Smart public transportation talent management platform and identification method

Abstract

The present invention discloses a smart public transportation preferential talent management platform and identification method, which specifically relates to the field of public transportation travel technology, and is used to solve the problem of insufficient efficiency of identity authentication and preferential qualification verification in high-density crowd gathering scenes in the prior art; by gridding crowd position data in a target area, calculating local information entropy values ​​and generating entropy value time series, evaluating multi-scale entropy fluctuations, and combining discrete degree analysis of residence time data, it is judged whether there is a potential risk of aggregation effect in the area; if a risk is detected, a processing efficiency improvement mechanism will be triggered; at the same time, the diffusion trend is identified through path direction consistency analysis, and the user's identity information and preferential qualifications are tracked in real time and dynamically to ensure the continuity of verification during the user's movement.

Description

Technical Field

[0001] The present invention relates to the field of public transportation technology, and more specifically, to a smart public transportation talent management platform and identification method. Background Art

[0002] In the existing smart public transportation system, user identity authentication and preferential qualification verification have become important links to ensure smooth travel and enjoy specific discounts. Among them, the "public transportation" in smart public transportation not only refers to the urban public transportation system, but also covers a variety of travel scenarios such as entry verification to public attractions such as gardens and parks, and riding buses and subways. "Huitai" refers to preferential or convenient measures provided to specific groups of people (such as students, etc.), so that they can enjoy exclusive travel discounts or convenient services. During peak hours, weekends, and holidays, there are often a large number of users traveling in places such as transportation hubs and public attractions. Existing technologies usually rely on fixed verification and verification processing mechanisms to verify the identity and preferential qualifications of users who meet the requirements of Huicai through centralized verification methods.

[0003] In practice, if we cannot promptly respond to the problem of insufficient processing efficiency of identity authentication and preferential qualification verification in scenarios with high-density crowds, users will have to queue and wait during the verification process, affecting travel efficiency. The continuity of identity information and qualifications of high-density crowds during dynamic dispersion cannot be guaranteed.

[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, the embodiments of the present invention provide a smart public transportation talent management platform and an identification method 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: The smart public transportation talent identification method comprises the following steps: By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area. By monitoring the length of stay data of users in the target area, analyzing the dispersion of the length of stay, and evaluating the convergence distribution of the length of stay of people in the target area; Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area; When there is a potential risk of aggregation effect in the target area, it triggers the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests; at the same time, by collecting the path direction data of users in the target area, using fuzzy clustering algorithm to analyze the direction consistency, and evaluate the fluctuation of path direction consistency in the target area; Based on the fluctuation of the consistency of the path direction in the target area, it is judged whether the target area has a tendency of the diffusion effect to begin. When the target area has a tendency of the diffusion effect to begin, the user's identity information and preferential qualifications are continuously and dynamically tracked.

[0007] In a preferred embodiment, by gridding the real-time crowd location data in the target area, calculating the local information entropy value of each grid and generating an entropy value time series, the multi-scale entropy fluctuation of the target area is evaluated, specifically: The sensor network collects the location data of people in the target area in real time and records the real-time location coordinates of each user in the grid; Divide the target area into multiple grids according to a preset size and construct a spatial position model for each grid; According to the distribution of crowd positions in the grid, calculate the local information entropy value of each grid; Among them, the calculation formula of local information entropy value is: ;in, Indicates Line The local information entropy value of the column grid, For the The number of different crowd positions observed within the grid, Indicates The probability of a position, and Represents the row and column position index of the grid, The numbers representing the different population locations observed within a particular grid; The local information entropy value of each grid is recorded in real time to generate the entropy value time series of each grid within the specified time range; The multi-scale analysis method is used to conduct multi-scale fluctuation assessment on the entropy value time series of each grid, and the multi-scale entropy fluctuation in the target area is quantitatively assessed.

[0008] In a preferred embodiment, a multi-scale analysis method is used to perform a multi-scale fluctuation assessment on the entropy value time series of each grid, and a multi-scale entropy fluctuation in the target area is quantitatively assessed, specifically: Set the time scale set, which is expressed as: ,in, is the set of time scales, Indicates The time scale of the hierarchy, Indicates the time scale level number in multiscale analysis; Calculate the multiscale entropy fluctuation index, which is expressed as: ;in, is the multiscale entropy fluctuation index, Indicates Line Column grid at time points The local information entropy value of is the index variable of the current step in the time scale, Indicates time, It refers to the The time range selected in the hierarchical time scale is used as the upper limit of each sliding step in the time series; Set the multiscale entropy fluctuation threshold; compare the multiscale entropy fluctuation index with the multiscale entropy fluctuation threshold: When the multiscale entropy fluctuation index is greater than the multiscale entropy fluctuation threshold, the risk of entropy analysis aggregation effect is judged to be high; when the multiscale entropy fluctuation index is less than or equal to the multiscale entropy fluctuation threshold, the risk of entropy analysis aggregation effect is judged to be normal.

