Marathon event data processing system and method based on big data

Abstract

The invention relates to the technical field of competition data processing, in particular to a marathon competition data processing system and method based on big data, in the marathon competition data processing system and method based on big data, physical consumption and health risks of athletes are comprehensively evaluated by combining environmental factors, the body loads of the athletes can be comprehensively known, and the health of the athletes is improved. Therefore, possible risks in the match can be more effectively judged, decision support is provided, match track resources are dynamically adjusted by combining a health risk assessment result and match track condition analysis, state changes of athletes are analyzed by adopting an isolated forest algorithm, match resource allocation is dynamically adjusted, and match resource scheduling is more scientific and real-time. According to the method and the system, the running flexibility and the coping capacity of the competition are enhanced, the load and resource requirements of different competition segments can be accurately predicted through the real-time health data and the time sequence analysis of the long-short-term memory network, the competition resource allocation is more reasonable and scientific, excessive or insufficient allocation of the resources is avoided, and the accuracy and the real-time performance of the competition resource scheduling are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of event data processing, and in particular, to a marathon event data processing system and method based on big data. Background Art

[0002] The technical field of event data processing aims to optimize event management and enhance the participation experience through advanced computer technologies, big data analysis, and artificial intelligence algorithms. By efficiently collecting, storing, and processing data, it provides information such as the real-time location, performance, and health status of athletes, helps event organizers keep track of the event dynamics in real time, ensures the fairness and security of the event, and at the same time provides data support for subsequent event optimization and planning.

[0003] The purpose of a marathon event data processing system based on big data is to achieve precise monitoring and optimized management of the event process, improve the efficiency of event organization, ensure the fairness of the event, and provide real-time feedback on the athletes' exercise status and performance data, helping event managers make timely decisions. By collecting the athletes' exercise data in real time, it conducts intelligent analysis on various types of event data, provides trend prediction and risk control for the event, and achieves the effect of optimizing event operation management and enhancing the experience of participants.

[0004] The prior art lacks precise real-time monitoring of the physical state of athletes, resulting in the failure to detect abnormalities in the health status of athletes in a timely manner during the event, increasing the probability of sudden health problems. In addition, the response of the prior art in resource scheduling is relatively slow. The scheduling of event resources such as supply stations and medical resources depends on the preliminary design plan at the beginning of the event, lacking the ability of dynamic adjustment based on real-time data, and unable to make precise adjustments according to real-time data in the middle or late stage of the event, resulting in improper resource allocation and increasing event risks. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose a marathon event data processing system and method based on big data.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: A marathon event data processing system based on big data includes: Health monitoring module: Real-time collect the physiological data of athletes through sensors, monitor the physical state of athletes, compare it with preset health thresholds, and generate health anomaly data; Comprehensively evaluate the physical energy consumption and health risks of athletes in combination with environmental factors, provide decision-making support for judging possible risks in the event, and dynamically adjust track resources; Risk assessment and early warning module: Based on the health abnormal data, combined with environmental factors, analyze the physical energy consumption and potential health risks of athletes, and judge whether there are health hazards according to the preset threshold, and generate a health risk assessment result; Dynamic scheduling module for event resources: Based on the health risk assessment result, analyze the health status of athletes and the demand for event resources. At the same time, real-time detect the crowded areas of athletes on the track and the pressure on supply stations, adjust the positions of medical staff and supply stations, and generate an event resource allocation plan; Real-time event optimization module: According to the event resource allocation plan, combined with real-time health data, use the isolation forest algorithm to evaluate the abnormality of the load of supply stations, analyze the changes in the status of athletes, and dynamically adjust the number and layout of supply stations. At the same time, real-time update the positions of medical sites and rescue vehicles, and generate a real-time event scheduling plan; Track and resource allocation evaluation module: Based on the real-time event scheduling plan, use the long short-term memory network to predict the load and resource requirements of different segments. Combine the matching degree of the load situation and resource allocation of each segment to evaluate the rationality of the track and resource allocation, and adjust the event arrangement through the resource usage situation, and generate a track resource allocation evaluation result; Dynamic decision execution module: According to the track resource allocation evaluation result, analyze various real-time data, judge the rationality of the track resource allocation, make dynamic adjustments through real-time feedback and issue decision instructions, and generate decision execution instructions.

[0007] As a further solution of the present invention, the health monitoring module includes: Data acquisition sub-module: Based on sensor devices, real-time collect the heart rate, step frequency, body temperature and movement speed data of athletes, perform data cleaning and synchronization, and generate a real-time physiological data set; Health threshold comparison sub-module: Based on the real-time physiological data set, compare item by item with the preset physiological health threshold of athletes, identify the data exceeding the threshold, and generate a health threshold comparison result; Health abnormal data generation sub-module: Based on the health threshold comparison result, judge whether there are health abnormalities exceeding the threshold, mark the abnormal items, and generate health abnormal data.

