), which includes an on-vehicle computing terminal, an external front camera of the vehicle, an external rear camera of the vehicle, an external left camera of the vehicle, an external right camera of the vehicle, a display, a GPS global positioning module, a 4G network communication module, and a PC remote monitoring center; the on-vehicle computing terminal includes an ARM controller, a DSP image stitching module, a DSP image detection module, a GPU module, and a CAN communication module; the DSP image stitching module is connected to the ARM controller and is used for real-time processing of the videos of the 4 external cameras of the vehicle and stitching them into a panoramic video; the DSP image detection module is connected to the ARM controller and is used for real-time processing of the video of the front camera inside the vehicle and identifying and detecting the passengers on the platform of the driverless bus so as to automatically park the vehicle; the 4G network communication module is connected to the ARM controller; the 4 external camera modules are used for collecting real-time videos around the vehicle. The system of this solution highly depends on the data collected by the external cameras of the vehicle. In case of bad weather (such as heavy rain, heavy snow, thick fog, etc.), the field of view of the cameras may be affected, resulting in unclear image acquisition, which in turn affects the accuracy of passenger detection and the normal operation of the system. The 4G network communication module is used to transmit information to the PC remote monitoring center. When the network signal is unstable or there is network congestion, data transmission delay may occur, affecting the real-time nature of remote monitoring and the timely control of the vehicle. Summary of the Invention

[0008] In view of the above problems, the present invention provides a digital platform passenger behavior analysis and personalized service push method, which can not only accurately collect and analyze information such as the behaviors, trajectories, and demands of passengers and pedestrians on the platform, but also inform in advance about the vehicle congestion degree and push personalized riding suggestions and other services, thereby improving the operation efficiency, service quality, and passenger travel experience of driverless buses.

[0009] In order to achieve the above object and other related objects, the technical solution provided by the present invention is as follows: A digital platform passenger behavior analysis and personalized service push method, the method includes: L1. When a passenger enters the digital platform, data information on the passenger's historical travel trajectory is obtained. Based on the cameras installed in the platform, image data information of the passenger is obtained in real time. Based on the infrared sensors installed in the platform, data information on the sequence of the passenger's position coordinates is obtained in real time. Based on the pressure sensors installed in the platform, data information on the time series of the pressure changes on the platform is obtained in real time; L2. Based on the image data information of the passenger, a skeletal key point detection model based on a dynamic learning factor is constructed to represent the passenger's limb movement feature matrix, and data information on the passenger's limb movement feature matrix is obtained; L3. Based on the data information on the passenger's limb movement feature matrix, the data information on the sequence of the passenger's position coordinates, and the data information on the time series of the pressure changes on the platform, a BP neural network regression prediction algorithm based on the Pearson correlation coefficient is used to predict the travel time of the passenger at the platform, and data information on the predicted travel time of the passenger at the platform is obtained; L4. Based on the data information on the predicted travel time of the passenger at the platform and the data information on the passenger's historical travel trajectory, a push service model for the passenger's travel is constructed, corresponding personalized push strategies are formulated, and they are pushed to the mobile APP or WeChat mini-program of the platform passengers through offline push.

[0010] Further, in step L2, the construction of the skeletal key point detection model based on a dynamic learning factor to represent the passenger's limb movement feature matrix includes: L21. The image data information of the passenger is input into a convolutional neural network model to extract the passenger's image features, and data information on the passenger's image feature matrix is obtained; L22. Based on the data information on the passenger's image feature matrix, a passenger skeletal key point extraction function Q is established, , where x is the data information on the passenger's image feature matrix, and α, β, and δ are weight coefficients, to extract the passenger's skeletal key points and obtain data information on the passenger's skeletal key point coordinates; L23. Based on the data information on the passenger's skeletal key point coordinates, a passenger limb movement representation function W is constructed, , where y is the data information on the passenger's skeletal key point coordinates, and γ1, γ2, and γ3 are dynamic learning factors, to represent the passenger's limb movement feature matrix and obtain data information on the passenger's limb movement feature matrix.

[0011] Further, the dynamic learning factors γ1, γ2, and γ3 are , , , Among them, y is the data information of the skeletal key point coordinates of the passenger.

[0012] Further, in step L3, the prediction of the travel time of the platform where the passenger is located by using the BP neural network regression prediction algorithm based on the Pearson correlation coefficient includes: L31. Based on the data information of the position coordinate sequence of the passenger and the data information of the pressure change time sequence of the platform, establish the Pearson correlation coefficient function R of two variables, , Among them, X is the data information of the position coordinate sequence of the passenger, Y is the data information of the pressure change time sequence of the platform, which characterizes the relationship between the two variables, and obtains the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform; L32. Input the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform and the data information of the limb movement feature matrix of the passenger into the BP neural network regression prediction model for training and learning to obtain the trained BP neural network regression prediction model; L33. Based on the trained BP neural network regression prediction model, input the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform and the data information of the limb movement feature matrix of the passenger to predict the travel time of the platform where the passenger is located, and obtain the data information of the predicted travel time of the platform where the passenger is located.