[0009] In a preferred embodiment, by monitoring the residence time data of users in the target area, analyzing the dispersion of the residence time, and evaluating the convergence distribution of the residence time of people in the target area, specifically: Collect the residence time data of each user in the target area in real time and pre-process the collected residence time data; Calculate the discrete degree of user stay time in the target area and use the variance formula to obtain the discrete distribution value of user stay time; Based on the discrete distribution value of user stay time, the concentration of user stay time is analyzed to evaluate the convergence distribution of the stay time of people in the target area: if the discrete distribution value of user stay time is less than the preset threshold , it is judged that the user's stay time is convergent; if the discrete distribution value of the user's stay time is greater than or equal to the preset threshold , it is judged that the user's stay time does not have convergence.

[0010] In a preferred embodiment, based on the multi-scale entropy fluctuation of the target area and the convergent distribution of the length of stay of people in the target area, it is judged whether the target area has a potential risk of aggregation effect, specifically: When the entropy analysis aggregation effect risk is high and the discrete distribution value of the user's stay time is less than the preset threshold When , it is determined that the target area has a potential risk of aggregation effect.

[0011] In a preferred embodiment, when there is a potential risk of aggregation effect in the target area, the requirement for improving the efficiency of processing identity authentication and preferential qualification verification requests is triggered, specifically: When a potential risk of aggregation effect is detected in the target area, an early warning signal is immediately triggered; After receiving the early warning signal, resources are automatically allocated dynamically; Adjust the priority sorting for identity verification and preferential eligibility verification; divide verification requests into batches and adopt parallel processing.

[0012] In a preferred embodiment, the path direction data of users in the target area are collected, and the direction consistency is analyzed using a fuzzy clustering algorithm to evaluate the fluctuation of the path direction consistency in the target area, specifically: Collect path direction data in real time, which includes the user's location information and path direction at different time points; The path direction data is input into the fuzzy clustering algorithm model, and the user groups are divided into high consistency groups and low consistency groups according to the consistency of the path direction; The proportion of the low consistency group to the total number of users is calculated, and the path direction consistency volatility is obtained as an indicator for evaluating the change of path direction consistency.

[0013] In a preferred embodiment, judging whether the target area has a trend of the diffusion effect about to begin based on the fluctuation of the consistency of the path direction in the target area is as follows: Compare the path direction consistency volatility with the preset threshold Compare and determine whether there is a trend of diffusion effect about to begin in the target area: If the path direction consistency volatility is greater than or equal to the preset threshold , it is determined that the target area has a tendency to start diffusion effect; If the path direction consistency volatility is less than the preset threshold , it is determined that the target area has no tendency of diffusion effect temporarily.

[0014] In a preferred embodiment, when the target area has a tendency to start a diffusion effect, the user's identity information and preferential qualifications are continuously and dynamically tracked, specifically: Monitor the user's latest location in real time; continuously and dynamically synchronize the user's identity information during the user's movement; update the user's preferential eligibility status in real time; when the user moves to an adjacent area, synchronize the identity information and preferential eligibility status to the adjacent area.

[0015] On the other hand, the present invention provides a smart public transportation talent management platform, including a multi-scale entropy analysis module, a stay time analysis module, an aggregate risk judgment module, a request processing enhancement module, a path consistency assessment module, a diffusion trend judgment module, and an identity qualification tracking module; Multi-scale entropy analysis module: by gridding the real-time crowd location data in the target area, calculating the local information entropy value of each grid and generating an entropy value time series, the multi-scale entropy fluctuation of the target area is evaluated; Dwell time analysis module: by monitoring the dwell time data of users in the target area, analyzing the dispersion of dwell time, and evaluating the convergence distribution of dwell time of people in the target area; Aggregation risk judgment module: Based on the multi-scale entropy fluctuations of the target area and the convergence distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area; Request processing improvement module: When there is a potential risk of aggregation effect in the target area, it triggers the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests; Path consistency evaluation module: collects path direction data of users in the target area, uses fuzzy clustering algorithm to analyze direction consistency, and evaluates the fluctuation of path direction consistency in the target area; Diffusion trend judgment module: judges whether the target area is about to start a diffusion effect based on the fluctuation of the path direction consistency in the target area; Identity qualification tracking module: When there is a trend of diffusion effect about to begin in the target area, the user's identity information and preferential qualifications are continuously and dynamically tracked.

[0016] Technical effects and advantages of the smart public transportation talent management platform and identification method of the present invention: 1. By combining the grid processing of crowd location data, the discrete analysis of the length of stay, and the dynamic monitoring of the consistency of the path direction, the processing efficiency of identity authentication and preferential qualification verification in high-density crowd gathering scenarios can be effectively improved. First of all, the present invention can identify the potential risks of aggregation effects in real time through the convergence evaluation of multi-scale entropy fluctuations and length of stay, trigger the processing efficiency improvement mechanism in advance during peak hours or crowded places, dynamically allocate system resources, and effectively respond to high-concurrency verification needs. Different from the traditional fixed processing mechanism, this method can intelligently identify and start the optimization of the verification process when the risk increases, reduce the waiting time for verification, ensure the smoothness of the identity authentication and preferential verification process of users in travel scenarios, and improve the overall travel efficiency.