[0008] As a further solution of the present invention, the risk assessment and early warning module includes: Physical energy consumption analysis sub-module: Based on the health abnormal data, estimate the physical energy consumption situation by analyzing the step frequency, movement speed and exercise intensity of athletes, and generate physical energy consumption data; Environmental factor analysis sub-module: Based on the health abnormal data, combined with the environmental data of the track, analyze the potential impact of the environment on the health status of athletes, and generate an environmental impact analysis result; Health risk assessment sub-module: Based on the physical energy consumption data and the environmental impact analysis results, combined with the preset health risk threshold, it determines whether there are health hazards and generates a health risk assessment result.

[0009] As a further solution of the present invention, the event resource dynamic scheduling module includes: Health status and demand analysis sub-module: Based on the health risk assessment result, it conducts real-time analysis on the health indicators of each athlete, and combines the current status and expected consumption of the athlete to calculate their respective resource demands, generating athlete resource demand data; Track density monitoring sub-module: Based on the athlete resource demand data, it monitors the positions and movements of athletes on the track in real time, obtains the distribution density of athletes through GPS data, predicts the track density areas, and generates track density distribution data; Resource allocation generation sub-module: Based on the track density distribution data, combined with the needs of athletes and the distribution of the track, it analyzes and optimizes the positions and quantities of medical staff and supply stations, adjusts the event resource allocation, and generates an event resource allocation plan.

[0010] As a further solution of the present invention, the real-time event optimization module includes: Event resource adaptation analysis sub-module: Based on the event resource allocation plan, it obtains the health data of athletes in real time, and combines with environmental changes to analyze the adaptation of resource allocation, generating resource adaptation data; Supply station dynamic adjustment sub-module: Based on the resource adaptation data, it monitors the demand changes of athletes, uses the isolation forest algorithm to analyze the load of supply stations, and dynamically adjusts the quantity, position and supply content of supply stations, generating a supply station adjustment plan; Track resource optimization sub-module: Based on the supply station adjustment plan, it obtains real-time updated event data, analyzes the distribution of medical stations and rescue vehicles, and adjusts the resource positions in real time, generating a real-time event scheduling plan.

[0011] As a further solution of the present invention, the isolation forest algorithm is calculated according to the formula:

[0012] Where: is the anomaly degree of the supply station load, is the supply station load data point is the path length in the isolation forest algorithm, is the total number of trees in the isolation forest, is the weight coefficient of the athlete's physical energy consumption, is the distance from the supply station to the current position of the athlete, is the importance coefficient of the supply station distance for load analysis, is the richness of the supplies provided by the supply station, is the influence weight coefficient of the supplies on the load change, is the influence of environmental factors on the load.

[0013] As a further solution of the present invention, the track and resource allocation evaluation module includes: Resource matching analysis sub-module: Based on the real-time event scheduling plan, using a long short-term memory network, analyze the load situation of each stage of the track, combine the preset resource demand data, quantify the resource allocation requirements for each stage, calculate the matching degree between the actual resources and the required resources, and generate resource matching degree data; Resource usage monitoring sub-module: Based on the resource matching degree data, collect and monitor the usage of various resources during the event in real time, including the consumption of supply stations, medical points, and rescue vehicles, calculate whether the actual consumption deviates from the expected allocation, and generate resource usage data; Configuration evaluation generation sub-module: Based on the resource usage data, analyze the rationality of the track and resource configuration, combine the track load and the actual usage of resources, determine whether the resource configuration needs to be adjusted, and generate the track resource configuration evaluation result.

[0014] As a further solution of the present invention, the long short-term memory network is calculated according to the formula:

[0015] Where: is the resource matching degree, is the stage The actually allocated resource quantity, is the stage The required resource quantity, is the total number of stages, is the stage The resource demand quantity affected by environmental factors, is the stage The weight coefficient of the resource quantity and the demand quantity, is the stage The resource demand quantity affected by the stage length and terrain factors, is the weight coefficient affected by environmental factors, is the weight coefficient affected by stage characteristics.

[0016] As a further solution of the present invention, the dynamic decision execution module includes: Resource rationality judgment sub-module: Based on the evaluation results of the track resource allocation, analyze the real-time data, combine the resource matching degree data and resource usage data, judge whether the current track resource allocation is reasonable, and generate resource rationality judgment data; Feedback adjustment sub-module: Based on the resource rationality judgment data, collect the actual feedback in the event in real time, and dynamically adjust the track and resources for the unreasonable part of the resource allocation to generate a resource adjustment plan; Decision instruction issuing sub-module: Based on the resource adjustment plan, issue adjustment instructions to each relevant department to ensure that the immediate changes in track resources can be accurately executed by each department, and generate decision execution instructions.

