A method, terminal, and medium for predicting shared bicycle rental and return demand based on multi-source data.

Abstract

This invention discloses a method, terminal, and medium for predicting shared bicycle rental and return demand based on multi-source data. The method includes: acquiring shared bicycle riding order data, PO I data, and historical weather data; and statistically analyzing the rental and return volume at each shared bicycle parking station on an hourly basis; mining the temporal, spatial, and weather-related features of shared bicycle parking demand at each station, and determining the relationship between these three features and the station's rental and return demand; constructing passenger flow characteristics for interactive and similar stations; building a rental and return demand prediction model based on the passenger flow characteristics of interactive and similar stations, and training the model using historical passenger flow data. This invention delves into the relationship between various influencing factors and the rental and return demand at shared bicycle parking stations, enabling rental and return volume prediction, providing a reasonable bicycle allocation quantity for shared bicycle scheduling, and helping to reduce scheduling costs for shared bicycle companies.

Description

Technical Field

[0001] This invention relates to the field of shared bicycle management and scheduling technology, and in particular to a method, terminal and medium for predicting shared bicycle borrowing and returning demand based on multi-source data. Background Technology

[0002] In recent years, shared bicycles have become increasingly popular among urban residents due to their advantages such as being green and low-carbon, flexible and convenient, inexpensive, and capable of achieving "last mile" delivery, gradually becoming an important part of urban residents' travel. On the one hand, the shared bicycle market has shown a good development trend in recent years, with the number of shared bicycle users, annual riding volume, and average daily riding trips increasing year by year.

[0003] The spatiotemporal imbalance of demand refers to the significant fluctuations in demand for shared bicycles over time and their uneven spatial distribution. As the number of shared bicycle users and riding volume increases year by year, this spatiotemporal imbalance becomes increasingly pronounced, frequently leading to situations where "no bicycles are available and nowhere to return them." This phenomenon not only affects the user experience but also poses challenges to operating companies and city management. Due to this spatiotemporal imbalance, shared bicycle companies need to incur high operational costs to balance the mismatch between supply and demand.

[0004] Currently, shared bike operators primarily employ a unified scheduling approach to address scheduling issues. This involves concentrating scheduling efforts on stations with a limited or large number of bikes within a specific timeframe. This strategy suffers from significant lag and lacks a refined scheduling strategy tailored to the actual borrowing and returning needs of each station. The fundamental solution lies in accurately predicting short-term borrowing and returning demand at each station and developing a cost-optimized scheduling strategy—achieving proactive and precise scheduling. Specifically, by accurately predicting short-term borrowing volume at a station, if demand is high in the next time period, the precise number of bikes needed should be deployed promptly. Conversely, by accurately predicting short-term returning volume at a station, if return demand is high in the next time period, the precise number of redundant bikes should be removed promptly.

[0005] Therefore, it is necessary to accurately predict the short-term demand for car rentals at stations and transform the original centralized scheduling strategy into demand-oriented precision scheduling, so as to provide more refined and reliable data support for scheduling and deployment. Summary of the Invention

[0006] The technical problem to be solved by this invention is to address the above-mentioned deficiencies of the prior art by providing a method, terminal, and medium for predicting the demand for shared bicycle rentals and returns based on multi-source data. The technical solution adopted by this invention is as follows:

[0007] In a first aspect, the present invention provides a method for predicting the demand for shared bicycle rentals and returns based on multi-source data, wherein the method includes:

[0008] Acquire shared bike riding order data, POI data, and historical weather data, and calculate the number of bikes borrowed and returned at each parking station on an hourly basis;

[0009] We will explore the temporal, spatial, and weather-related characteristics of bike-sharing station borrowing and returning demand, and determine the relationship between these three characteristics and station borrowing and returning demand.

[0010] Based on the relationship between the three dimensions of characteristics and the demand for car rental and return at the stations, we construct the passenger flow characteristics of interactive stations and similar stations.

[0011] Based on the passenger flow characteristics of the interactive stations and similar stations, a bike-sharing demand prediction model is constructed, and historical passenger flow data is used to train the bike-sharing demand model. The bike-sharing demand model is used to predict the bike-sharing demand at shared bike stations.

