A method for predicting epidemic pathogens based on big data

Abstract

The present invention discloses a method for predicting epidemic pathogens based on big data, which specifically relates to the field of epidemic pathogen prediction, including: collecting passenger flow, health declaration, boarding and alighting records and environmental data of transportation hubs through standardized protocols, performing cleaning and spatiotemporal correlation processing, and generating a multi-factor dynamic parameter set. Based on the SEIR model framework, combined with the spatiotemporal data of transportation hubs, a multidimensional spatiotemporal description of the pathogen transmission chain is generated; a spatiotemporal dynamic network is constructed, the weight of the transmission path is adjusted, and the priority of the transmission risk is identified; the dynamic relationship between nodes is tracked by using health declaration and boarding and alighting records, and the trajectory of the main transmission path is constructed; the propagation chain expansion coefficient between paths is fitted by nonlinear regression, and a regional transmission risk prediction model is established. The model is adjusted and verified in real time, high-risk paths are predicted and early warnings are given, providing data support and decision-making basis for epidemic pathogen spread prevention and control.

Description

Technical Field

[0001] The present invention relates to the technical field of epidemic pathogen prediction, and more specifically, the present invention relates to an epidemic pathogen prediction method based on big data. Background Art

[0002] Since the spread of pathogens has certain spatiotemporal characteristics, the periodic changes in passenger flow between transportation hubs make the spread of pathogens between different transportation routes present a certain regular correlation, and the flow density and passenger flow direction between transportation hubs have significant characteristics, which provides favorable conditions for the spread of pathogens in the transportation hub network. Existing epidemiological prediction models are mainly based on the plane or regional spread of pathogens, and lack specific analysis of the spread of transportation hub path dimensions.

[0003] Therefore, how to propose a pathogen transmission risk prediction method based on the transmission path of pathogens in transportation hubs as the analysis dimension through the passenger flow characteristics of transportation hubs, passenger health declarations, and monitoring of pathogen transmission characteristics in big data information, so as to more accurately identify potential high-risk paths and time points, realize early risk warning and dynamic adjustment for transportation hub sites, and provide scientific decision-making support for disease prevention and control, is an urgent problem to be solved.

[0004] In order to solve the above problems, a technical solution is now provided. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an epidemic pathogen prediction method based on big data to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] S1: Collect passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of transportation hubs in the monitoring area through standardized data collection protocol interfaces, perform data cleaning and normalization, associate data with time and geographic space dimensions, establish multi-factor dynamic parameter sets, and store them in layers according to time and space attributes;

[0008] S2: Based on the framework of the SEIR model, a basic model of pathogen transmission is established, and multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multi-dimensional spatiotemporal description of the pathogen transmission chain;

[0009] S3: Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed, and the weight of each edge of the network is dynamically adjusted to characterize the transmission intensity of the path and identify the risk priority of the pathogen transmission path;

[0010] S4: Based on the passenger health declaration data of each node and the passenger boarding and alighting records of each node, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path;

[0011] S5: Based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficient between all adjacent paths is iteratively fitted by the nonlinear regression method to establish a regional propagation risk prediction model based on the weighted algorithm;

[0012] S6: cyclically compare and verify the prediction results of the regional transmission risk prediction model with the current actual calculation results, and adjust the model parameters until the model reaches the set accuracy value;

[0013] S7: Predict the transmission intensity of pathogen transmission paths within the set monitoring area in real time, display high-risk transmission paths of pathogens and corresponding transportation hubs, and provide early warning of pathogen spread for transportation hubs in all high-risk transmission paths.