[0017] A big-data-based data processing method for marathon events. The big-data-based data processing method for marathon events is executed based on the above-mentioned big-data-based marathon event data processing system, and includes the following steps: Step 1: Based on the heart rate, stride frequency, body temperature and movement speed data of athletes, collect and conduct comparative analysis in real time, judge whether each data point exceeds the preset health threshold, perform anomaly identification, and generate health anomaly data; Step 2: Combine the health anomaly data, conduct correlation analysis based on environmental factors, calculate the physical energy consumption value of the athlete, and compare it with the health threshold to generate a health risk assessment result; Step 3: Based on the health risk assessment result, analyze the health status of the athlete and the event resource requirements, monitor the pressure on the concentrated areas of athletes and supply stations on the track in real time, and dynamically adjust the positions and quantities of medical staff and supply stations to generate an event resource allocation plan; Step 4: According to the event resource allocation plan, collect the health data of athletes in real time and conduct trend analysis of the changes, use the isolation forest algorithm to judge the health changes of athletes, dynamically adjust the layout of supply stations, and update the positions of medical stations and rescue vehicles in real time to generate a real-time event scheduling plan; Step 5: Based on the real-time event scheduling plan, apply the long short-term memory network, combine the load conditions of each stage and the matching degree of resource allocation, analyze the rationality of the track resource allocation, evaluate the resource usage, and combine the real-time feedback data to dynamically adjust the allocation plan and issue decision instructions to generate decision execution instructions.

[0018] Compared with the prior art, the advantages and positive effects of the present invention are as follows: 1. In the present invention, by comprehensively evaluating the physical energy consumption and health risks of athletes in combination with environmental factors, the physical load of athletes can be comprehensively understood, and then the risks that may occur in the event can be more effectively judged, providing decision-making support. By combining the health risk assessment results and the analysis of the track conditions, the track resources are dynamically adjusted to ensure that the medical resources and supply stations can meet the event requirements; 2. In the present invention, the isolated forest algorithm is used to analyze the state changes of athletes, dynamically adjust the allocation of event resources, make the scheduling of event resources more scientific and real-time, and enhance the flexibility and response ability of event operation; 3. In the present invention, through the time series analysis of real-time health data and long short-term memory network, the load and resource requirements of different race segments can be accurately predicted, making the allocation of event resources more reasonable and scientific, avoiding over-allocation or under-allocation of resources, and improving the accuracy and real-time nature of event resource scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is the system flow chart of the present invention; Figure 2 is the schematic diagram of the system framework of the present invention; Figure 3 is the schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present 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 only used to explain the present invention and are not used to limit the present invention.

[0021] Please refer to Figure 1 , the present invention provides a technical solution: a marathon event data processing system based on big data includes: Health monitoring module: Real-time collect the physiological data of athletes through sensors, monitor the physical state of athletes, compare with the preset health thresholds, and generate health abnormal data; Comprehensively evaluate the physical energy consumption and health risks of athletes in combination with environmental factors, provide decision-making support for judging the risks that may occur in the event, and dynamically adjust the track resources; Risk assessment and early warning module: Based on the health abnormal data, in combination with environmental factors, analyze the physical energy consumption and potential health risks of athletes, and judge whether there are health hazards according to the preset thresholds, and generate health risk assessment results; Event resource dynamic scheduling module: Based on the health risk assessment results, analyze the health status of athletes and the event resource requirements, and at the same time, real-time detect the dense areas of athletes on the track and the pressure of supply stations, adjust the positions of medical staff and supply stations, and generate event resource allocation plans; Real-time Event Optimization Module: According to the event resource allocation plan, combined with real-time health data, using the Isolation Forest algorithm, evaluate the abnormality of the supply station load, analyze the changes in the athletes' states, and dynamically adjust the quantity and layout of the supply stations. At the same time, update the positions of the medical stations and rescue vehicles in real time, and generate a real-time event scheduling plan; Track and Resource Allocation Evaluation Module: Based on the real-time event scheduling plan, using the Long Short-Term Memory network, predict the load and resource requirements of different segments, combine the matching degree of the load situation and resource allocation in each segment, evaluate the rationality of the track and resource allocation, and adjust the event arrangement through the resource usage situation, and generate the track resource allocation evaluation result; Dynamic Decision Execution Module: According to the track resource allocation evaluation result, analyze various real-time data, judge the rationality of the track resource allocation, make dynamic adjustments through real-time feedback and issue decision instructions, and generate decision execution instructions.