[0012] In one implementation, the temporal, spatial, and weather-related characteristics of shared bicycle rental and return demand at stations are mined to determine the relationship between these three characteristics and the rental and return demand at stations, including:

[0013] Based on shared bicycle riding order data, the relationship between weekdays, weekends, holidays, and hourly periods and the demand for bicycle rental and return at stations is determined from the time dimension.

[0014] Based on POI data, the relationship between the geographical location, surrounding facilities, population density, and road network density of shared bicycle parking stations and the demand for bicycle rental and return at the stations is determined from a spatial perspective.

[0015] Based on historical weather data, the relationship between temperature, precipitation, and ground wind speed and the demand for car rentals and returns at stations is determined from a weather perspective.

[0016] In one implementation, based on shared bicycle riding order data, the relationship between weekdays, weekends, holidays, and hourly periods and the demand for bicycle rentals and returns at stations is determined from a time perspective, including:

[0017] By analyzing the time-varying curves of the number of bikes borrowed and returned at the station throughout the day on weekdays, weekends, and holidays through data visualization, we can identify the bike borrowing and returning demand patterns at different times.

[0018] Calculate the Spearman correlation coefficient between the passenger flow in this time period and the passenger flow in the previous N time periods;

[0019] Calculate the Spearman correlation coefficient between passenger flow in this time period and passenger flow in the same time period on different workdays of the same week, as well as passenger flow in the same time period on the same workday of different weeks.

[0020] In one implementation, based on POI data, the relationship between the geographical location, surrounding facilities, population density, and road network density of shared bicycle parking stations and the demand for bike rentals at these stations is determined from a spatial perspective, including:

[0021] Data visualization analysis was used to create a heatmap of car rental and return demand in different geographical locations during the same period;

[0022] Correlation analysis was used to determine the strength of the correlation between the population size, road network density, and distribution of various POI facilities around the station and the demand for car rentals and returns at the station.

[0023] In one implementation, based on historical weather data, the relationship between temperature, precipitation, and ground wind speed and the demand for bike rentals at stations is determined from a weather perspective, including:

[0024] The data visualization was used to analyze the scatter plots of different temperatures, ground wind speeds, and rainfall in relation to the demand for car rentals and returns at the stations.

[0025] This study explores the strong correlation between temperature, ground wind speed, rainfall, and the demand for bike rentals at stations based on correlation analysis.

[0026] In one implementation, based on the relationship between three dimensions of features and the demand for car rentals and returns at stations, interactive station passenger flow features and similar station passenger flow features are constructed, including:

[0027] Based on several shared bicycle riding order data, we screened out the OD pairs of closely interacting sites, and based on correlation analysis, we determined the most relevant time periods of influence of the interactive sites and constructed the passenger flow characteristics of the interactive sites.

[0028] A similarity measurement method for passenger flow sequences between two stations based on Euclidean distance is used to identify pairs of stations with similar characteristics in passenger flow trends and passenger flow volume, and to construct passenger flow characteristics of similar stations.

[0029] In one implementation, a car rental demand prediction model is constructed based on the passenger flow characteristics of the interactive site and similar sites, and the car rental demand model is trained using historical passenger flow data, including:

[0030] Spatiotemporal graph convolutional networks are used to model the passenger flow characteristics of interactive sites and similar sites, and to extract the temporal and spatial variations of car rental and return demand.

[0031] The temporal and spatial variation features output by the spatiotemporal graph convolutional network are input into the XGBoost model. High-dimensional features extracted from weather, spatial and temporal features are used to construct a car rental demand prediction model, which is a combination of the spatiotemporal graph convolutional network and the XGBoost model.

[0032] The car rental demand model was trained using historical passenger flow data, and the hyperparameters of the spatiotemporal graph convolutional network and the XGBoost model were tuned sequentially using grid search and Bayesian optimization.

[0033] Secondly, embodiments of the present invention also provide a system for predicting the demand for shared bicycle rentals and returns based on multi-source data, wherein the system includes:

[0034] The data statistics module is used to acquire shared bicycle riding order data, POI data, and historical weather data, and to count the number of shared bicycles borrowed and returned at each parking station on an hourly basis.