[0013] Further, the BP neural network regression prediction model includes an input layer, a hidden layer and an output layer, and the neuron function P of the hidden layer is , Among them, r1 is the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform, r2 is the data information of the limb movement feature matrix of the passenger, and ω1, ω2 and ω3 are the weight factors of the BP neural network.

[0014] Further, the constraint conditions of the weight factors ω1, ω2 and ω3 of the BP neural network are .

[0015] Further, in step L4, the construction of the push service model for passenger travel and the formulation of corresponding personalized push strategies include: L41. Normalize the data information of the predicted travel time of the passenger at the platform and the data information of the passenger's historical travel trajectory to obtain the data information of the travel time of the passenger at the platform and the historical travel trajectory of the passenger after normalization processing; L42. Based on the data information of the travel time of the passenger at the platform and the data information of the passenger's historical travel trajectory after normalization processing, establish the objective optimization function G of the passenger travel trajectory, , where h1 is the data information of the travel time of the passenger at the platform after normalization processing, h2 is the data information of the passenger's historical travel trajectory after normalization processing, and ρ1, ρ2, and ρ3 are penalty factors; L43. Optimize the passenger's travel trajectory based on the objective optimization function G of the passenger travel trajectory to obtain the data information of the optimized passenger travel trajectory, thereby formulating corresponding personalized push strategies.

[0016] Further, the corresponding personalized push strategies include push strategies based on platform actions, push strategies based on travel habits, push strategies based on real-time scenarios, and push strategies based on interest preferences. The penalty factors ρ1, ρ2, and ρ3 are, , , , where h1 is the data information of the travel time of the passenger at the platform after normalization processing, and h2 is the data information of the passenger's historical travel trajectory after normalization processing.

[0017] To achieve the above and other related purposes, the present invention also provides a digital platform passenger behavior analysis and personalized service push system, including a computer device, which is programmed or configured to execute the steps of any one of the digital platform passenger behavior analysis and personalized service push methods.

[0018] To achieve the above and other related purposes, the present invention also provides a computer-readable storage medium, on which a computer program is stored that is programmed or configured to execute the steps of any one of the digital platform passenger behavior analysis and personalized service push methods.

[0019] The present invention has the following positive effects: 1. The present invention constructs a complete digital platform passenger behavior analysis and personalized service push system for driverless buses, which covers the digital platform perception layer, data transmission layer, data processing center, and personalized service push module. These layers work together to achieve the full-process function from passenger behavior data collection to personalized service push.

[0020] 2. The present invention adopts a multi-sensor fusion technology, and through devices such as cameras, infrared sensors, and pressure sensors, comprehensively collects the behavior data of passengers on the platform, providing a rich data source for in-depth analysis of passenger behavior.

[0021] 3. The present invention predicts the departure time of passengers at the platform by using a BP neural network regression prediction algorithm based on the Pearson correlation coefficient, and combines the construction of a skeleton key point detection model based on a dynamic learning factor to characterize the limb movement feature matrix of passengers. It not only formulates a personalized service push strategy based on platform actions, travel habits, real-time scenarios, and interest preferences, but also can provide diverse and personalized service content according to the different needs and actual situations of passengers. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a schematic diagram of the push scenario of the present invention; Figure 2 is a schematic diagram of the method flow of the present invention; Figure 3 is a schematic diagram of the construction of a skeleton key point detection model based on a dynamic learning factor of the present invention; Figure 4 is a schematic diagram of the flow of the BP neural network regression prediction algorithm based on the Pearson correlation coefficient of the present invention; Figure 5 is a schematic diagram of the flow of the construction of a push service model for passenger travel of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] The following describes exemplary embodiments of the present disclosure, including various details of the embodiments of the present disclosure to facilitate understanding. They should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for clarity and conciseness, the description below omits the description of well-known functions and structures.

