Heatstroke community hosting demand analysis method based on school-enterprise linkage

Abstract

The invention discloses a summer community trusteeship demand analysis method based on school-enterprise linkage in the technical field of electronic data analysis management, which comprises the following steps: acquiring desensitization data of enterprise employees within a certain time period based on a human resource management system, encoding the desensitization data to obtain encoded data, and storing the encoded data in a database; and collecting trusteeship demand data corresponding to the coded data by adopting a personnel flow system, and performing fuzzy matching on the coded data and the trusteeship demand data by adopting a fuzzy matching method to obtain a matched data set. According to the method, the data string is formed by combining the desensitization data including the post function, the commuting route and the attendance duration, the hosting demand data and the thermodynamic diagram of the hosting place, and the communication path of the electronic map is subjected to dotted labeling, so that the density distribution diagram of the hosting demand can be quickly obtained; the summer hosting resources can be reasonably allocated, and the authenticity and accuracy of hosting demand analysis can be improved.

Description

Technical Field

[0001] This invention relates to the field of electronic data analysis and management technology, specifically a method for analyzing the needs of summer community care programs based on university-enterprise collaboration. Background Technology

[0002] Summer childcare is of great significance in alleviating the current difficulties faced by dual-income families in raising children. When schools provide venue resources, teachers share campus resources through online teaching, and communities participate in summer childcare by gathering college volunteers, and when companies track and survey childcare services and analyze childcare needs, they can effectively gain a deeper understanding of the fluctuations in employee turnover rates in the following year.

[0003] Existing community childcare demand analysis relies on manual surveys or simple statistical models. Because traditional regression models are difficult to handle the correlation data between employees' job functions, commuting routes and attendance hours and childcare needs, the authenticity and accuracy of existing childcare demand analysis are insufficient. It is difficult to effectively adapt to the needs of personnel with different job functions, commuting routes and attendance hours, and it cannot effectively reduce employee turnover. Summary of the Invention

[0004] The purpose of this invention is to provide a method for analyzing the demand for summer community care services based on school-enterprise collaboration. By using anonymized data including job functions, commuting routes and attendance duration, care service demand data, and heat maps of care service locations to form a data string, and marking the communication paths on the electronic map in a point-like manner, it is beneficial to quickly obtain a density distribution map of care service demand, to rationally allocate summer care service resources, and to improve the authenticity and accuracy of care service demand analysis, and to reduce the turnover rate of enterprise employees.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] Firstly, this invention provides a method for analyzing the demand for summer community care services based on university-enterprise collaboration, including:

[0007] Based on the human resources management system, anonymized data of enterprise employees within a certain period is obtained and encoded to obtain coded data. Then, the HR process system collects the corresponding managed services demand data. Fuzzy matching is used to perform fuzzy matching between the coded data and the managed services demand data to obtain a matching dataset. The anonymized data includes job functions, commuting routes, and attendance duration, while the managed services demand data includes the number of employees requiring managed services and their ages. Using anonymized data of enterprise employees helps to analyze the urgency of managed services demand while protecting employee privacy, and improves the authenticity and accuracy of managed services demand analysis.

[0008] Geographic heatmap technology is used to obtain heatmaps of after-school care schools and residential communities corresponding to the matching dataset. A point-to-point topology structure of the after-school care school and residential community heatmaps is obtained based on a geographic information system. A graph neural network is used to associate the matching dataset with the topology structure to obtain a related data string. By using geographic heatmap technology to obtain the heatmaps of after-school care schools and residential communities corresponding to the matching dataset, the distribution of schools and communities can be quickly determined, which is beneficial for integrating after-school care resources in schools and communities. Furthermore, by using a graph neural network to associate the matching dataset with the topology structure to obtain the related data string, it is beneficial for obtaining the actual after-school care needs of employees based on their actual commuting times and routes.

[0009] Based on electronic maps, the associated data is strung together and marked as dots on the commuting routes of the electronic maps using annotation methods, resulting in labeled data with marked points. Kernel density analysis is used to obtain a density distribution map of the labeled data. Kernel radius and bandwidth parameters are set, and the number of schools and communities covered by the density distribution map is found. Based on preset judgment rules, the urgency of after-school care needs is determined, and after-school care resources are allocated. By obtaining labeled data with marked points and combining it with the density distribution map of the labeled data, it is easier to quickly determine the correlation between the occupational attributes of employees with after-school care needs and their needs. The school and community resources are statistically analyzed to obtain a resource ledger. After-school care resources are allocated based on the resource ledger, thereby reducing the childcare pressure on employees during the summer.