[0014] In a preferred embodiment, in S1, the passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of the transportation hub in the monitoring area are collected through a standardized data collection protocol interface, data cleaning and normalization are performed, the data are associated with the time dimension and the geographic space dimension, a multi-factor dynamic parameter set is established, and hierarchical storage is performed according to the time and space attributes, specifically including:

[0015] Collect data from transportation hubs in the monitoring area through a standardized data collection protocol interface, including passenger flow, passenger health declaration information, passenger boarding and alighting records, and environmental monitoring data, and upload them to the central database for centralized archiving;

[0016] Perform cleaning operations on the collected transportation hub data according to the preset cleaning rules, and standardize the data fields according to the unified format structure and coding;

[0017] Based on the operating hours and geographic locations of all transportation hubs in the monitoring area, the data is associated with the set time dimension and geographic space dimension, and the data is given an accurate timestamp and a spatial code corresponding to the hub location;

[0018] The normalized data are marked as a multi-factor dynamic parameter set, stored in a regional-time window layer according to their spatiotemporal attributes, and an index file is generated as a basis for data retrieval.

[0019] In a preferred embodiment, in S2, a basic model of pathogen transmission is established based on the framework of the SEIR model, and multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multidimensional spatiotemporal description of the pathogen transmission chain, specifically including:

[0020] Based on the framework of the SEIR model, combined with the passenger flow and geographic location of the transportation hub data in the central database, a basic model of pathogen transmission was established to preliminarily simulate the flow of pathogens between different transportation hubs;

[0021] According to the location information of each transportation hub, the multi-factor dynamic parameter set of each transportation hub is retrieved, and the multi-factor dynamic parameter set is introduced into the basic model of pathogen transmission. Combined with the geographical location of each transportation hub and the time series changes of passenger flow, a multi-dimensional spatiotemporal description of the transmission chain in the basic model of pathogen transmission is generated. The multi-dimensional spatiotemporal description includes the geographical location, connection paths and passenger flow change trends between the connection paths of each transportation hub station.

[0022] In a preferred embodiment, in S3, a spatiotemporal dynamic network is constructed based on the multidimensional spatiotemporal description of the pathogen transmission chain, and the weight of each edge of the network is dynamically adjusted to characterize the path transmission intensity. The identification of the risk priority of the pathogen transmission path specifically includes:

[0023] Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed. Each network node in the network diagram represents a transportation hub, and the edge represents the specific diffusion path of the pathogen transmission chain.

[0024] Set the pathogen type to be monitored, and based on the relevant symptoms caused by the pathogen type, exclude the health declaration data of passengers of non-monitored pathogen types, and analyze the passenger flow trend based on the passenger flow time series change data in the pathogen transmission chain;

[0025] Based on passenger flow trends and screened passenger health declaration data, combined with environmental monitoring data, a comprehensive assessment is conducted to dynamically adjust the weight of each edge to characterize the path transmission intensity of pathogens;

[0026] A graph search algorithm is used to analyze the propagation chain between nodes, and the risk priority of the propagation path is identified according to the order of the path propagation intensity.

[0027] In a preferred embodiment, in S4, based on the passenger health declaration data of each node and the node passenger boarding and disembarking records, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path.

[0028] In a preferred embodiment, in S5, based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficient between all adjacent paths is iteratively fitted by a nonlinear regression method, and a regional propagation risk prediction model based on a weighted algorithm is established, which specifically includes:

[0029] Obtain the dynamic main path trajectory of pathogens in the spatiotemporal dynamic network and the change in the path propagation intensity adjacent to the main path trajectory in real time;

[0030] The sliding monitoring interval window size is set, and the correlation of the propagation intensity changes in the continuous sliding monitoring interval windows of all adjacent paths in the spatiotemporal dynamic network is counted. The propagation chain expansion coefficients between all adjacent paths are iteratively fitted with the main propagation path trajectory as the center point through the nonlinear regression method.

[0031] According to the transmission characteristics of the pathogen type to be monitored and combined with the expansion coefficient between the transmission chains, a regional transmission risk prediction model based on a weighted algorithm is established.