[0024] Embodiment 1: As Figure 1 or Figure 2 shown, a method for digital platform passenger behavior analysis and personalized service push, the method includes: L1. When a passenger enters the digital platform, data information on the passenger's historical travel trajectory is obtained. Based on the in-platform built-in camera, image data information of the passenger is obtained in real time. Based on the in-platform built-in infrared sensor, data information on the sequence of the passenger's position coordinates is obtained in real time. Based on the in-platform built-in pressure sensor, data information on the time series of the pressure change on the platform is obtained in real time; L2. Based on the image data information of the passenger, a bone key-point detection model based on a dynamic learning factor is constructed to characterize the limb movement feature matrix of the passenger, and data information on the limb movement feature matrix of the passenger is obtained; L3. Based on the data information on the limb movement feature matrix of the passenger, the data information on the sequence of the passenger's position coordinates, and the data information on the time series of the pressure change on the platform, a BP neural network regression prediction algorithm based on the Pearson correlation coefficient is used to predict the travel time of the passenger at the platform, and data information on the predicted travel time of the passenger at the platform is obtained; L4. Based on the data information on the predicted travel time of the passenger at the platform and the data information on the passenger's historical travel trajectory, a push service model for the passenger's travel is constructed, a corresponding personalized push strategy is formulated, and it is pushed to the passenger's mobile APP or WeChat mini-program at the platform through offline push.

[0025] In this embodiment, as Figure 3 shown, in step L2, the construction of the bone key-point detection model based on a dynamic learning factor to characterize the limb movement feature matrix of the passenger includes: L21. The image data information of the passenger is input into a convolutional neural network model to extract the image features of the passenger, and data information on the image feature matrix of the passenger is obtained; L22. Based on the data information on the image feature matrix of the passenger, a bone key-point extraction function Q of the passenger is established, , where x is the data information on the image feature matrix of the passenger, and α, β, and δ are weight coefficients, to extract the bone key-points of the passenger, and data information on the bone key-point coordinates of the passenger is obtained; L23. Based on the data information on the bone key-point coordinates of the passenger, a limb movement characterization function W of the passenger is constructed, , where y is the data information on the bone key-point coordinates of the passenger, and γ1, γ2, and γ3 are dynamic learning factors, to characterize the limb movement feature matrix of the passenger, and data information on the limb movement feature matrix of the passenger is obtained.

[0026] In this embodiment, the dynamic learning factors γ1, γ2, and γ3 are , , , Among them, y is the data information of the coordinates of the passenger's skeletal key points.

[0027] In this embodiment, as Figure 4 shown, in step L3, the prediction of the travel time of the platform where the passenger is located by using the BP neural network regression prediction algorithm based on the Pearson correlation coefficient includes: L31. Based on the data information of the position coordinate sequence of the passenger and the data information of the pressure change time sequence of the platform, establish the Pearson correlation coefficient function R of the two variables, , where X is the data information of the position coordinate sequence of the passenger, Y is the data information of the pressure change time sequence of the platform, and the relationship between the two variables is characterized to obtain the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform; L32. Input the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform and the data information of the limb movement feature matrix of the passenger into the BP neural network regression prediction model for training and learning to obtain a trained BP neural network regression prediction model; L33. Based on the trained BP neural network regression prediction model, input the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform and the data information of the limb movement feature matrix of the passenger to predict the travel time of the platform where the passenger is located, and obtain the data information of the predicted travel time of the platform where the passenger is located.

[0028] In this embodiment, the BP neural network regression prediction model includes an input layer, a hidden layer, and an output layer. The neuron function P of the hidden layer is , where r1 is the data information of the relationship matrix between the position coordinate of the passenger and the pressure change of the platform, r2 is the data information of the limb movement feature matrix of the passenger, and ω1, ω2, and ω3 are the weight factors of the BP neural network.

[0029] In this embodiment, the constraint conditions for the weight factors ω1, ω2, and ω3 of the BP neural network are .

[0030] Embodiment 2: On the basis of the method for analyzing the behavior of passengers on a digital platform and pushing personalized services in Embodiment 1, the present invention will be further described and explained below.

[0031] As Figure 1 or Figure 2 shown, a method for analyzing passengers' behavior on a digital platform and pushing personalized services, the method comprising: L1. When a passenger enters the digital platform, data information on the passenger's historical travel trajectory is obtained, image data information of the passenger is obtained in real time based on the built-in cameras on the platform, data information on the sequence of position coordinates of the passenger is obtained in real time based on the built-in infrared sensors on the platform, and data information on the time series of pressure changes on the platform is obtained in real time based on the built-in pressure sensors on the platform; L2. Based on the image data information of the passenger, a skeletal key point detection model based on a dynamic learning factor is constructed to characterize the limb movement feature matrix of the passenger, and data information on the limb movement feature matrix of the passenger is obtained; L3. Based on the data information of the limb movement feature matrix of the passenger, the data information of the sequence of position coordinates of the passenger, and the data information of the time series of pressure changes on the platform, a BP neural network regression prediction algorithm based on the Pearson correlation coefficient is used to predict the travel time of the passenger on the platform, and data information on the predicted travel time of the passenger on the platform is obtained; L4. Based on the data information of the predicted travel time of the passenger on the platform and the data information of the passenger's historical travel trajectory, a push service model for the passenger's travel is constructed, a corresponding personalized push strategy is formulated, and it is pushed to the passenger's mobile APP or WeChat mini-program through offline network.