[0017] 2. The present invention monitors the consistency of the user path direction in the target area, timely determines and identifies the beginning trend of the diffusion effect, and starts a continuous dynamic tracking mechanism during the dynamic dispersion of users. This mechanism ensures that when users move from the gathering area to different directions, identity authentication and preferential qualification information can be updated in real time and seamlessly connected, avoiding verification delays or interruptions caused by location changes, and ensuring the continuity and accuracy of verification. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a schematic diagram of the intelligent public transportation talent identification method of the present invention; Figure 2 It is a structural diagram of the smart public transportation talent management platform of the present invention. DETAILED DESCRIPTION

[0019] 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. Example

[0020] Figure 1 The present invention provides a smart public transportation talent identification method, which includes the following steps: By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area.

[0021] By monitoring the residence time data of users in the target area, analyzing the dispersion of the residence time, and evaluating the convergent distribution of the residence time of people in the target area.

[0022] Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area.

[0023] When there is a potential risk of aggregation effect in the target area, the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests is triggered; at the same time, by collecting the path direction data of users in the target area, the fuzzy clustering algorithm is used to analyze the direction consistency, and the fluctuation of the path direction consistency in the target area is evaluated.

[0024] Based on the fluctuation of the consistency of the path direction in the target area, it is judged whether the target area has a tendency of the diffusion effect to begin. When the target area has a tendency of the diffusion effect to begin, the user's identity information and preferential qualifications are continuously and dynamically tracked.

[0025] By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area, specifically: The sensor network collects the location data of people in the target area in real time and records the real-time location coordinates of each user in the grid: The sensor network installed in the target area collects the location data of all people in the area in real time. The sensors used can be infrared sensors, laser positioning devices or cameras that support positioning functions to ensure the real-time and accuracy of the crowd location data. Each user will be detected by the sensor after entering the target area, and the user's location data in the target area will be collected in the form of a coordinate pair.

[0026] Among them, the target area refers to the specific location or place where the system monitors crowd gathering and movement activities, usually key locations where crowd gathering may occur, such as bus hubs, subway stations, and scenic spot entrances.

[0027] Divide the target area into multiple grids according to the preset size and construct the spatial position model of each grid: Specifically, assuming that the target area is a rectangular area with a width of 10m and a length of 20m, the target area is divided into 5*4 grid units at equal intervals. The side lengths of each grid unit are 2m and 5m respectively. The spatial location model in each grid unit will record the location distribution and other characteristic parameters of the users in the grid. This division method ensures the refined processing of the location distribution in the target area.

[0028] According to the distribution of crowd positions within the grid, the local information entropy value of each grid is calculated to quantify the randomness and density of crowd distribution: The calculation of local information entropy value can be achieved by the following formula: ;in, Indicates Line The local information entropy value of the column grid, For the The number of different crowd positions observed within the grid, Indicates The probability of a position (that is, the ratio of the number of people at that position to the total number of people in the grid), and Represents the row and column position index of the grid, Numbers representing the different population locations observed within a particular grid.

[0029] The local information entropy value of each grid is recorded in real time to generate the entropy value time series of each grid within the specified time range: Specifically, a time interval is set, and the local information entropy value of each grid is recorded once every time interval.

[0030] The entropy value time series records the changes in information entropy of each grid at each time point, which is helpful for fluctuation assessment in subsequent multi-scale analysis.

[0031] The multi-scale analysis method is used to evaluate the multi-scale fluctuation of the entropy value time series of each grid, and the multi-scale entropy fluctuation in the target area is quantitatively evaluated: Specifically, the multi-scale analysis method is used to process the entropy time series and calculate its fluctuations at different time scales.

[0032] Set the time scale set, which is expressed as: ,in, is the set of time scales, Indicates The time scale of the hierarchy, Indicates the time scale level number in multiscale analysis.

[0033] Calculate the multiscale entropy fluctuation index, which is expressed as: ;in, is the multiscale entropy fluctuation index, which indicates that Line Column grid based on time scale The calculated average information entropy value; Indicates Line Column grid at time points The local information entropy value of is the index variable of the current step in the time scale; Indicates time; It refers to the The time range selected in the hierarchical time scale is used as the upper limit of each sliding step in the time series. For example, when , the range is 0 to 4.

[0034] By analyzing the entropy value at each time scale, the multi-scale entropy change characteristics of the target area can be evaluated, providing a basis for judging the aggregation effect of the target area.

[0035] The larger the multiscale entropy fluctuation index, the more intense the multiscale entropy fluctuation in the target area, indicating that the uncertainty and randomness of the population distribution in the grid are higher. This means that the activities of the population in the area are more dispersed and more mobile, and there may be potential aggregation effects or dynamic changes.