[0032] In a preferred embodiment, in S6, the prediction results of the regional transmission risk prediction model are cyclically compared and verified with the current actual calculation results, and the model parameters are adjusted until the model reaches the set accuracy value, which specifically includes:

[0033] The regional transmission risk prediction model predicts the path transmission intensity of all paths within the next sliding monitoring interval window;

[0034] Compare and verify the prediction results of the regional transmission risk prediction model in the previous sliding monitoring interval window with the current actual calculation results, adjust the model parameters according to the prediction deviation, and repeat the prediction and comparison adjustment process until the prediction accuracy of the model reaches the set accuracy value.

[0035] In a preferred embodiment, in S7, the path propagation intensity of the pathogen propagation path in the set monitoring area is predicted in real time, the high-risk propagation path of the pathogen and the corresponding transportation hub are displayed, and the early warning of pathogen spread is provided for all transportation hubs in the high-risk propagation path, specifically including:

[0036] Deploy the regional transmission risk prediction model to the server to predict the transmission intensity of the pathogen transmission path in the set monitoring area in real time;

[0037] The warning levels are divided according to the set predicted path transmission intensity warning level gradient, and GIS is used to display the high-risk transmission paths of pathogens and corresponding transportation hubs in real time, and the transmission hotspot map is dynamically updated;

[0038] Provide early warning of pathogen spread for transportation hubs corresponding to nodes in all high-risk transmission paths shown in the transmission hotspot map.

[0039] Technical effects and advantages of the epidemic pathogen prediction method based on big data of the present invention:

[0040] Through dynamic data collection and analysis, the transmission patterns of pathogens in different time periods can be accurately identified. Through multi-factor modeling, passenger flow, health declaration, passenger boarding and alighting data and environmental factors are combined to establish a transmission risk prediction model based on the path dimension, which can reflect the transmission of pathogens between transportation hubs in real time. The model dynamically adjusts the transmission intensity and risk assessment of each path to ensure that the priority of the transmission path is consistent with the actual situation, effectively improving the accuracy and timeliness of risk identification. The model also combines the boarding and alighting records and health declaration data of each transportation hub node to track the dynamic relationship between nodes and monitor potential transmission chains in real time. By performing nonlinear regression analysis on the intensity changes of the transmission path, the prediction accuracy of the path transmission intensity is further optimized. The regional transmission risk prediction model established by the weighted algorithm can provide more accurate risk assessment and realize dynamic adjustment under different time and space conditions. Through continuous cycle verification and parameter optimization, the accuracy of the model is continuously improved, which can provide real-time and reliable decision-making support for the epidemic prevention and control measures of transportation hubs, ensuring that the risk of pathogen transmission is timely warned and effectively controlled. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a schematic diagram of an epidemic pathogen prediction method based on big data of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] Example 1

[0044] Figure 1 The present invention provides a method for predicting epidemic pathogens based on big data, which includes the following steps:

[0045] S1: Collect passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of transportation hubs in the monitoring area through standardized data collection protocol interfaces, perform data cleaning and normalization, associate data with time and geographic space dimensions, establish multi-factor dynamic parameter sets, and store them in layers according to time and space attributes;

[0046] S2: Based on the framework of the SEIR model, a basic model of pathogen transmission is established, and multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multi-dimensional spatiotemporal description of the pathogen transmission chain;

[0047] S3: Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed, and the weight of each edge of the network is dynamically adjusted to characterize the transmission intensity of the path and identify the risk priority of the pathogen transmission path;

[0048] S4: Based on the passenger health declaration data of each node and the passenger boarding and alighting records of each node, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path;

[0049] S5: Based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficient between all adjacent paths is iteratively fitted by the nonlinear regression method to establish a regional propagation risk prediction model based on the weighted algorithm;

[0050] S6: cyclically compare and verify the prediction results of the regional transmission risk prediction model with the current actual calculation results, and adjust the model parameters until the model reaches the set accuracy value;

[0051] S7: Predict the transmission intensity of pathogen transmission paths within the set monitoring area in real time, display high-risk transmission paths of pathogens and corresponding transportation hubs, and provide early warning of pathogen spread for transportation hubs in all high-risk transmission paths.