[0010] As a further aspect of the present invention: the method of obtaining de-identified data of enterprise employees within a certain time period based on the human resource management system, and encoding the de-identified data to obtain encoded data, includes:

[0011] Set a specific time period, within the past six months;

[0012] Based on data on employee job functions, commuting routes, and attendance duration obtained from the human resources management system over the past six months, these data are sequentially coded using a first, second, and third classification code, resulting in first-coded data, second-coded data, and third-coded data. The coded data comprises these three categories. By constructing a linked data string from these three categories, along with the commuting route code and attendance duration code, a structured processing of employee occupational attribute characteristics is achieved. This ensures that each employee's occupational attribute characteristics can be fully represented through a single data entry, providing a standardized data foundation for spatial annotation and density analysis on electronic maps.

[0013] As a further aspect of the present invention: the point-to-point topology structure of the heat map of the managed school and the residential community obtained based on the geographic information system includes:

[0014] Set up a point-to-point topology template that includes the first node, the second node, and the connection path;

[0015] Input the heat map data of the host school into the first node of the point-to-point topology template, and input the heat map of the residential community into the second node;

[0016] Set a dataset splitting rule, and use a data splitting method based on the splitting rule to split the matching dataset into multiple matching data units;

[0017] By inputting matching data units corresponding to the first and second nodes along the connection path using a graph neural network, a related data string is formed. By inputting the heat map data of the boarding school into the first node of the point-to-point topology template, inputting the heat map of the residential community into the second node, and inputting matching data units corresponding to the first and second nodes along the connection path using a graph neural network, a related data string can be formed. The correlation between data can be intuitively matched through the annotation of the point-to-point topology map and its connection path.

[0018] As a further aspect of the present invention: the dataset splitting rule is to set each matching data unit to include coded data and the hosting requirement data corresponding to the coded data.

[0019] As a further aspect of the present invention: the step of determining the urgency of hosting needs and allocating hosting resources based on preset judgment rules includes:

[0020] Based on the density distribution map, kernel radius, and bandwidth parameters, the preset judgment rules are set;

[0021] Based on preset judgment rules, the urgency of the hosting needs is divided into low demand, medium demand, and high demand;

[0022] When the management demand corresponding to the commuting route is determined to be low demand, the nearest neighbor algorithm is used to merge the commuting route corresponding to low demand into the nearest medium demand commuting route.

[0023] When the demand for after-school care corresponding to the commuting route is determined to be medium demand, the school resources and community resources corresponding to medium demand are compared, and the corresponding school resources and / or community resources are matched based on the number of people seeking after-school care to obtain the after-school care resources for medium demand.

[0024] When the demand for after-school care is determined to be high along the commuting route, the school and community resources corresponding to the high demand will be designated as high-demand after-school care resources.

[0025] As a further aspect of the present invention: the preset judgment rule based on the density distribution map, kernel radius, and bandwidth parameters includes:

[0026] The number of markers on the same commuting route on the statistical density distribution map is determined by using a data collector to obtain resident population data under kernel radius and bandwidth parameters. When the number of markers does not exceed one ten-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be low demand.

[0027] When the number of marked points exceeds one ten-thousandth but is less than one thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be medium demand.

[0028] When the number of marked points exceeds one-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be high.

[0029] As a further aspect of the present invention: the step of statistically analyzing the school resources and community resources to obtain a resource ledger, and allocating managed resources based on the resource ledger, includes:

[0030] The resources are recorded by statistically analyzing the capacity of after-school care facilities, teacher allocation and available time periods of the schools, as well as the area of ​​after-school care venues, safety facilities and opening hours of the communities.

[0031] The managed resource data in the resource ledger is stored in a structured database, and the managed resources are dynamically allocated to the schools and communities with the highest matching degree based on the urgency of the managed needs using an optimized matching algorithm.

[0032] Obtain meteorological data for a certain period in the future, determine whether to trigger an adjustment operation of the resource availability coefficient based on the meteorological data, and adjust the resource availability coefficient after the adjustment operation is triggered. The certain period is several days.

[0033] As a further aspect of the present invention: the step of acquiring meteorological data for a certain future period, determining whether to trigger an adjustment operation of the resource availability coefficient based on the meteorological data, and adjusting the resource availability coefficient after triggering the adjustment operation, includes:

[0034] Meteorological data for the next few days is obtained from an online weather forecasting platform;

[0035] Compare meteorological data with preset extreme weather data;

[0036] When at least one meteorological data point matches the characteristics of the extreme weather data, an adjustment operation is triggered and the resource availability coefficient is increased.

[0037] As a further aspect of the present invention: the step of statistically analyzing the school resources and community resources to obtain a resource ledger, and allocating managed resources based on the resource ledger, further includes:

[0038] Build a correlation model between hosting needs and employee turnover rate to predict employee turnover rate.

[0039] As a further aspect of the present invention: the association model is:

[0040] λ = β0 + β1 × η + β2 × α + ε, where λ is the employee turnover rate, β0 is a constant term, η is the hosting demand satisfaction rate, β1 is the hosting demand satisfaction rate coefficient, α is the historical turnover rate, β2 is the historical turnover rate coefficient, and ε is the random error term. The constant term is used to represent the baseline level of the turnover rate, the hosting demand satisfaction rate coefficient is used to represent the coefficient of the hosting demand satisfaction rate, and β1 is used to measure the degree of change of the employee turnover rate λ when the hosting demand satisfaction rate changes by one unit.