[0036] Set the multiscale entropy fluctuation threshold; compare the multiscale entropy fluctuation index with the multiscale entropy fluctuation threshold: When the multiscale entropy fluctuation index is greater than the multiscale entropy fluctuation threshold, it is judged that the risk of entropy analysis aggregation effect is high; this indicates that the distribution uncertainty and randomness of the population in the region are high, and the activities of the population tend to be complex and dispersed, which increases the risk of large-scale aggregation.

[0037] When the multiscale entropy fluctuation index is less than or equal to the multiscale entropy fluctuation threshold, the risk of entropy analysis aggregation effect is judged to be normal; this means that the activities of people in the area are relatively stable and controllable, the distribution is low in randomness, and has not reached a high-risk aggregation state.

[0038] Among them, the multiscale entropy fluctuation threshold is a reference value set according to historical data, environmental characteristics and crowd flow patterns in the target area. Specifically, the system collects fluctuation data of the multiscale entropy fluctuation index in the daily operation of the target area, and records the multiscale entropy fluctuations under normal crowd activities and high-density aggregation. Through statistical analysis of these data, a benchmark fluctuation value reflecting the aggregation trend is determined, namely the multiscale entropy fluctuation threshold.

[0039] By monitoring the length of stay data of users in the target area, analyzing the dispersion of the length of stay, and evaluating the convergence distribution of the length of stay of people in the target area, specifically: Collect the residence time data of each user in the target area in real time, and pre-process the collected residence time data: The residence time data of each user is collected in real time through sensor devices, which can be infrared sensors and video surveillance cameras. Specifically, the sensor records the time when the user enters the target area and the time when the user leaves the target area. The residence time of each user can be defined as the difference between the entry and exit time, which is used to characterize the residence time of the user in the target area. The system records the residence time data of each user in the database for subsequent data processing and analysis.

[0040] Preprocess the collected dwell time data: detect and remove outliers in the data. Outliers can be identified by setting a reasonable range for dwell time (for example, setting upper and lower limits based on historical data). When a user's dwell time data exceeds this range, the data is marked as an outlier and removed. Secondly, for missing data caused by equipment or network problems, an interpolation algorithm is used to fill in the missing dwell time data. The interpolation method can use mean interpolation or nearest neighbor-based interpolation to ensure the integrity of the dwell time data to the greatest extent. The preprocessed dwell time data will be further used for the calculation of the degree of discreteness.

[0041] Calculate the discrete degree of user stay time in the target area, and use the variance formula to obtain the discrete distribution value of user stay time: Based on the preprocessed dwell time data, the discrete distribution value of the user's dwell time is calculated to quantify the dispersion or concentration of the user's dwell time. The specific formula is: ;in, It is the discrete distribution value of user’s residence time, which reflects the dispersion degree of residence time in the whole user group; Indicates the total number of all users in the target area; In the target area The length of stay of each user; The average length of time users stay in the target area; The ID of the user in the target area.

[0042] The discrete distribution value of user residence time reflects the distribution fluctuation of user residence time in the target area.

[0043] If the discrete distribution value of user stay time is large, it means that the difference in user stay time is large; if the discrete distribution value of user stay time is small, it means that the user stay time is relatively concentrated. The calculation of dispersion can provide a basis for the subsequent analysis of the convergence of stay time.

[0044] Based on the discrete distribution values ​​of user dwell time, the concentration of user dwell time is analyzed to evaluate the convergence distribution of dwell time of people in the target area: Based on the calculated discrete distribution value of user stay time, the concentration of user stay time is further analyzed. When the discrete distribution value of user stay time is small, it indicates that the user stay time is concentrated in the target area; while a larger discrete distribution value of user stay time indicates that the user stay time distribution is more dispersed. Therefore, according to the size of the discrete distribution value of user stay time, the convergence of the stay time of people in the target area can be evaluated.

[0045] Specifically, if the discrete distribution value of the user's stay time is less than the preset threshold , then it is judged that the user stay time is similar, that is, most users stay in the target area for a similar length of time. Conversely, if the discrete distribution value of the user stay time is greater than or equal to the preset threshold , it is judged that the user stay time does not have convergence, that is, there are significant differences in the user stay time.

[0046] If the user residence time is similar, it means that the residence time of the crowd in the area is similar; if the user residence time is not similar, it means that there are large differences in the residence time of the crowd.

[0047] The preset threshold It is a reference value obtained based on the statistical analysis of the historical stay time data of the target area. Specifically, by collecting a large amount of stay time data of people under normal circumstances, the mean and standard deviation of the discrete distribution of the stay time are calculated, and a critical value that reflects the convergence of the crowd is set. Usually, the preset threshold It is set as the mean plus a certain multiple of the standard deviation to ensure that once the dispersion is lower than the threshold, it means that the population’s stay time is concentrated and the convergence is strong, thus serving as a basis for judging the aggregation effect.