[0052] In S1, the passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of transportation hubs in the monitoring area are collected through the standardized data collection protocol interface, and data cleaning and normalization are performed. The data are associated with the time dimension and geographic space dimension, and a multi-factor dynamic parameter set is established, and hierarchical storage is performed according to time and space attributes.

[0053] A standardized data collection protocol interface is used to collect data from all transportation hubs within the set monitoring area. This interface supports multiple data formats, including JSON and XML formats, and can process heterogeneous data from different sensors and data sources. Data sources include: passenger flow monitoring systems in transportation hubs, health declaration systems (such as passenger temperature monitoring and health status records), passenger boarding and alighting record systems (including electronic ticketing and manual records), and environmental monitoring equipment (such as air quality detection, temperature and humidity sensors).

[0054] A rule engine is used to automate data cleaning operations. The steps of data cleaning include deduplication, filling missing values, correcting format errors, and handling outliers. Taking health declaration data as an example, if the temperature field in the passenger's health declaration is negative or exceeds the normal range (such as a temperature exceeding 42°C), the data will be marked as abnormal. During the formatting process, all data fields will be converted to a unified standard format. For example, all date and time formats will be unified to "YYYY-MM-DD HH:MM:SS", and all text fields will be uniformly encoded in UTF-8 to ensure consistency and readability of data storage and subsequent analysis. For structured data (such as passenger boarding and disembarking records), unified encoding rules will be generated based on the data fields to ensure that all records can be uniformly identified during cross-regional analysis.

[0055] Both passenger flow data and health declaration data need to be labeled according to specific time, given precise timestamps, and the Global Positioning System (GPS) used to assign spatial codes to each piece of data based on the geographic location of each transportation hub.

[0056] After all data cleaning and association work is completed, the data will be marked as a multi-factor dynamic parameter set. This set contains multiple factors related to the spread of pathogens, such as passenger flow, health data, environmental factors, etc. A hierarchical storage method based on time windows is used to store data in different time periods and spatial levels. In practice, the data is divided into time windows by hour (such as morning peak, evening peak, and non-peak hours), and the data is allocated to different regional levels according to the geographical location of the transportation hub. A distributed database (such as Hadoop or HBase) is used for storage, and index files are generated for each layer of data for fast retrieval.

[0057] In S2, a basic model of pathogen transmission is established based on the framework of the SEIR model. Multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multidimensional spatiotemporal description of the pathogen transmission chain.

[0058] By using the framework of the SEIR model, the pathogen transmission process is divided into four stages: susceptible (S), exposed (E), infected (I), and recovered (R). According to the transmission characteristics of the epidemic pathogen types currently notified by public health agencies, the following basic transmission model (including the three stages of susceptible, exposed, and infected) is established:

[0059]

[0060] In the formula, is the number of susceptible people, is the number of latent populations, is the number of infected people, N is the total population, is the infection rate (depending on the contact rate and the probability of transmission), is the inverse of the incubation period (rate from exposure to infection), is the recovery rate, t is the set monitoring time variable, and the above data are provided by the actual data released by public health agencies for each type of epidemic pathogen.

[0061] According to the location information of each transportation hub, the multi-factor dynamic parameter set of each transportation hub is retrieved and introduced. This parameter set includes multiple factors related to the spread of pathogens, such as passenger flow, environmental monitoring data, passenger health declaration, etc. The specific operation includes automatically calling the dynamic parameter set related to the geographical location and passenger flow stored in the database through the location information of the transportation hub, and introducing these parameter sets into the basic model of pathogen transmission.

[0062] For example, by introducing dynamic adjustment of traffic flow, the propagation parameters Expanded to a dynamic function that depends on spatiotemporal traffic flow , specifically:

[0063]

[0064] in, Traffic flow at location (x, y) at time t, The number of people at position (x, y). The remaining multi-factor dynamic parameters can be introduced in a similar way to help achieve detailed control of the pathogen transmission process, making the transmission model closer to the actual scenario.