[0035] The relationship analysis module is used to mine the time dimension, spatial dimension, and weather dimension characteristics of shared bicycle station borrowing and returning demand, and to determine the relationship between the three dimensions of characteristics and the station borrowing and returning demand.

[0036] The feature construction module is used to construct interactive site passenger flow features and similar site passenger flow features based on the relationship between the three-dimensional features and the station's car rental and return demand.

[0037] The demand forecasting module is used to construct a bike-sharing demand forecasting model based on the passenger flow characteristics of the interactive station and similar stations, and to train the bike-sharing demand model using historical passenger flow data. The bike-sharing demand model is used to predict the bike-sharing demand at shared bike stations.

[0038] Thirdly, embodiments of the present invention also provide a terminal, wherein the terminal includes a memory, a processor, and a multi-source data-based shared bicycle rental and return demand prediction program stored in the memory and executable on the processor. When the processor executes the multi-source data-based shared bicycle rental and return demand prediction program, it implements the steps of the multi-source data-based shared bicycle rental and return demand prediction method of any of the above solutions.

[0039] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a bicycle rental and return demand prediction program based on multi-source data, and when the bicycle rental and return demand prediction program based on multi-source data is executed by a processor, it implements the steps of the bicycle rental and return demand prediction method based on multi-source data as described in any of the above schemes.

[0040] Beneficial Effects: Compared with existing technologies, this invention provides a method for predicting shared bicycle rental and return demand based on multi-source data. First, this invention acquires shared bicycle riding order data, POI data, and historical weather data, and then statistically analyzes the rental and return volume at each shared bicycle parking station on an hourly basis. Next, it mines the temporal, spatial, and weather-related features of shared bicycle station rental and return demand, determining the relationship between these three features and station rental and return demand. Then, based on the relationship between these three features and station rental and return demand, it constructs passenger flow characteristics for interactive stations and similar stations. Finally, it constructs a rental and return demand prediction model based on the interactive and similar station passenger flow characteristics, and trains the model using historical passenger flow data. This model is used to predict the rental and return demand at shared bicycle stations. This invention uses multi-source data, including shared bicycle riding order data, POI data, and historical weather data, to deeply explore the relationship between various influencing factors and the demand for shared bicycle parking stations, enabling the prediction of bicycle rental and return volume. This provides a reasonable number of bicycles to be allocated for shared bicycle scheduling, which helps reduce the scheduling costs for shared bicycle companies. Attached Figure Description

[0041] Figure 1 This is a flowchart of a preferred embodiment of the method for predicting the demand for shared bicycle rentals and returns based on multi-source data provided in this invention.

[0042] Figure 2 This is a flowchart of multi-source data acquisition and hourly data statistics for stations in one implementation of the present invention.

[0043] Figure 3 This is a heat map of car rental demand during the morning rush hour in various streets of Shenzhen, according to one implementation of the present invention.

[0044] Figure 4 This is a framework diagram of the STGCN_XGBoost combined prediction model in one implementation of the present invention.

[0045] Figure 5 This is a prediction effect diagram of the STGCN_XGBoost model in one implementation of the present invention.

[0046] Figure 6 This is a schematic diagram of the architecture of a shared bicycle rental and return demand prediction system based on multi-source data provided in an embodiment of the present invention.

[0047] Figure 7 This is a schematic diagram of the terminal provided in the embodiment of the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0049] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content, operations, or steps, nor does it require execution in the described order. For example, some operations or steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0050] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0051] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. For example, the first control information and the second control information are only used to distinguish different control information and do not limit their order.

[0052] Those skilled in the art will understand that the words "first" and "second" do not limit the quantity or the order of execution, and that the words "first" and "second" do not necessarily imply that they are different.