[0032] In this embodiment, as Figure 5 shown, in step L4, the construction of the push service model for the passenger's travel and the formulation of the corresponding personalized push strategy include: L41. Based on the data information of the predicted travel time of the passenger on the platform and the data information of the passenger's historical travel trajectory, normalization processing is performed to obtain the data information of the travel time of the passenger on the platform and the data information of the passenger's historical travel trajectory after normalization processing; L42. Based on the data information of the travel time of the passenger on the platform and the data information of the passenger's historical travel trajectory after normalization processing, an objective optimization function G of the passenger's travel trajectory is established, , where h1 is the data information of the travel time of the passenger on the platform after normalization processing, h2 is the data information of the passenger's historical travel trajectory after normalization processing, and ρ1, ρ2, and ρ3 are penalty factors; L43. Based on the objective optimization function G of the passenger's travel trajectory, the travel trajectory of the passenger is optimized to obtain the data information of the optimized travel trajectory of the passenger, thereby formulating a corresponding personalized push strategy.

[0033] In this embodiment, the corresponding personalized push strategies include a push strategy based on platform actions, a push strategy based on travel habits, a push strategy based on real-time scenarios, and a push strategy based on interest preferences. The penalty factors ρ1, ρ2, and ρ3 are , , , where h1 is the data information of the travel time of the passenger at the platform after normalization processing, and h2 is the data information of the historical travel trajectory of the passenger after normalization processing.

[0034] In this embodiment, the push strategy based on platform actions is shown in the following table:

[0035] In this embodiment, the push strategy based on travel habits is shown in the following table:

[0036] In this embodiment, the push strategy based on real-time scenarios is shown in the following table:

[0037] In this embodiment, the push strategy based on interest preferences is shown in the following table:

[0038] In this embodiment, the present invention provides a digital platform passenger behavior analysis and personalized service push system, including a computer device, which is programmed or configured to execute the steps of any one of the digital platform passenger behavior analysis and personalized service push methods.

[0039] In this embodiment, the present invention provides a computer-readable storage medium, on which a computer program is stored that is programmed or configured to execute any one of the digital platform passenger behavior analysis and personalized service push methods.

[0040] Any reference to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), among others.

[0041] In summary, the present invention can not only accurately collect and analyze information such as the behaviors, trajectories, and demands of passengers and pedestrians on the platform, but also provide services such as early notification of vehicle congestion and personalized ride suggestions, thereby improving the operation efficiency, service quality, and passenger travel experience of driverless buses.

[0042] The above specific embodiments do not constitute a limitation on the protection scope of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

1. A method for analyzing passengers' behaviors at a digital platform and pushing personalized services, characterized in that, The method includes: L1. When a passenger enters the digital platform, data information on the passenger's historical travel trajectory is obtained. Image data information of the passenger is obtained in real time based on the built-in cameras on the platform. Data information on the sequence of the passenger's position coordinates is obtained in real time based on the built-in infrared sensors on the platform. Data information on the time series of the pressure change on the platform is obtained in real time based on the built-in pressure sensors on the platform; L2. Based on the image data information of the passenger, a skeleton key-point detection model based on a dynamic learning factor is constructed to characterize the passenger's limb movement feature matrix, and data information on the passenger's limb movement feature matrix is obtained; L3. Based on the data information of the passenger's limb movement feature matrix, the data information of the sequence of the passenger's position coordinates, and the data information of the time series of the pressure change on the platform, a BP neural network regression prediction algorithm based on the Pearson correlation coefficient is used to predict the travel time of the passenger at the platform, and data information on the predicted travel time of the passenger at the platform is obtained; L4. Based on the data information of the predicted travel time of the passenger at the platform and the data information of the passenger's historical travel trajectory, a push service model for the passenger's travel is constructed, corresponding personalized push strategies are formulated, and they are pushed to the mobile APP or WeChat mini-program of the platform passengers through network-free push; 2. The digital platform passenger behavior analysis and personalized service push method according to claim 1, characterized in that In step L2, the construction of the skeleton key-point detection model based on the dynamic learning factor to characterize the passenger's limb movement feature matrix includes: L21. The image data information of the passenger is input into a convolutional neural network model to extract the passenger's image features, and data information on the passenger's image feature matrix is obtained; L22. Based on the data information of the passenger's image feature matrix, a function Q for extracting the passenger's skeleton key points is established, ， where x is the data information of the passenger's image feature matrix, α, β, and δ are weight coefficients, the passenger's skeleton key points are extracted, and data information on the coordinates of the passenger's skeleton key points is obtained; L23. Based on the data information of the coordinates of the passenger's skeleton key points, a function W for characterizing the passenger's limb movements is constructed, ， where y is the data information of the coordinates of the passenger's skeleton key points, γ1, γ2, and γ3 are dynamic learning factors, the passenger's limb movement feature matrix is characterized, and data information on the passenger's limb movement feature matrix is obtained.