[0065] Specifically, passenger flow data is used to estimate the frequency of contact between susceptible and infected people in each transportation hub, while geographic location information determines the transmission routes and potential transmission chains of pathogens from one hub to another. The model gradually simulates the flow of pathogens between different hubs by continuously updating the states of susceptible, exposed, infected, and recovered, and combines the geographic location information of each transportation hub with the time series changes of passenger flow to generate transmission chains between hubs. These transmission chains reflect the spatiotemporal paths of pathogens flowing between different transportation hubs.

[0066] In S3, a spatiotemporal dynamic network is constructed based on the multidimensional spatiotemporal description of the pathogen transmission chain, and the weight of each edge of the network is dynamically adjusted to characterize the path transmission intensity and identify the risk priority of the pathogen transmission path.

[0067] Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed by treating each transportation hub as a node and setting edges according to the connection and transmission path between hubs in the pathogen transmission chain. Each edge represents a specific transmission path and is dynamically assigned values ​​based on passenger flow, health status and environmental factors.

[0068] After constructing the spatiotemporal dynamic network, the monitored pathogen type, such as influenza virus, is set, and the passenger health declaration data is screened based on the relevant symptoms caused by the type (such as fever, cough, etc.). By comparing the symptom information in the passenger health declaration, the health declaration data that is not related to the set monitored pathogen type is eliminated, so that only the health information related to the monitored pathogen is retained for subsequent analysis.

[0069] In addition, by analyzing the time series change data of passenger flow at transportation hubs, we can capture the changing trend of passenger flow (such as changes in the morning peak and evening peak hours) and the direction of passenger flow, and derive the passenger flow trend. This trend analysis is based on dynamic data sources (such as real-time traffic monitoring systems). By statistically analyzing the directionality of passenger flow in different time periods, we can identify high-density and low-density areas, thereby providing key support for the prediction of pathogen transmission paths.

[0070] The screened passenger health declaration data is combined with environmental monitoring data (such as temperature, humidity, air quality, etc.) to conduct a comprehensive evaluation of the transmission intensity through the weighted sum calculation method. The weight of each edge is dynamically adjusted according to the passenger flow trend and transmission intensity of different hubs. The weight value reflects the intensity and potential risk of the transmission path. For example, during the morning peak hours, the passenger flow is large and the air quality is poor, the risk of pathogen transmission is high, and the corresponding transmission path weight value will be large. During non-peak hours, the passenger flow is small and the air quality is good, then the transmission path weight value will be small.

[0071] Graph search algorithms (such as Dijkstra or A*) are used to analyze the transmission chain between transportation hub nodes. The specific operation is as follows: Based on the weight of each path (i.e., transmission strength), the algorithm calculates the shortest path from the source node to the target node to identify the main path of pathogen transmission. For each transmission path, the risk priority is automatically sorted by the size of the transmission strength.

[0072] In S4, based on the passenger health declaration data of each node and the passenger boarding and alighting records at each node, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path.

[0073] In terms of the node passenger boarding and alighting records, the system will record the boarding information of each passenger in real time, and mark the specific boarding and alighting time and location. By analyzing the health declaration data and boarding and alighting records of each transportation hub node, a dynamic relationship between nodes is established. The specific operation is to track the increase and decrease of passengers whose health declaration symptoms meet the set monitoring pathogen symptoms at each station based on the passenger flow data of each node (such as boarding and alighting records), further improve the dynamic relationship between nodes of the spatiotemporal dynamic network, ensure more accurate calculation of the transmission intensity of each transmission path, and mark the transmission path that is finally identified and automatically sorted by risk priority as the main transmission path trajectory, and regard all hubs on the path as important transmission hubs.