[0053] It should also be understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0054] The current shared bicycle operation suffers from problems such as "no bikes available and nowhere to return bikes" due to inaccurate demand forecasting. Traditional demand forecasting methods often rely on a single data source or simple time series models, ignoring complex factors affecting shared bicycle rental and return demand, such as weather changes, spatial distribution, and passenger flow characteristics. Therefore, this invention proposes a shared bicycle rental and return demand forecasting method based on multi-source data, utilizing multi-source data to improve the accuracy and applicability of forecasts. Specifically, this embodiment first acquires shared bicycle riding order data, POI data, and historical weather data, and then statistically analyzes the rental and return volume of each shared bicycle parking station on an hourly basis. Then, it mines the time dimension, spatial dimension, and weather dimension characteristics of shared bicycle station rental and return demand, determining the relationship between these three dimensions and station rental and return demand. Finally, based on the relationship between these three dimensions and station rental and return demand, it constructs passenger flow characteristics for interactive stations and passenger flow characteristics for similar stations. Finally, a bike-sharing demand prediction model is constructed based on the passenger flow characteristics of the interactive stations and similar stations. This model is then trained using historical passenger flow data and is used to predict the bike-sharing demand at shared bike stations. This invention utilizes multi-source data, including shared bike riding order data, POI data, and historical weather data, to deeply explore the relationship between various influencing factors and the bike-sharing demand at shared bike parking stations. This enables the prediction of bike-sharing volume, providing a reasonable number of bikes for shared bike dispatching and helping to reduce dispatching costs for shared bike companies.

[0055] The multi-source data-based shared bicycle rental and return demand prediction method of this embodiment can be applied to terminals, including intelligent product terminals such as mobile phones, smart TVs, and computers. Figure 1 As shown in the figure, the method for predicting shared bicycle rental and return demand based on multi-source data in this embodiment includes the following steps:

[0056] Step S100: Obtain shared bicycle riding order data, POI data, and historical weather data, and calculate the number of shared bicycles borrowed and returned at each parking station on an hourly basis.

[0057] In this embodiment, web crawling technology can be used to obtain shared bicycle riding order data, POI (Point of Interest) data (referring to point-based data in internet electronic maps), and historical weather data from various open data platforms. The shared bicycle riding order data includes attributes such as riding order number, start time, start location latitude and longitude coordinates, end time, and end location latitude and longitude coordinates. POI data includes attributes such as name, address, coordinates, and category. Historical weather data includes attributes such as region, time, temperature, wind speed, humidity, and rainfall.

[0058] Next, the origin and destination coordinates of the shared bike ride order data were separated. Then, the `Scikit-learn_NearestNeighbors` function from the Python library was used to assign the bike rental and return points to the nearest shared bike parking stations. Finally, the `pandas_resample` function was used to aggregate the time series data hourly to count the number of bike rentals and returns for each station. Specifically... Figure 2 As shown, this embodiment can read data by connecting to MongoDB data to obtain cycling order data, subway station entry and exit passenger flow data and weather data, then determine the nearest matching parking stations, and then perform statistics based on 1-hour time granularity to count the number of bikes borrowed and returned at each station.

[0059] Step S200: Extract the time dimension, spatial dimension, and weather dimension characteristics of shared bicycle station borrowing and returning demand, and determine the relationship between the three dimensions and the station borrowing and returning demand.

[0060] In this embodiment, the acquired shared bicycle riding order data, POI data, and historical weather data are first preprocessed. After preprocessing, the above three dimensions and the station's bike-sharing demand are analyzed from the perspectives of time, space, and weather.

[0061] Specifically, this embodiment first determines the relationship between shared bicycle riding order data and station rental and return demand from a time perspective, considering weekdays, weekends, holidays, and hourly segments. First, it uses data visualization analysis to analyze the time-varying curves of station rental and return volume throughout the day on weekdays, weekends, and holidays, identifying rental and return demand patterns at different times, such as weekday commuting peaks or weekend leisure usage demand. Then, it analyzes the similarity characteristics of passenger flow trends based on correlation analysis, specifically calculating the Spearman correlation coefficient between the passenger flow in the current time period and the passenger flow in the previous N time periods. Finally, it analyzes the periodic characteristics of passenger flow based on correlation analysis, specifically calculating the Spearman correlation coefficient between the passenger flow in the current time period and the passenger flow in the same time period on different weekdays within the same week, as well as the passenger flow in the same time period on different weekdays within the same week.