[0022] Please refer to Figure 2 , the health monitoring module includes: Data Acquisition Sub-module: Based on sensor devices, real-time collect the athletes' heart rate, stride frequency, body temperature and movement speed data, perform data cleaning and synchronization, and generate a real-time physiological data set; Health Threshold Comparison Sub-module: Based on the real-time physiological data set, compare item by item with the preset athletes' physiological health thresholds, identify the data exceeding the thresholds, and generate the health threshold comparison result; Health Abnormal Data Generation Sub-module: Based on the health threshold comparison result, judge whether there are health abnormalities exceeding the thresholds, mark the abnormal items, and generate health abnormal data; Data Acquisition Sub-module: Based on sensor devices, use the timed sampling method to perform real-time collection of the athletes' heart rate, stride frequency, body temperature and movement speed data, connect to the microcontroller through the serial communication protocol of the sensor device, and preprocess and store the data through a data caching mechanism. Use a calibration algorithm to filter the noise of the collected raw data, and perform data synchronization. Synchronize the data in chronological order through the timestamp algorithm, and generate a real-time physiological data set; Health Threshold Comparison Sub-module: Based on the real-time physiological data set, use the threshold judgment algorithm. Among them, the preset physiological health threshold data is compared with the real-time data through the linear interpolation algorithm, use the item-by-item comparison method to compare each physiological data with the preset threshold, and use conditional judgment statements to judge whether each data item exceeds the threshold, and generate the health threshold comparison result; Health Abnormality Data Generation Sub-module: Based on the health threshold comparison results, use anomaly detection algorithms to screen the data in the comparison results through the Z-score anomaly detection method, determine whether there are health abnormalities exceeding the threshold, and use a marking algorithm to mark the abnormal data items to generate health abnormality data.

[0023] Please refer to Figure 2 , the risk assessment and early warning module includes: Physical Energy Consumption Analysis Sub-module: Based on the health abnormality data, estimate the physical energy consumption situation by analyzing the athlete's stride frequency, movement speed, and exercise intensity, and generate physical energy consumption data; Environmental Factor Analysis Sub-module: Based on the health abnormality data, combine the environmental data of the track, analyze the potential impact of the environment on the athlete's health status, and generate environmental impact analysis results; Health Risk Assessment Sub-module: Based on the physical energy consumption data and environmental impact analysis results, combine the preset health risk threshold, determine whether there are health hazards, and generate health risk assessment results; Physical Energy Consumption Analysis Sub-module: Based on the health abnormality data, adopt the acceleration sensor data analysis algorithm, combine the athlete's stride frequency, movement speed, and exercise intensity data, perform weighted calculations on the stride frequency and movement speed by using a linear regression model, dynamically adjust the exercise intensity by using an exponential decay model, and estimate the physical energy consumption by using the Kalman filter algorithm to generate physical energy consumption data; Environmental Factor Analysis Sub-module: Based on the health abnormality data, combine the track environmental data, adopt the multiple regression analysis method, analyze the relationship between the athlete's health status and environmental variables, fit the environmental data by using the maximum likelihood estimation method, and evaluate the potential impact of the environment on the athlete's health by using the regression coefficient to generate environmental impact analysis results; Health Risk Assessment Sub-module: Based on the physical energy consumption data and environmental impact analysis results, combine the preset health risk threshold, use the decision tree algorithm to classify the physical energy consumption and environmental impact data, train the decision tree by using the CART algorithm, select the best split point by using the information gain method, and optimize the complexity of the decision tree by using the pruning algorithm to generate health risk assessment results.

[0024] Please refer to Figure 2 , the event resource dynamic scheduling module includes: Health Status and Demand Analysis Sub-module: Based on the health risk assessment results, perform real-time analysis on the health indicators of each athlete, and combine the athlete's current status and expected consumption to calculate their respective resource requirements and generate athlete resource requirement data; Track density monitoring sub-module: Based on the athlete resource demand data, it monitors the positions and movements of athletes on the track in real time, obtains the distribution density of athletes through GPS data, predicts the dense areas of the track, and generates track density distribution data; Resource allocation generation sub-module: Based on the track density distribution data, combined with the needs of athletes and the distribution of the track, it analyzes and optimizes the positions and quantities of medical staff and supply stations, adjusts the event resource allocation, and generates an event resource allocation plan; Health status and demand analysis sub-module: Based on the health risk assessment results, it uses a linear model to analyze the health indicators of each athlete in real time. Combining the current status and expected consumption of the athlete, it calculates their respective resource demands through multiple regression analysis, weights the athlete demands using a weighted average algorithm, and combines the physiological indicators and consumption expectations of the athlete to generate athlete resource demand data; Track density monitoring sub-module: Based on the athlete resource demand data, it monitors the positions and movements of athletes on the track in real time, obtains the position information of athletes in real time through the GPS data acquisition module, uses the K-means clustering algorithm to perform clustering analysis on the athlete distribution, uses the DBSCAN algorithm to detect the density of athletes on the track, predicts the dense areas of the track, and generates track density distribution data; Resource allocation generation sub-module: Based on the track density distribution data, combined with the athlete needs and the distribution of the track, it uses the integer programming algorithm to optimize the analysis of the positions and quantities of medical staff and supply stations, uses the simulated annealing algorithm to adjust the parameters of the resource allocation, and allocates the event resources through the greedy algorithm to generate an event resource allocation plan.