[0074] In S5, based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficients between all adjacent paths are iteratively fitted by the nonlinear regression method to establish a regional propagation risk prediction model based on the weighted algorithm.

[0075] The dynamic propagation main path trajectory of pathogens in the spatiotemporal dynamic network and the change in path propagation intensity adjacent to the main path trajectory are obtained in real time.

[0076] The sliding monitoring interval window size is set (the sliding window size is automatically adjusted according to the actual transportation cycle of the transportation hub), and the change in the adjacent path propagation intensity is obtained. The system uses a nonlinear regression method for fitting. The regression method analyzes the propagation chain expansion coefficient between the main propagation path and the adjacent path through an iterative process. The change in the propagation intensity of each adjacent path is used as the independent variable, and the expansion coefficient of the propagation chain is used as the dependent variable, and the nonlinear regression model is used for fitting. In this way, the system can capture the propagation relationship between adjacent paths and its influencing factors, thereby fitting the expansion coefficient of the propagation chain.

[0077] Based on the expansion coefficient of the fitted adjacent path transmission chain and the transmission characteristics of the pathogen type to be monitored, a regional transmission risk prediction model based on a weighted algorithm is established. The weighted algorithm calculates the overall transmission risk (expressed according to the predicted transmission intensity of adjacent paths) by assigning different weight values ​​to different paths. During the calculation process, the weight of each path depends not only on its transmission intensity, but also on the transmission characteristics of the pathogen, which include the infectivity weight and the latent period weight setting. The following is a specific example:

[0078] For example, influenza virus is transmitted through air (aerosol transmission) and direct contact, so its infectiousness weight is relatively high, set to 1.5, which means that it is easier to spread than viruses transmitted through general contact. At the same time, the incubation period of influenza virus is relatively long. Therefore, although there are no symptoms during the incubation period, the virus can still be transmitted. Therefore, the weight of the incubation period is set to 1.2.

[0079] In S6, the prediction results of the regional transmission risk prediction model are compared and verified with the current actual calculation results in a loop, and the model parameters are adjusted until the model reaches the set accuracy value.

[0080] The regional transmission risk prediction model predicts the path transmission intensity of all paths within the next sliding monitoring interval window.

[0081] Compare and verify the prediction results of the regional transmission risk prediction model in the previous sliding monitoring interval window with the current actual calculation results, adjust the model parameters according to the prediction deviation, and repeat the prediction and comparison adjustment process until the prediction accuracy of the model reaches the set accuracy value (specific settings are made according to the transmission ability of the set monitored pathogens, and higher accuracy standards are set for pathogens with strong transmission capabilities).

[0082] In S7, the path transmission intensity of the pathogen transmission path within the set monitoring area is predicted in real time, the high-risk transmission paths of the pathogens and the corresponding transportation hubs are displayed, and early warning of pathogen spread is provided for all transportation hubs in the high-risk transmission paths.

[0083] First, the regional transmission risk prediction model is deployed on the central server. The server uses efficient computing resources and real-time data processing capabilities to regularly obtain the latest data from data sources (such as passenger flow at transportation hubs, health declarations, and environmental monitoring systems), and processes and analyzes them. The prediction model runs in real time, dynamically calculating the transmission path and transmission intensity of pathogens through input data (such as passenger flow, temperature, humidity, etc.). To ensure the efficient operation of the model, the model adopts a distributed computing architecture that supports large-scale concurrent data processing and real-time feedback.

[0084] Based on the set transmission intensity warning level classification of the transmission path, the system makes real-time predictions on the transmission intensity of each transmission path. The transmission intensity is dynamically updated based on information such as pathogen type, passenger flow data, and health declarations at transportation hubs to form real-time transmission intensity data for each path. According to the preset warning level gradient (such as low risk, medium risk, and high risk), the system compares the transmission intensity of each path with the warning threshold to divide the corresponding warning level.