[0062] Furthermore, this embodiment, based on POI data, determines the relationship between the geographical location of shared bicycle parking stations, surrounding facilities (such as subway stations, shopping malls, schools, etc.), population density, road network density, and the demand for bicycle rentals and returns at these stations from a spatial perspective. In this embodiment, data visualization analysis is first used to analyze heatmaps of rental and return demand at different geographical locations during the same time period, such as... Figure 3 This represents the heat map of car rental demand on different streets during the morning rush hour. Secondly, based on correlation analysis, the strength of the correlation between the population size, road network density, and the distribution of various POI facilities around the stations and the demand for car rentals at the stations is determined.

[0063] Furthermore, this embodiment, based on historical weather data, determines the relationship between temperature, precipitation, and surface wind speed and the demand for bike rentals at designated stations from a weather perspective. This embodiment first uses data visualization to analyze scatter plots of different temperatures, surface wind speeds, and rainfall amounts with the demand for bike rentals at designated stations. Then, based on correlation analysis, it explores the strength of the correlation between temperature, surface wind speed, rainfall, and the demand for bike rentals at designated stations.

[0064] Step S300: Based on the relationship between the three-dimensional features and the station's car rental and return needs, construct the passenger flow features of interactive stations and similar stations.

[0065] This embodiment can filter out closely interacting site-to-origin (OD) pairs based on several shared bicycle ride order data, and determine the most relevant time periods of influence for the interacting sites based on correlation analysis, thus constructing the passenger flow characteristics of the interacting sites. Specifically, by statistically analyzing OD data from shared bicycle ride order data, the average proportion of the number of times from site O to site D within N days to the total number of bikes borrowed from site O is the site-to-OD interaction intensity, as shown in the following formula:

[0066]

[0067] Where S represents the interaction strength between OD sites; This represents the number of times the data travels from station O to station D on day i. Total number of car rentals at site O; N days;

[0068] The formula for the characteristics of passenger flow at the same time at interactive sites is as follows:

[0069]

[0070] in, For interactive site i at time t; Let be the interaction strength between site i and site j at time t; The number of vehicles borrowed at station j at time t.

[0071] Furthermore, a similarity measurement method for passenger flow sequences between two stations based on Euclidean distance is used to identify pairs of stations with similar characteristics in passenger flow change trends and passenger flow volume, and to construct passenger flow characteristics of similar stations.

[0072] Specifically, the Euclidean distance between the daily car rental time series of station i and station j is calculated, and a similarity threshold σ is set; if the similarity exceeds σ, the similarity is set to 0. The formula for calculating station similarity is as follows:

[0073]

[0074] Among them, P ij S represents the similarity between site i and site j; ijS is the Euclidean distance between the passenger flow time series of stations i and j; σ is the Euclidean distance threshold, S min For the minimum Euclidean distance, S max This represents the maximum Euclidean distance.

[0075] The characteristic formula for similar sites at the same time is as follows:

[0076]

[0077] in, P represents the passenger flow of similar stations to station i at time t; ij For site ij similarity; The number of vehicles borrowed at station j at time t.

[0078] Step S400: Construct a bike-sharing demand prediction model based on the passenger flow characteristics of the interactive station and similar stations, and train the bike-sharing demand model using historical passenger flow data. The bike-sharing demand model is used to predict the bike-sharing demand at shared bike stations.

[0079] This embodiment utilizes a spatiotemporal graph convolutional network to model the passenger flow characteristics of interactive stations and similar stations, extracting the temporal and spatial variations in car rental demand. Then, the temporal and spatial variation features output by the spatiotemporal graph convolutional network are input into an XGBoost model. High-dimensional features extracted from weather, spatial, and temporal characteristics are used to construct a car rental demand prediction model, which is a combination of the spatiotemporal graph convolutional network and the XGBoost model.

[0080] Specifically, this embodiment utilizes a Spatiotemporal Graph Convolutional Network (STGCN) to model the spatiotemporal dependencies between cycling stations. It learns the spatial relationships between stations through graph convolution (GCN) and combines this with a temporal convolutional network (TCN) to model time-series features, extracting characteristics of demand changes over time and space. For example... Figure 4 As shown. Next, the spatiotemporal features output by the STGCN model (Spatiotemporal Graph Convolutional Network) are input into the XGBoost model to further enhance its ability to capture nonlinear features. The XGBoost model can effectively utilize high-dimensional features extracted from weather, spatial, and temporal features, and make fine-grained predictions of car rental demand.