[0025] Please refer to Figure 2 , the real-time event optimization module includes: Event resource adaptation analysis sub-module: Based on the event resource allocation plan, it obtains the health data of athletes in real time, and combines with environmental changes to analyze the adaptation of resource allocation, and generates resource adaptation data; Supply station dynamic adjustment sub-module: Based on the resource adaptation data, it monitors the demand changes of athletes, uses the isolation forest algorithm to analyze the load of supply stations, dynamically adjusts the quantity, position and supply content of supply stations, and generates a supply station adjustment plan; Track resource optimization sub-module: Based on the supply station adjustment plan, it obtains real-time updated event data, analyzes the distribution of medical stations and rescue vehicles, and adjusts the resource positions in real time to generate a real-time event scheduling plan; Event Resource Adaptation Analysis Sub-module: Based on the event resource configuration plan, it obtains the athletes' health data in real time, combines with environmental changes, uses the multi-dimensional regression analysis method to fit the athletes' health data and environmental variables, evaluates the adaptation of various resources using the weighted summation algorithm, adapts various types of resources to the athletes' needs by calculating the correlation matrix, and generates resource adaptation data; Supply Station Dynamic Adjustment Sub-module: Based on the resource adaptation data, it monitors the changing needs of athletes, analyzes the supply station load data using the Isolation Forest algorithm, evaluates the supply station load using the anomaly detection method, conducts real-time monitoring by setting the sliding window algorithm, and dynamically adjusts the supply station locations using the K-means clustering algorithm to generate a supply station adjustment plan; Track Resource Optimization Sub-module: Based on the supply station adjustment plan, it obtains real-time updated event data, analyzes the distribution of medical stations and rescue vehicles using the resource allocation optimization algorithm, makes real-time adjustments to the resource locations using the simulated annealing algorithm, and optimizes the scheduling strategy of track resources using the genetic algorithm to generate a real-time event scheduling plan.

[0026] Isolation Forest algorithm, according to the formula:

[0027] Where: is the anomaly degree of the supply station load, is the supply station load data point The path length in the Isolation Forest algorithm, is the total number of trees in the Isolation Forest, is the weight coefficient of the athletes' physical energy consumption, is the distance from the supply station to the current location of the athlete, is the importance coefficient of the supply station distance for load analysis, is the richness of the supplies provided by the supply station, is the influence weight coefficient of the supplies on the load change, is the influence of environmental factors on the load; Execution process: First, collect the real-time data of athletes, including physical energy consumption, exercise intensity, and current location, and calculate the load status of the supply station. Then calculate the path length of each supply station data point. The shorter the path, the more abnormal the load. Then calculate the current distance between the supply station and the athlete, and assign a weight according to the influence of this distance. At the same time, evaluate the richness of supplies of the supply station, and adjust the load influence through the weight . Environmental factors are also included in the analysis and the weights are adjusted accordingly , finally, synthesize the parameters to obtain the abnormality degree of each supply station , and average through the evaluation results to generate a supply station adjustment plan to ensure the dynamic adjustment of the number, location, and provided material content of supply stations during the marathon event.