[0048] Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area, specifically: When the entropy analysis aggregation effect risk is high and the discrete distribution value of the user's stay time is less than the preset threshold When , it is determined that the target area has a potential risk of aggregation effect.

[0049] It is more rigorous to use the "and" condition, because the simultaneous satisfaction of these two conditions can more clearly indicate the possibility of aggregation effect. The large multiscale entropy fluctuation index indicates that the distribution of the population is highly random and fluid, while the low dispersion of the residence time indicates that the residence time of users is converging. The simultaneous satisfaction of the two conditions indicates that the population is not only randomly distributed in terms of location, but also relatively concentrated in terms of residence time. The risk of aggregation effect is more significant, and thus the judgment is more accurate. This can reduce misjudgments, issue risk warnings under truly high aggregation risks, and improve the accuracy of judgments.

[0050] When there is a potential risk of agglomeration effect in the target area, requirements for improving the efficiency of processing identity verification and preferential qualification verification requests are triggered, specifically: When a potential risk of aggregation effect is detected in the target area, an early warning signal is immediately triggered to enter a state of improving the efficiency of processing requests: This early warning signal will trigger subsequent steps such as resource allocation, sorting adjustment and parallel processing to cope with the large-scale identity verification and preferential verification needs brought about by high-density crowds.

[0051] After receiving the early warning signal, resources are automatically allocated dynamically, allocating more computing resources to improve the processing capacity of identity authentication and preferential qualification verification: Specifically, based on the warning level and the number of requests, the total amount of computing resources required for the current load is calculated, and additional resources are scheduled to supplement processing capacity. Resource allocation is dynamically adjusted based on the load situation and the state of the resource pool, calling more processor cores, distributed computing resources, and storage resources to improve verification efficiency. For example, the number of processor threads can be increased, and idle processing nodes in distributed resources can be called to ensure processing stability and efficiency under high concurrency requests. The resource allocation process aims to achieve rapid response to large-scale requests by scheduling additional resources to ensure sufficient processing capacity to deal with emergencies.

[0052] Adjust the priority sorting for identity verification and benefit eligibility verification; batch verification requests and process them in parallel to increase the overall processing speed of identity verification and benefit verification: After the dynamic allocation of resources is completed, the current identity authentication and preferential qualification verification requests are prioritized to ensure that urgent or important verification requests are processed first. Specifically, the requests will be re-ordered according to the source of the request, the user identity category (urgent needs, etc.), or the timeliness of the current verification needs. The sorting algorithm can be based on preset priority rules, and requests with higher priorities are assigned to the priority processing queue so that they are processed first when resources are sufficient, ensuring that urgent needs are responded to first in high-density crowd scenarios. The adjusted requests will enter the parallel processing process in turn to ensure the continuity and effectiveness of the processing.

[0053] Based on the reordered request queue, the verification requests are divided into batches for parallel processing. In high-density scenarios, processing requests in batches in parallel can greatly improve the overall processing speed. In specific operations, the requests are grouped according to the set batch size, and enter the parallel processing process according to the batches. Each batch is assigned to an independent computing thread or node for processing. The multi-threaded architecture or distributed resources that the parallel processing method relies on can process multiple request batches at the same time, improving the responsiveness in high-concurrency scenarios. This batch division and parallel processing design ensures that in the case of a sudden increase in the number of verification requests, not only can the processing speed be improved, but the load can also be effectively dispersed to avoid resource overload.

[0054] By collecting the path direction data of users in the target area, the fuzzy clustering algorithm is used to analyze the direction consistency and evaluate the fluctuation of the path direction consistency in the target area. Specifically: Collect path direction data in real time. The path direction data includes the user's location information and path direction at different time points: Get the user at a certain time Location information .

[0055] By calculating the position difference between adjacent time points, the user's moving direction is obtained, which is expressed as: ;in, It's time point The user's movement direction, and Respectively represent the user at a time point The horizontal and vertical coordinates of and Respectively represent the previous time point The horizontal and vertical coordinates of .

[0056] Through continuous collection, the path direction data of each user over a period of time is obtained, providing a basis for subsequent analysis.

[0057] The path direction data is input into the fuzzy clustering algorithm model, and the user groups are divided into high consistency groups and low consistency groups according to the consistency of the path direction: Organize the route direction data of all users within a specific time window , forming a data set ;in, Indicates The path direction of each user, is the total number of users, The user ID.

[0058] According to the analysis requirements, the number of clusters is set to 2, that is, the user group is divided into two categories: a high consistency group and a low consistency group.

[0059] Initialize the membership matrix, expressed as: ;in, The membership matrix, Indicates Users belong to The membership of a cluster satisfies the condition and ;in, is the number of clusters, is the cluster number.

[0060] Using the fuzzy C-means algorithm, iterative update is performed according to the following steps: Calculate cluster centers: ;in, It is The center of the cluster, is the fuzzy index (also called fuzzification coefficient) in the fuzzy clustering algorithm, is the fuzzy index, usually .