[0081] This embodiment uses 90 days of historical passenger flow data to train the obtained car rental demand model. The 90-day historical passenger flow data is used as the training dataset, divided into training, validation, and test sets at 70%, 15%, and 15% respectively. To ensure the temporal continuity of the data and the model's generalization ability, a sliding window method is chosen to generate training samples. The processed spatiotemporal weather features are input into the model for training. First, the spatiotemporal features are initially learned in the STGCN model, and then the output is input into the XGBoost model. This combined model improves prediction accuracy.

[0082] Next, hyperparameter tuning was performed on the spatiotemporal graph convolutional network (STGCN) and the XGBoost model using grid search and Bayesian optimization, respectively. Specifically, in the STGCN model, hyperparameters such as the number of convolutional layers and kernel size were adjusted to optimize the extraction of spatiotemporal features. Simultaneously, in the XGBoost model, key parameters such as the learning rate and tree depth were adjusted using grid search or Bayesian optimization methods to improve the overall model performance. This embodiment also allows for feature selection and combination based on temporal, spatial, and weather characteristics to ensure that the features input to the model effectively represent changes in demand. The sensitivity of the model to demand fluctuations was improved by gradually eliminating redundant features and introducing new features (such as temperature and holidays). Evaluation metrics such as MAPE (Mean Absolute Percentage Error) and RMSE (Root Mean Square Error) were used to quantify the model's prediction accuracy on the validation and test sets. By comparing the prediction performance under different parameter combinations, the model was further optimized to ensure the predictive performance of the car rental demand prediction model on the test data. Figure 5 As shown.

[0083] In summary, this embodiment first acquires shared bicycle riding order data, POI data, and historical weather data, and then statistically analyzes the hourly rental and return volume at each shared bicycle parking station. Next, it mines the temporal, spatial, and weather-related characteristics of shared bicycle parking station demand to determine the relationship between these three dimensions and the demand. Then, based on the relationship between these three dimensions and the demand, it constructs passenger flow characteristics for interactive and similar stations. Finally, it builds a rental and return demand prediction model based on these characteristics and trains it using historical passenger flow data. This model is used to predict the rental and return demand at shared bicycle parking stations. This embodiment, through multi-source data such as shared bicycle riding order data, POI data, and historical weather data, deeply explores the relationship between various influencing factors and the rental and return demand at shared bicycle parking stations, achieving rental and return volume prediction. This provides a reasonable number of bicycles for shared bicycle dispatching, helping to reduce dispatching costs for shared bicycle companies.

[0084] Based on the above embodiments, the present invention also provides a multi-source data-based shared bicycle rental and return demand prediction system, such as... Figure 6 As shown, the system includes: a data statistics module 10, a relationship analysis module 20, a feature construction module 30, and a demand prediction module 40. Specifically, the data statistics module is used to acquire shared bicycle riding order data, POI data, and historical weather data, and to statistically analyze the hourly rental and return volume of each shared bicycle parking station. The relationship analysis module 20 is used to mine the time-dimensional, spatial-dimensional, and weather-dimensional features of shared bicycle station rental and return demand, respectively, and determine the relationship between these three dimensions and station rental and return demand. The feature construction module 30 is used to construct interactive station passenger flow features and similar station passenger flow features based on the relationship between the three dimensions and station rental and return demand. The demand prediction module 40 is used to construct a rental and return demand prediction model based on the interactive station passenger flow features and similar station passenger flow features, and to train the rental and return demand model using historical passenger flow data. The rental and return demand model is used to predict the rental and return demand of shared bicycle stations.

[0085] The working principle of each module in the multi-source data-based shared bicycle rental and return demand prediction system of this embodiment is the same as the principle of each step in the above method embodiment, and will not be repeated here.