[0085] Based on the display of the transmission heat map, the system will issue early warnings for all transportation hubs in the high-risk transmission paths. For each marked important transmission hub, different emergency measures will be triggered according to its warning level. For example, for transportation hubs with high risk levels, the system will issue emergency warnings.

[0086] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0087] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium may be a solid-state hard disk.

[0088] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0089] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0090] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0091] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, and may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0092] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0093] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0094] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0095] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting epidemic pathogens based on big data, characterized in that: The steps include: S1: Collect passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of transportation hubs in the monitoring area through standardized data collection protocol interfaces, perform data cleaning and normalization, associate data with time and geographic space dimensions, establish multi-factor dynamic parameter sets, and store them in layers according to time and space attributes; S2: Based on the framework of the SEIR model, a basic model of pathogen transmission is established, and multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multi-dimensional spatiotemporal description of the pathogen transmission chain; S3: Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed, and the weight of each edge of the network is dynamically adjusted to characterize the transmission intensity of the path and identify the risk priority of the pathogen transmission path; S4: Based on the passenger health declaration data of each node and the passenger boarding and alighting records of each node, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path; S5: Based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficient between all adjacent paths is iteratively fitted by the nonlinear regression method to establish a regional propagation risk prediction model based on the weighted algorithm; S6: cyclically compare and verify the prediction results of the regional transmission risk prediction model with the current actual calculation results, and adjust the model parameters until the model reaches the set accuracy value; S7: Predict the transmission intensity of pathogen transmission paths within the set monitoring area in real time, display high-risk transmission paths of pathogens and corresponding transportation hubs, and provide early warning of pathogen spread for transportation hubs in all high-risk transmission paths.

2. The method for predicting epidemic pathogens based on big data according to claim 1, characterized in that: In S1, the passenger flow, health declaration, passenger boarding and alighting records and environmental monitoring data of the transportation hubs in the monitoring area are collected through the standardized data collection protocol interface, and data cleaning and normalization are performed. The data are associated with the time dimension and the geographic space dimension, and a multi-factor dynamic parameter set is established. The data is stored in layers according to the time and space attributes. Specifically, it includes: Collect data from transportation hubs in the monitoring area through a standardized data collection protocol interface, including passenger flow, passenger health declaration information, passenger boarding and alighting records, and environmental monitoring data, and upload them to the central database for centralized archiving; Perform cleaning operations on the collected transportation hub data according to the preset cleaning rules, and standardize the data fields according to the unified format structure and coding; Based on the operating hours and geographic locations of all transportation hubs in the monitoring area, the data is associated with the set time dimension and geographic space dimension, and the data is given an accurate timestamp and a spatial code corresponding to the hub location; The normalized data are marked as a multi-factor dynamic parameter set, stored in a regional-time window layer according to their spatiotemporal attributes, and an index file is generated as a basis for data retrieval.

3. The method for predicting epidemic pathogens based on big data according to claim 2, characterized in that: In S2, a basic model of pathogen transmission is established based on the framework of the SEIR model. Multi-factor dynamic parameters are automatically introduced according to the spatiotemporal index of each transportation hub to generate a multidimensional spatiotemporal description of the pathogen transmission chain, including: Based on the framework of the SEIR model, combined with the passenger flow and geographic location of the transportation hub data in the central database, a basic model of pathogen transmission was established to preliminarily simulate the flow of pathogens between different transportation hubs; According to the location information of each transportation hub, the multi-factor dynamic parameter set of each transportation hub is retrieved, and the multi-factor dynamic parameter set is introduced into the basic model of pathogen transmission. Combined with the geographical location of each transportation hub and the time series changes of passenger flow, a multi-dimensional spatiotemporal description of the transmission chain in the basic model of pathogen transmission is generated. The multi-dimensional spatiotemporal description includes the geographical location, connection paths and passenger flow change trends between the connection paths of each transportation hub station.