[0028] Please refer to Figure 2 , the track and resource allocation evaluation module includes:[[]] Resource matching analysis sub-module: Based on the real-time event scheduling plan, use the long short-term memory network to analyze the load situation of each stage of the track, combine the preset resource demand data, quantify the resource allocation requirements of each stage, calculate the matching degree between the actual resources and the required resources, and generate resource matching degree data; Resource usage monitoring sub-module: Based on the resource matching degree data, collect and monitor the usage of various resources during the event in real time, including the consumption of supply stations, medical points, and rescue vehicles, calculate whether the actual consumption deviates from the expected allocation, and generate resource usage data; Configuration evaluation generation sub-module: Based on the resource usage data, analyze the rationality of the track and resource configuration, combine the track load and the actual usage of resources, judge whether the resource configuration needs to be adjusted, and generate the track resource configuration evaluation result; Resource matching analysis sub-module: Based on the real-time event scheduling plan, use the long short-term memory network to analyze the load situation of each stage of the track, adjust the model weights through the backpropagation algorithm, combine the preset resource demand data, quantify the resource allocation requirements of each stage, and calculate the matching degree between the actual resources and the required resources by the weighted average method to generate resource matching degree data; Resource usage monitoring sub-module: Based on the resource matching degree data, collect and monitor the usage of various resources during the event in real time, obtain the consumption data of supply stations, medical points, and rescue vehicles through the sensor network, use the Kalman filter algorithm to smooth the resource usage data, and use the time series analysis method to calculate the deviation degree between the actual consumption and the expected allocation to generate resource usage data; Configuration evaluation generation sub-module: Based on the resource usage data, analyze the rationality of the track and resource configuration, combine the track load data and the actual usage of resources, evaluate the resource configuration by the weighted scoring method, and judge whether the resource configuration needs to be adjusted through the binary classification algorithm to generate the track resource configuration evaluation result.

[0029] The long short-term memory network, according to the formula:

[0030] where: is the resource matching degree, is the stage The actually allocated resource quantity, For a stage The required resource quantity, Is the total number of stages, For a stage The resource demand quantity affected by environmental factors, For a stage The weight coefficient of the resource quantity and the demand quantity, For a stage The resource demand quantity affected by the stage length and terrain factors, Is the weight coefficient affected by environmental factors, Is the weight coefficient affected by stage characteristics; Execution process: First, collect the actually allocated resources of each stage And the required resources , then adjust according to the environmental factors of the stage And stage characteristics . The influence of each factor on the resource demand is adjusted by the weight coefficients , And To ensure that the characteristics of different stages can be reflected in the resource matching degree calculation. Then sum the weighted resource demand values of each stage, and calculate the ratio with the required resource quantity of the corresponding stage. Finally, obtain the overall resource matching degree of all stages , providing a reference basis for the scheduling and optimization of event resources to ensure the reasonable allocation of event resources.

[0031] Please refer to Figure 2 , the dynamic decision-making execution module includes: Resource rationality judgment sub-module: Based on the evaluation results of the track resource configuration, analyze various real-time data, and combine the resource matching degree data and resource usage data to judge whether the current track resource configuration is reasonable, and generate resource rationality judgment data; Feedback adjustment sub-module: Based on the resource rationality judgment data, collect the actual feedback in the event in real time, and perform dynamic adjustment of the track and resources for the unreasonable part of the resource configuration, and generate a resource adjustment plan; Decision instruction release sub-module: Based on the resource adjustment plan, issue adjustment instructions to each relevant department to ensure that the immediate changes in track resources can be accurately executed by each department, and generate decision execution instructions; Resource Rationality Judgment Sub-module: Based on the evaluation results of track resource allocation, analyze various real-time data, combine resource matching degree data and resource usage data, evaluate the rationality of resource allocation by using multiple linear regression analysis, calculate the correlation between various resource allocations and actual usage by using the Pearson correlation coefficient, and use the threshold judgment algorithm to determine whether the resource allocation is reasonable, generating resource rationality judgment data; Feedback Adjustment Sub-module: Based on the resource rationality judgment data, collect the actual feedback in the event in real time, identify unreasonable resource allocations by using sensor feedback data and real-time monitoring data, use the dynamic adjustment algorithm to adjust the track and resource allocations in real time, and optimize the resource adjustment strategy by combining the genetic algorithm, generating a resource adjustment plan; Decision Instruction Publishing Sub-module: Based on the resource adjustment plan, issue adjustment instructions to each relevant department, timely transmit the adjustment instructions to each executing department by using the information flow scheduling algorithm, distribute the instructions by using the message queue algorithm, and ensure the accurate execution of the instructions by combining the transaction processing mechanism, generating decision execution instructions.