[0086] The modules in the aforementioned multi-source data-based shared bicycle rental and return demand prediction system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the terminal in hardware form or independent of it, or stored in the terminal's memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0087] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 7 As shown. The terminal may include one or more processors 100 ( Figure 7 (Only one is shown in the image), a memory 101, and a computer program 102 stored in the memory 101 and executable on one or more processors 100, such as a sleep analysis program based on multi-sensor data. When one or more processors 100 execute the computer program 102, they can implement the various steps in the sleep analysis method embodiment based on multi-sensor data. Alternatively, when one or more processors 100 execute the computer program 102, they can implement the functions of various modules / units in the sleep analysis system embodiment based on multi-sensor data, which is not limited here.

[0088] In one embodiment, the processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0089] In one embodiment, memory 101 may be an internal storage unit of an electronic device, such as a hard drive or RAM. Memory 101 may also be an external storage device of the electronic device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, memory 101 may include both internal and external storage units. Memory 101 is used to store computer programs and other programs and data required by the terminal. Memory 101 can also be used to temporarily store data that has been output or will be output.

[0090] Those skilled in the art will understand that Figure 7 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0091] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, operational databases, or other media used in the embodiments provided by this invention can 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 a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual operating data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting shared bicycle rental and return demand based on multi-source data, characterized in that, The method includes: Acquire shared bicycle riding order data, point data from POI internet electronic maps, and historical weather data, and calculate the number of shared bicycles borrowed and returned at each parking station on an hourly basis; The study analyzes the temporal, spatial, and weather-related characteristics of shared bicycle rental and return demand at various stations to determine the relationship between these three dimensions and the demand. Specifically, this includes: Based on shared bicycle riding order data, the relationship between weekdays, weekends, holidays, and hourly periods and the demand for bicycle rental and return at stations is determined from the time dimension. Based on POI data, the relationship between the geographical location, surrounding facilities, population density, and road network density of shared bicycle parking stations and the demand for bicycle rental and return at the stations is determined from a spatial perspective. Based on historical weather data, the relationship between temperature, precipitation, and ground wind speed and the demand for car rental and return at the stations is determined from a weather perspective. Based on the relationship between the three dimensions of characteristics and the demand for car rentals and returns at stations, we construct passenger flow characteristics for interactive stations and passenger flow characteristics for similar stations, specifically including: Based on several shared bicycle riding order data, we screened out the OD pairs of closely interacting sites, and based on correlation analysis, we determined the most relevant time periods of influence of the interactive sites and constructed the passenger flow characteristics of the interactive sites. A pairwise station passenger flow sequence similarity measurement method based on Euclidean distance is used to identify pairs of stations with similar characteristics in passenger flow change trends and passenger flow volume, and to construct passenger flow characteristics of similar stations. A bike-sharing demand prediction model is constructed based on the passenger flow characteristics of the interactive stations and similar stations. This model is trained using historical passenger flow data. The bike-sharing demand prediction model is used to predict the bike-sharing demand at shared bike stations, specifically including: The spatiotemporal graph convolutional network is used to model the passenger flow characteristics of interactive sites and similar sites, extract the features of time and space changes in car rental demand, learn the spatial relationship between sites through graph convolution, and combine it with the temporal convolutional network to model time series features, extract the features of demand changes with time and space. The temporal and spatial variation features output by the spatiotemporal graph convolutional network are input into the XGBoost model. High-dimensional features extracted from weather, spatial and temporal features are used to construct a car rental demand prediction model, which is a combination of the spatiotemporal graph convolutional network and the XGBoost model. The car rental demand prediction model was trained using historical passenger flow data, and the hyperparameters of the spatiotemporal graph convolutional network and the XGBoost model were tuned sequentially using grid search and Bayesian optimization.

2. The method for predicting shared bicycle rental and return demand based on multi-source data according to claim 1, characterized in that, Constructing visitor flow characteristics for interactive sites includes: By analyzing shared bike ride order data and OD (Original Demand) data, the average percentage of trips from station O to station D within N days, relative to the total number of bikes borrowed from station O, represents the station-OD interaction intensity. The formula is: in, The strength of interaction between OD sites; This represents the number of times the data travels from station O to station D on day i. This represents the total number of car rentals at station O. For the number of days; The formula for the characteristics of visitor flow at the same time on an interactive site is: in, For interactive site i at time t; Let be the interaction strength between site i and site j at time t; The number of vehicles borrowed at station j at time t.