3. The digital platform passenger behavior analysis and personalized service push method according to claim 2, characterized in that: The dynamic learning factors γ1, γ2, and γ3 are ， ， ， where y is the data information of the coordinates of the passenger's skeleton key points.

4. The method for analyzing the behavior of passengers on a digital platform and pushing personalized services according to claim 1, characterized in that, In step L3, the use of the BP neural network regression prediction algorithm based on the Pearson correlation coefficient to predict the travel time of the passenger at the platform includes: L31. Based on the data information of the sequence of the passenger's position coordinates and the data information of the time series of the pressure change on the platform, a Pearson correlation coefficient function R of two variables is established, , where X is the data information of the sequence of the passenger's position coordinates, Y is the data information of the time series of the pressure change on the platform, the relationship between the two variables is characterized, and data information on the relationship matrix between the passenger's position coordinates and the pressure change on the platform is obtained; L32. Input the data information of the relationship matrix between the position coordinates of the passenger and the pressure change on the platform and the data information of the passenger's body movement feature matrix into the BP neural network regression prediction model for training and learning to obtain a trained BP neural network regression prediction model; L33. Based on the trained BP neural network regression prediction model, input the data information of the relationship matrix between the position coordinates of the passenger and the pressure change on the platform and the data information of the passenger's body movement feature matrix to predict the travel time of the platform where the passenger is located, and obtain the data information of the predicted travel time of the platform where the passenger is located.

5. The digital platform passenger behavior analysis and personalized service push method according to claim 4, characterized in that: The BP neural network regression prediction model includes an input layer, a hidden layer, and an output layer. The neuron function P of the hidden layer is ， where r1 is the data information of the relationship matrix between the position coordinates of the passenger and the pressure change on the platform, r2 is the data information of the passenger's body movement feature matrix, and ω1, ω2, and ω3 are the weight factors of the BP neural network.

6. The digital platform passenger behavior analysis and personalized service push method according to claim 5, characterized in that: The constraint conditions for the weight factors ω1, ω2, and ω3 of the BP neural network are 。 7. The digital platform passenger behavior analysis and personalized service push method according to claim 1, characterized in that In step L4, the construction of the push service model for passenger travel and the formulation of corresponding personalized push strategies include: L41. Based on the data information of the predicted travel time of the platform where the passenger is located and the data information of the passenger's historical travel trajectory, perform normalization processing to obtain the data information of the travel time of the platform where the passenger is located and the data information of the passenger's historical travel trajectory after normalization processing; L42. Based on the data information of the travel time of the platform where the passenger is located and the data information of the passenger's historical travel trajectory after normalization processing, establish an objective optimization function G for the passenger travel trajectory, ， where h1 is the data information of the travel time of the platform where the passenger is located after normalization processing, h2 is the data information of the passenger's historical travel trajectory after normalization processing, and ρ1, ρ2, and ρ3 are penalty factors; L43. Based on the objective optimization function G of the passenger travel trajectory, optimize the passenger's travel trajectory to obtain the data information of the optimized passenger's travel trajectory, thereby formulating corresponding personalized push strategies.

8. The method for analyzing the behavior of passengers on a digital platform and pushing personalized services according to claim 7, characterized in that: The corresponding personalized push strategies include push strategies based on platform actions, push strategies based on travel habits, push strategies based on real-time scenarios, and push strategies based on interest preferences. The penalty factors ρ1, ρ2, and ρ3 are ， ， ， where h1 is the data information of the travel time of the platform where the passenger is located after normalization processing, and h2 is the data information of the passenger's historical travel trajectory after normalization processing.

9. A digital platform passenger behavior analysis and personalized service push system, including a computer device, characterized in that, The computer device is programmed or configured to execute the steps of the digital platform passenger behavior analysis and personalized service push method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program programmed or configured to execute the digital platform passenger behavior analysis and personalized service push method according to any one of claims 1 to 8.

Images