[0032] Please refer to Figure 3 , a method for processing marathon event data based on big data, which is executed based on the above-mentioned big data-based marathon event data processing system, including the following steps: Step 1: Based on the heart rate, cadence, body temperature and movement speed data of athletes, collect and compare them in real time, determine whether each data point exceeds the preset health threshold, perform anomaly identification, generating health anomaly data; Step 2: Combine the health anomaly data, conduct correlation analysis based on environmental factors, calculate the physical energy consumption value of athletes, and compare it with the health threshold, generating a health risk assessment result; Step 3: Based on the health risk assessment result, analyze the health status of athletes and the event resource requirements, monitor the athlete-dense areas and the pressure on supply stations on the track in real time, dynamically adjust the positions and quantities of medical staff and supply stations, generating an event resource allocation plan; Step 4: According to the event resource allocation plan, collect the health data of athletes in real time and conduct trend analysis, use the isolation forest algorithm to judge the health changes of athletes, dynamically adjust the layout of supply stations, and update the positions of medical sites and rescue vehicles in real time, generating a real-time event scheduling plan; Step 5: Based on the real-time event scheduling plan, apply the long short-term memory network, combine the load conditions of each stage and the matching degree of resource allocation, analyze the rationality of track resource allocation, evaluate the resource usage, and combine the real-time feedback data to dynamically adjust the allocation plan and issue decision instructions, generating decision execution instructions.

[0033] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the relevant art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A marathon event data processing system based on big data, characterized by: The system comprises: Health monitoring module: collects athletes’ physiological data in real time through sensors, monitors athletes’ physical condition, compares it with preset health thresholds, and generates health abnormality data; Combined with environmental factors, comprehensive assessment of athletes' physical energy consumption and health risks is carried out to provide decision-making support for judging possible risks in the event and dynamically adjust track resources; Risk assessment and early warning module: based on the abnormal health data and combined with environmental factors, thresholds are set and the athlete's physical energy consumption and potential health risks are analyzed to generate health risk assessment results; Dynamic scheduling module for event resources: Based on the health risk assessment results, it analyzes the health status of athletes and the demand for event resources, detects the pressure of athlete-dense areas and supply stations on the track in real time, adjusts the location of medical personnel and supply stations, and generates an event resource allocation plan; Real-time event optimization module: Based on the event resource allocation plan, combined with real-time health data, the isolation forest algorithm is used to evaluate the abnormality of the supply station load, analyze the changes in the athlete's status, and dynamically adjust the number and layout of the supply stations. At the same time, the location of medical stations and rescue vehicles is updated in real time to generate a real-time event scheduling plan; Track and resource allocation evaluation module: Based on the real-time event scheduling plan, the long short-term memory network is used to predict the load and resource requirements of different stages, and the rationality of the track and resource allocation is evaluated by combining the load situation of each stage with the matching degree of resource allocation. The event arrangement is adjusted according to the resource usage situation, and the track resource allocation evaluation result is generated; Dynamic decision execution module: According to the track resource configuration evaluation results, analyze various real-time data, judge the rationality of track resource configuration, make dynamic adjustments and issue decision instructions through real-time feedback, and generate decision execution instructions.

2. The marathon event data processing system based on big data according to claim 1 is characterized in that: The health monitoring module includes: Data collection submodule: collects athletes’ physiological data through sensor devices, cleans and synchronizes the data, and generates real-time physiological data sets; Health threshold comparison submodule: based on the real-time physiological data set, compare it item by item with the preset athlete's physiological health threshold, identify the data exceeding the threshold, and generate the health threshold comparison result; Health abnormality data generation submodule: Based on the health threshold comparison result, determine whether there is a health abnormality exceeding the threshold, mark the abnormal item, and generate health abnormality data.

3. The marathon event data processing system based on big data according to claim 1 is characterized in that: The risk assessment and early warning module includes: Physical energy consumption analysis submodule: based on the abnormal health data, by analyzing the athlete's step frequency, exercise speed and exercise intensity, the physical energy consumption is estimated and the physical energy consumption data is generated; Environmental factor analysis submodule: based on the abnormal health data and combined with the environmental data of the track, analyze the potential impact of the environment on the health status of the athletes and generate environmental impact analysis results; Health risk assessment submodule: Based on the physical energy consumption data and environmental impact analysis results, combined with the preset health risk threshold, determine whether there are health risks and generate health risk assessment results.

4. The marathon event data processing system based on big data according to claim 1 is characterized in that: The event resource dynamic scheduling module includes: Health status and demand analysis submodule: Based on the health risk assessment results, the health indicators of each athlete are analyzed in real time, and the respective resource requirements are calculated based on the athlete's current status and expected consumption to generate athlete resource demand data; Track density monitoring submodule: Based on the athlete resource demand data, monitor the athlete position and flow on the track in real time, obtain the athlete distribution density through GPS data, predict the track density area, and generate track density distribution data; Resource allocation generation submodule: Based on the track density distribution data, combined with the needs of athletes and the distribution of tracks, analyze and optimize the location and number of medical personnel and supply stations, adjust the allocation of event resources, and generate an event resource allocation plan.