3. The method for predicting shared bicycle rental and return demand based on multi-source data according to claim 1, characterized in that, Construct passenger flow characteristics for similar stations, including: Calculate the Euclidean distance between the daily car rental time series of stations i and j, and set an Euclidean distance threshold. ,Exceed A similarity score of 0 indicates a similarity of zero. The formula for calculating site similarity is: in, Let i be the similarity between site i and site j; Let Euclidean distance be the time series distance of passenger flow between stations i and j. The Euclidean distance threshold is... The minimum Euclidean distance. The maximum Euclidean distance; The formula for passenger flow characteristics at similar stations at the same time is: in, The passenger flow of similar stations to station i at time t; For site ij similarity; The number of vehicles borrowed at station j at time t.

4. The method for predicting shared bicycle rental and return demand based on multi-source data according to claim 1, characterized in that, Based on shared bicycle riding order data, the relationship between weekdays, weekends, holidays, and hourly periods and the demand for bicycle rentals and returns at stations is determined from a time perspective, including: By analyzing the time-varying curves of the number of bikes borrowed and returned at the station throughout the day on weekdays, weekends, and holidays through data visualization, we can identify the bike borrowing and returning demand patterns at different times. Calculate the Spearman correlation coefficient between the passenger flow in this time period and the passenger flow in the previous several time periods; Calculate the Spearman correlation coefficient between passenger flow in this time period and passenger flow in the same time period on different workdays of the same week, as well as passenger flow in the same time period on the same workday of different weeks.

5. The method for predicting shared bicycle rental and return demand based on multi-source data according to claim 1, characterized in that, Based on POI data, the spatial relationship between the geographical location, surrounding facilities, population density, and road network density of shared bicycle parking stations and the demand for bike rentals and returns is determined, including: Data visualization analysis was used to create a heatmap of car rental and return demand in different geographical locations during the same period; Correlation analysis was used to determine the strength of the correlation between the population size, road network density, and distribution of various POI facilities around the station and the demand for car rentals and returns at the station.

6. The method for predicting shared bicycle rental and return demand based on multi-source data according to claim 1, characterized in that, Based on historical weather data, the relationship between temperature, precipitation, and ground wind speed and the demand for bike rentals at stations is determined from a weather perspective, including: The data visualization was used to analyze scatter plots of different temperatures, ground wind speeds, and rainfall in relation to the demand for car rentals and returns at the stations. This study explores the strong correlation between temperature, ground wind speed, rainfall, and the demand for bike rentals at stations based on correlation analysis.

7. A shared bicycle rental and return demand prediction system based on multi-source data, characterized in that, The system is used to implement the steps of the method for predicting the demand for shared bicycle rentals based on multi-source data as described in any one of claims 1-6, and the system includes: The data statistics module is used to acquire shared bicycle riding order data, POI data, and historical weather data, and to count the number of shared bicycles borrowed and returned at each parking station on an hourly basis. The relationship analysis module is used to mine the time dimension, spatial dimension, and weather dimension characteristics of shared bicycle station borrowing and returning demand, and to determine the relationship between the three dimensions of characteristics and the station borrowing and returning demand. The feature construction module is used to construct interactive site passenger flow features and similar site passenger flow features based on the relationship between the three-dimensional features and the station's car rental and return demand. The demand prediction module is used to construct a bike-sharing demand prediction model based on the passenger flow characteristics of the interactive station and similar stations, and to train the bike-sharing demand prediction model using historical passenger flow data. The bike-sharing demand prediction model is used to predict the bike-sharing demand at shared bike stations.

8. A terminal, characterized in that, The terminal includes a memory, a processor, and a multi-source data-based shared bicycle rental and return demand prediction program stored in the memory and executable on the processor. When the processor executes the multi-source data-based shared bicycle rental and return demand prediction program, it implements the steps of the multi-source data-based shared bicycle rental and return demand prediction method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a demand prediction program for shared bicycles based on multi-source data. When the demand prediction program for shared bicycles based on multi-source data is executed by a processor, it implements the steps of the demand prediction method for shared bicycles based on multi-source data as described in any one of claims 1-6.

Images