4. The method for predicting epidemic pathogens based on big data according to claim 3, characterized in that: In S3, a spatiotemporal dynamic network is constructed based on the multidimensional spatiotemporal description of the pathogen transmission chain, and the weight of each edge of the network is dynamically adjusted to characterize the path transmission intensity. The risk priority of the pathogen transmission path is identified specifically including: Based on the multidimensional spatiotemporal description of the pathogen transmission chain, a spatiotemporal dynamic network is constructed. Each network node in the network diagram represents a transportation hub, and the edge represents the specific diffusion path of the pathogen transmission chain; Set the pathogen type to be monitored, and based on the relevant symptoms caused by the pathogen type, exclude the health declaration data of passengers of non-monitored pathogen types, and analyze the passenger flow trend based on the passenger flow time series change data in the pathogen transmission chain; Based on passenger flow trends and screened passenger health declaration data, combined with environmental monitoring data, a comprehensive assessment is conducted to dynamically adjust the weight of each edge to characterize the path transmission intensity of pathogens; A graph search algorithm is used to analyze the propagation chain between nodes, and the risk priority of the propagation path is identified according to the order of the path propagation strength.

5. The method for predicting epidemic pathogens based on big data according to claim 4, characterized in that: In S4, based on the passenger health declaration data of each node and the passenger boarding and alighting records at each node, the dynamic relationship between nodes is tracked, and the dynamic main transmission path trajectory of pathogens in the spatiotemporal dynamic network is constructed in combination with the risk priority identification order of the transmission path.

6. The method for predicting epidemic pathogens based on big data according to claim 5, characterized in that: In S5, based on the change in path propagation intensity between the main path trajectory and the adjacent paths, the propagation chain expansion coefficients between all adjacent paths are iteratively fitted by a nonlinear regression method, and a regional propagation risk prediction model based on a weighted algorithm is established, which specifically includes: Obtain the dynamic main path trajectory of pathogens in the spatiotemporal dynamic network and the change in the path propagation intensity adjacent to the main path trajectory in real time; The sliding monitoring interval window size is set, and the correlation of the propagation intensity changes in the continuous sliding monitoring interval windows of all adjacent paths in the spatiotemporal dynamic network is counted. The propagation chain expansion coefficients between all adjacent paths are iteratively fitted with the main propagation path trajectory as the center point through the nonlinear regression method. According to the transmission characteristics of the pathogen type to be monitored and combined with the expansion coefficient between the transmission chains, a regional transmission risk prediction model based on a weighted algorithm is established.

7. The method for predicting epidemic pathogens based on big data according to claim 6, characterized in that: In S6, the prediction results of the regional transmission risk prediction model are compared and verified with the current actual calculation results in a loop, and the model parameters are adjusted until the model reaches the set accuracy value, specifically including: The regional transmission risk prediction model predicts the path transmission intensity of all paths within the next sliding monitoring interval window; Compare and verify the prediction results of the regional transmission risk prediction model in the previous sliding monitoring interval window with the current actual calculation results, adjust the model parameters according to the prediction deviation, and repeat the prediction and comparison adjustment process until the prediction accuracy of the model reaches the set accuracy value.

8. The method for predicting epidemic pathogens based on big data according to claim 7, characterized in that: In S7, the transmission intensity of the pathogen transmission paths in the set monitoring area is predicted in real time, the high-risk transmission paths of pathogens and the corresponding transportation hubs are displayed, and early warning of pathogen spread is provided for all transportation hubs in the high-risk transmission paths. Specifically, the following are included: Deploy the regional transmission risk prediction model to the server to predict the transmission intensity of the pathogen transmission path in the set monitoring area in real time; The warning levels are divided according to the set predicted path transmission intensity warning level gradient, and GIS is used to display the high-risk transmission paths of pathogens and corresponding transportation hubs in real time, and the transmission hotspot map is dynamically updated; Provide early warning of pathogen spread for transportation hubs corresponding to nodes in all high-risk transmission paths shown in the transmission hotspot map.

Images