5. The marathon event data processing system based on big data according to claim 1 is characterized in that: The real-time event optimization module includes: Event resource adaptation analysis submodule: Based on the event resource allocation plan, the health data of athletes is obtained in real time, and the adaptation of resource allocation is analyzed in combination with environmental changes to generate resource adaptation data; A supply station dynamic adjustment submodule: Based on the resource adaptation data, monitor the changes in athletes' needs, use the isolation forest algorithm to analyze the load of the supply station, dynamically adjust the number, location and supply content of the supply station, and generate a supply station adjustment plan; Track resource optimization submodule: Based on the supply station adjustment plan, obtain real-time updated event data, analyze the distribution of medical stations and rescue vehicles, adjust resource locations in real time, and generate a real-time event scheduling plan.

6. The marathon event data processing system based on big data according to claim 1 is characterized in that: The isolation forest algorithm, according to the formula: in: is the abnormality of the supply station load, Load data points for refueling stations Path length in the Isolation Forest algorithm, is the total number of trees in the isolation forest, is the weight coefficient of the athlete’s physical energy consumption, is the distance from the supply station to the athlete’s current location, is the importance coefficient of the distance to the supply station for load analysis, The richness of the supply content provided to the supply station, is the weight coefficient of the impact of replenishment content on load change, The impact of environmental factors on the load.

7. The marathon event data processing system based on big data according to claim 1 is characterized in that: The track and resource allocation evaluation module includes: Resource matching analysis submodule: Based on the real-time event scheduling scheme, the long short-term memory network is used to analyze the load of each stage of the track, and the resource allocation requirements of each stage are quantified in combination with the preset resource demand data, and the matching degree between the actual resources and the required resources is calculated to generate resource matching degree data; Resource usage monitoring submodule: Based on the resource matching data, real-time collection and monitoring of the usage of various resources during the event, including the consumption of supply stations, medical points and rescue vehicles, calculation of whether the actual consumption and allocation deviate from expectations, and generation of resource usage data; Configuration evaluation generation submodule: Based on the resource usage data, analyze the rationality of the track and resource configuration, combine the track load and the actual usage of resources, determine whether the resource configuration needs to be adjusted, and generate the track resource configuration evaluation result.

8. The marathon event data processing system based on big data according to claim 1 is characterized in that: The long short-term memory network is based on the formula: in: For resource matching, For the stage The actual amount of resources allocated, For the stage The amount of resources required, is the total number of stages, For the stage The demand for resources affected by environmental factors, For the stage The weight coefficient of resource quantity and demand quantity, For the stage The resource requirements are affected by the length of the stage and the terrain. is the weight coefficient of environmental factors, is the weight coefficient of the stage characteristics.

9. The marathon event data processing system based on big data according to claim 1, characterized in that: The dynamic decision execution module includes: Resource rationality judgment submodule: Based on the track resource configuration evaluation results, analyze various real-time data, combine resource matching data and resource usage data, judge whether the current track resource configuration is reasonable, and generate resource rationality judgment data; Feedback adjustment submodule: Based on the resource rationality judgment data, the actual feedback in the event is collected in real time, and the track and resources are dynamically adjusted for the unreasonable resource allocation part to generate a resource adjustment plan; Decision-making instruction issuing submodule: Based on the resource adjustment plan, issue adjustment instructions to relevant departments to ensure that the real-time changes of track resources can be accurately executed by various departments and generate decision-making execution instructions.

10. A method for processing marathon event data based on big data, characterized in that: The marathon event data processing system based on big data according to any one of claims 1 to 9 comprises the following steps: Step 1: Based on the athlete's heart rate, cadence, body temperature and exercise speed data, real-time collection and comparative analysis are performed to determine whether each data point exceeds the preset health threshold, perform abnormality identification, and generate health abnormality data; Step 2: Combine the abnormal health data, perform correlation analysis based on environmental factors, calculate the athlete's physical energy consumption value, and compare it with the health threshold to generate a health risk assessment result; Step 3: Based on the health risk assessment results, analyze the athletes’ health status and event resource requirements, monitor the athlete-dense areas and supply station pressures on the track in real time, dynamically adjust the location and number of medical personnel and supply stations, and generate an event resource allocation plan; Step 4: According to the event resource allocation plan, the health data of athletes are collected in real time and the trend of changes is analyzed. The isolation forest algorithm is used to determine the changes in the health of athletes, the layout of the supply stations is dynamically adjusted, the locations of medical stations and rescue vehicles are updated in real time, and a real-time event scheduling plan is generated; Step 5: Based on the real-time event scheduling plan, the long short-term memory network is applied, and the matching degree of the load conditions of each stage and the resource configuration is combined to analyze the rationality of the track resource configuration, evaluate the resource usage, and dynamically adjust the configuration plan in combination with the real-time feedback data, and issue decision instructions to generate decision execution instructions.

Images