[0061] Compute the distance metric: ;in, For the The path direction of the user is The distance between cluster centers.

[0062] Update the membership matrix: ;in, Indicates Users belong to The membership degree of a cluster reflects the consistency between the user path direction and the cluster center, and its value ranges from 0 to 1; For the User and The distance between cluster centers; Represents the index used to calculate the distance between the user path direction data and each cluster center; Iteration termination condition: When the change in the membership matrix is ​​less than the preset threshold When , the iteration process terminates.

[0063] The preset threshold is a small positive number used to control the convergence of the algorithm.

[0064] The proportion of the low consistency group to the total number of users is calculated, and the path direction consistency volatility is obtained as an indicator for evaluating the change of path direction consistency.

[0065] After completing fuzzy clustering, calculate the proportion of low consistency groups to the total number of users, and obtain the path direction consistency volatility as an indicator for evaluating changes in path direction consistency. The specific calculation is as follows: Set the low consistency group to the corresponding cluster number , then the number of users in the low consistency group is calculated as: ;in, is the number of users in the low consistency group, For the The membership degree of a user to a low consistency group.

[0066] Calculate path direction consistent volatility: ;in, is the path direction consistency volatility, which reflects the proportion of low consistency groups in the total users.

[0067] The greater the path direction consistency fluctuation rate, the more inconsistent the path direction consistency fluctuation in the target area. This means that the proportion of low consistency groups has increased, indicating that the path directions of the crowd are more dispersed and diverse, which may indicate that the crowd is spreading in different directions.

[0068] Based on the fluctuation of path direction consistency in the target area, it is judged whether the target area has a trend of diffusion effect about to begin, specifically: Compare the path direction consistency volatility with the preset threshold Compare and determine whether there is a trend of diffusion effect about to begin in the target area: If the path direction consistency volatility is greater than or equal to the preset threshold , it is determined that the target area has a tendency that the diffusion effect is about to begin.

[0069] If the path direction consistency volatility is less than the preset threshold , it is determined that the target area has no tendency of diffusion effect temporarily.

[0070] Preset Threshold It is set based on historical data and empirical analysis. Usually, a critical value is selected. When the path direction consistency volatility exceeds this value, it indicates that the proportion of low-consistency groups is high, which may indicate that the crowd is beginning to disperse in different directions.

[0071] When the target area has a tendency to start a diffusion effect, the continuous and dynamic tracking of the user's identity information and preferential qualifications is to maintain the continuity and accuracy of the verification process when the user moves rapidly and disperses. The diffusion of the crowd means that the user may quickly leave the current area or move in different directions. At this time, if real-time tracking is not performed, the information of identity verification and preferential qualification verification may be invalid or delayed, resulting in verification interruption or verification errors, which in turn affects the user's experience and travel efficiency. Through dynamic tracking, the location information and identity status of each user can be updated in time to ensure that the verification of identity verification and preferential qualifications can be completed smoothly even during the movement, avoiding delays or verification failures caused by changes in user location.

[0072] When the target area shows a trend of diffusion effect, the user's identity information and preferential qualifications are continuously and dynamically tracked, specifically: The user's latest location is monitored in real time through sensors and positioning devices to ensure accurate location information is obtained when the user moves for subsequent verification.

[0073] As users move around, the user's identity information is continuously and dynamically synchronized to ensure a consistent authentication state across different locations and nodes.

[0074] Update the user's preferential eligibility status in real time to ensure the continuity of preferential eligibility during user movement and prevent verification interruptions due to location changes.

[0075] When a user moves to other adjacent areas, the identity information and discount eligibility status will be synchronized to the adjacent areas to ensure seamless connection of verification status in different areas.

[0076] The user's location information, identity information and preferential eligibility status are distributedly stored and managed to ensure the integrity and continuity of data in dynamic tracking.

[0077] The aggregation effect refers to the fact that at certain times and places, users' travel behaviors will quickly concentrate on the same transportation hub or place, such as bus stops, subway stations, or scenic spot entrances. In this case, a large number of users may try to verify their identities or use preferential qualifications at the same time in a short period of time. The diffusion effect refers to the rapid dispersion of people from a gathering point to different travel destinations, such as entering a car from a subway platform or getting on a bus from a bus stop. This rapid dynamic change requires the system to be able to continuously and dynamically track users' identity information and preferential qualifications to avoid verification delays or interruptions caused by user movement.

[0078] Among them, in the verification of user identity information and preferential qualifications, it can be achieved in the following ways: Users choose to enter real-name information through OCR on their mobile phones. The system automatically recognizes the user's identity information and stores it to ensure the authenticity and accuracy of the identity information.

[0079] The user selects the Alipay authorization method, and the system obtains the user's relevant information from Alipay and performs verification. This verification is used to confirm the correspondence between the user's identity and the Alipay account.

[0080] The user selects the China Higher Education Student Information and Career Center student status verification code for manual verification. The system obtains the user's student status information through the China Higher Education Student Information and Career Center interface and verifies the authenticity of the information to confirm the user's student status.

[0081] The user selects the precise authorization method of the State Council Office for identity authentication. The system completes the user's liveness detection through the real person authentication SDK and generates an authorization signature through the authorization issuance SDK. The background authorization submission function submits the user data to the State Council Office for verification.

[0082] Once the State Council Office has passed the verification, the system will generate a user-specific QR code according to the QR code specifications of the Ministry of Transport. Users can use this QR code to scan their faces to enter parks, take buses and subways, and other activities. Example

[0083] The difference between Example 2 of the present invention and Example 1 is that this example introduces the smart public transportation talent management platform.

[0084] Figure 2A structural schematic diagram of the smart public transportation talent management platform of the present invention is given, which includes a multi-scale entropy analysis module, a stay time analysis module, an aggregation risk judgment module, a request processing enhancement module, a path consistency evaluation module, a diffusion trend judgment module and an identity qualification tracking module.

[0085] Multi-scale entropy analysis module: By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area.

[0086] Dwell time analysis module: By monitoring the dwell time data of users in the target area, analyzing the dispersion of dwell time, and evaluating the convergent distribution of dwell time of people in the target area.

[0087] Aggregation risk judgment module: Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area.

[0088] Request processing improvement module: When there is a potential risk of aggregation effect in the target area, it triggers the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests.

[0089] Path consistency evaluation module: By collecting the path direction data of users in the target area, the fuzzy clustering algorithm is used to analyze the direction consistency and evaluate the fluctuation of the path direction consistency in the target area.

[0090] Diffusion trend judgment module: Based on the fluctuation of path direction consistency in the target area, it is judged whether there is a trend of diffusion effect about to begin in the target area.

[0091] Identity qualification tracking module: When there is a trend of diffusion effect about to begin in the target area, the user's identity information and preferential qualifications are continuously and dynamically tracked.

[0092] 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.

[0093] 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.

[0094] 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.

[0095] 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.

[0096] 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.

[0097] 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.

[0098] 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.

[0099] 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.

[0100] 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.

[0101] 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 smart public transportation talent identification method, characterized in that: The steps include: By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area. By monitoring the length of stay data of users in the target area, analyzing the dispersion of the length of stay, and evaluating the convergence distribution of the length of stay of people in the target area; Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area; When there is a potential risk of aggregation effect in the target area, it triggers the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests; at the same time, by collecting the path direction data of users in the target area, using fuzzy clustering algorithm to analyze the direction consistency, and evaluate the fluctuation of path direction consistency in the target area; Based on the fluctuation of the consistency of the path direction in the target area, it is judged whether the target area has a tendency of the diffusion effect to begin. When the target area has a tendency of the diffusion effect to begin, the user's identity information and preferential qualifications are continuously and dynamically tracked.

2. The smart public transportation talent identification method according to claim 1 is characterized in that: By gridding the real-time crowd location data in the target area, the local information entropy value of each grid is calculated and the entropy value time series is generated to evaluate the multi-scale entropy fluctuation of the target area, specifically: The sensor network collects the location data of people in the target area in real time and records the real-time location coordinates of each user in the grid; Divide the target area into multiple grids according to a preset size and construct a spatial position model for each grid; According to the distribution of crowd positions in the grid, calculate the local information entropy value of each grid; Among them, the calculation formula of local information entropy value is: ;in Indicates Line The local information entropy value of the column grid, For the The number of different crowd positions observed within the grid, Indicates The probability of a position, and Represents the row and column position index of the grid, The numbers representing the different population locations observed within a particular grid; The local information entropy value of each grid is recorded in real time to generate the entropy value time series of each grid within the specified time range; The multi-scale analysis method is used to conduct multi-scale fluctuation assessment on the entropy value time series of each grid, and the multi-scale entropy fluctuation in the target area is quantitatively assessed.

3. The smart public transportation talent identification method according to claim 2 is characterized in that: The multi-scale analysis method is used to evaluate the multi-scale fluctuation of the entropy value time series of each grid, and the multi-scale entropy fluctuation in the target area is quantitatively evaluated, specifically: Set the time scale set, which is expressed as: ,in, is the set of time scales, Indicates The time scale of the hierarchy, Indicates the time scale level number in multiscale analysis; Calculate the multiscale entropy fluctuation index, which is expressed as: ;in, is the multiscale entropy fluctuation index, Indicates Line Column grid at time points The local information entropy value of is the index variable of the current step in the time scale, Indicates time, It refers to the The time range selected in the hierarchical time scale is used as the upper limit of each sliding step in the time series; Set the multiscale entropy fluctuation threshold; compare the multiscale entropy fluctuation index with the multiscale entropy fluctuation threshold: When the multiscale entropy fluctuation index is greater than the multiscale entropy fluctuation threshold, the risk of entropy analysis aggregation effect is judged to be high; when the multiscale entropy fluctuation index is less than or equal to the multiscale entropy fluctuation threshold, the risk of entropy analysis aggregation effect is judged to be normal.

4. The smart public transportation talent identification method according to claim 3 is characterized in that: By monitoring the length of stay data of users in the target area, analyzing the dispersion of the length of stay, and evaluating the convergence distribution of the length of stay of people in the target area, specifically: Collect the residence time data of each user in the target area in real time and pre-process the collected residence time data; Calculate the discrete degree of user stay time in the target area and use the variance formula to obtain the discrete distribution value of user stay time; Based on the discrete distribution value of user stay time, the concentration of user stay time is analyzed to evaluate the convergence distribution of the stay time of people in the target area: if the discrete distribution value of user stay time is less than the preset threshold , it is judged that the user's stay time is convergent; if the discrete distribution value of the user's stay time is greater than or equal to the preset threshold , it is judged that the user's stay time does not have convergence.

5. The smart public transportation talent identification method according to claim 4 is characterized in that: Based on the multi-scale entropy fluctuations of the target area and the convergent distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area, specifically: When the entropy analysis aggregation effect risk is high and the discrete distribution value of the user's stay time is less than the preset threshold When , it is determined that the target area has a potential risk of aggregation effect.

6. The smart public transportation talent identification method according to claim 5 is characterized in that: When there is a potential risk of agglomeration effect in the target area, requirements for improving the efficiency of processing identity verification and preferential qualification verification requests are triggered, specifically: When a potential risk of aggregation effect is detected in the target area, an early warning signal is immediately triggered; After receiving the early warning signal, resources are automatically allocated dynamically; Adjust the priority sorting for identity verification and preferential eligibility verification; divide verification requests into batches and adopt parallel processing.

7. The smart public transportation talent identification method according to claim 6 is characterized in that: By collecting the path direction data of users in the target area, the fuzzy clustering algorithm is used to analyze the direction consistency and evaluate the fluctuation of the path direction consistency in the target area. Specifically: Collect path direction data in real time, which includes the user's location information and path direction at different time points; The path direction data is input into the fuzzy clustering algorithm model, and the user groups are divided into high consistency groups and low consistency groups according to the consistency of the path direction; The proportion of the low consistency group to the total number of users is calculated, and the path direction consistency volatility is obtained as an indicator for evaluating the change of path direction consistency.

8. The smart public transportation talent identification method according to claim 7 is characterized in that: Based on the fluctuation of path direction consistency in the target area, it is judged whether the target area has a trend of diffusion effect about to begin, specifically: Compare the path direction consistency volatility with the preset threshold Compare and determine whether there is a trend of diffusion effect about to begin in the target area: If the path direction consistency volatility is greater than or equal to the preset threshold , it is determined that the target area has a tendency to start diffusion effect; If the path direction consistency volatility is less than the preset threshold , it is determined that the target area has no tendency of diffusion effect temporarily.

9. The smart public transportation talent identification method according to claim 8 is characterized in that: When the target area shows a trend of diffusion effect, the user's identity information and preferential qualifications are continuously and dynamically tracked, specifically: Monitor the user's latest location in real time; continuously and dynamically synchronize the user's identity information during the user's movement; update the user's preferential eligibility status in real time; when the user moves to an adjacent area, synchronize the identity information and preferential eligibility status to the adjacent area.

10. A smart public transportation talent management platform, used to implement the smart public transportation talent identification method according to any one of claims 1 to 9, characterized in that: It includes multi-scale entropy analysis module, stay time analysis module, aggregate risk judgment module, request processing improvement module, path consistency assessment module, diffusion trend judgment module and identity qualification tracking module; Multi-scale entropy analysis module: by gridding the real-time crowd location data in the target area, calculating the local information entropy value of each grid and generating an entropy value time series, the multi-scale entropy fluctuation of the target area is evaluated; Dwell time analysis module: by monitoring the dwell time data of users in the target area, analyzing the dispersion of dwell time, and evaluating the convergence distribution of dwell time of people in the target area; Aggregation risk judgment module: Based on the multi-scale entropy fluctuations of the target area and the convergence distribution of the length of stay of the people in the target area, it is judged whether there is a potential risk of aggregation effect in the target area; Request processing improvement module: When there is a potential risk of aggregation effect in the target area, it triggers the requirement to improve the efficiency of processing identity authentication and preferential qualification verification requests; Path consistency evaluation module: collects path direction data of users in the target area, uses fuzzy clustering algorithm to analyze direction consistency, and evaluates the fluctuation of path direction consistency in the target area; Diffusion trend judgment module: judges whether the target area is about to start a diffusion effect based on the fluctuation of the path direction consistency in the target area; Identity qualification tracking module: When there is a trend of diffusion effect about to begin in the target area, the user's identity information and preferential qualifications are continuously and dynamically tracked.

Images