[0041] Secondly, the present invention also provides a system applied to the summer community care demand analysis method based on school-enterprise collaboration as described in the above scheme. The system includes a data encoding module, a data matching module, a graph neural network module, a care demand judgment module, and a care resource allocation module. The data encoding module is configured to acquire anonymized data of enterprise employees within a certain time period based on a human resource management system, encode the anonymized data to obtain encoded data, and collect care demand data corresponding to the encoded data using a personnel process system. The input of the data matching module is electrically connected to the output of the data encoding module. The data matching module is configured to use fuzzy matching to perform fuzzy matching between the encoded data and the care demand data to obtain a matching dataset. The anonymized data includes job functions, commuting routes, and attendance duration, while the care demand data includes the number of caregivers and their ages. The input of the graph neural network module is electrically connected to the output of the data matching module. The graph neural network module is configured to use geographic heatmap technology. The following steps are performed: First, obtain the heatmaps of the childcare schools and residential communities corresponding to the matching dataset. Second, obtain the point-to-point topology of the childcare school and residential community heatmaps based on a geographic information system. Third, use a graph neural network to associate the matching dataset with the topology to obtain a linked data string. The input of the childcare demand judgment module is electrically connected to the output of the graph neural network module. The childcare demand judgment module is configured to use an annotation method based on an electronic map to mark the linked data string as points on the commuting route of the electronic map, obtaining labeled data with marked points. A kernel density analysis method is used to obtain the density distribution map of the labeled data. Kernel radius and bandwidth parameters are set, and the number of schools and communities covered by the density distribution map is found. The urgency of the childcare demand is judged based on preset judgment rules, and childcare resources are allocated. Finally, the input of the childcare resource allocation module is electrically connected to the output of the childcare demand judgment module. The childcare resource allocation module is configured to statistically analyze the school and community resources to obtain a resource ledger, and allocate childcare resources based on the resource ledger.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] 1. In this invention, by using desensitized data including job functions, commuting routes and attendance duration, childcare demand data, and heat maps of childcare locations to form a data string, the communication paths of the electronic map are marked in a point-like manner. This facilitates the rapid acquisition of a density distribution map of childcare demand, helps to rationally allocate summer childcare resources, and improves the authenticity and accuracy of childcare demand analysis. At the same time, in conjunction with an employee turnover rate prediction model, the necessity of summer childcare can be evaluated, which helps to reduce the turnover rate of employees already employed by the company.

[0044] 2. In this invention, by using anonymized data including job functions, commuting routes, and attendance duration, it is beneficial to analyze the urgency of outsourcing needs while protecting the privacy of enterprise employees. This also helps to improve the authenticity and accuracy of outsourcing demand analysis. By constructing a related data string from the first coded data, the second coded data, the third coded data, the commuting route code, and the attendance duration code, the structured processing of the occupational attribute characteristics of enterprise employees is achieved. This ensures that the occupational attribute characteristics of each enterprise employee can be fully represented through a single data entry, which is beneficial to laying a standardized data foundation for spatial annotation and density analysis on electronic maps. Attached Figure Description

[0045] Figure 1 This is a diagram illustrating the method steps of the present invention;

[0046] Figure 2 This is a system structure diagram of the present invention.

[0047] In the diagram: 1. Data encoding module; 2. Data matching module; 3. Graph neural network module; 4. Hosting requirement judgment module; 5. Hosting resource allocation module. Detailed Implementation

[0048] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0049] Example:

[0050] Please see Figure 1 In this embodiment of the invention, a method for analyzing the demand for summer community care based on school-enterprise collaboration includes the following steps:

[0051] S1: Obtain de-identified data of enterprise employees within a certain period of time based on the human resources management system, and encode the de-identified data to obtain coded data. Use the human resources process system to collect the hosting demand data corresponding to the coded data, and use the fuzzy matching method to perform fuzzy matching between the coded data and the hosting demand data to obtain a matching dataset. The de-identified data includes job functions, commuting routes and attendance duration, and the hosting demand data includes the number of people to be hosted and the age of the people to be hosted.

[0052] S2: Using geographic heatmap technology, we obtain the heatmaps of the after-school care schools and residential communities corresponding to the matching datasets. Based on the geographic information system, we obtain the point-to-point topology of the after-school care school heatmaps and residential community heatmaps. We then use graph neural networks to associate the matching datasets with the topology to obtain the associated data strings.

[0053] S3: Based on the electronic map, the annotation method is used to mark the associated data as points on the commuting route of the electronic map to obtain the labeled data with the marked points. The kernel density analysis method is used to obtain the density distribution map of the labeled data. The kernel radius and bandwidth parameters are set and the number of schools and communities covered by the density distribution map is found. The urgency of the after-school care demand is judged based on the preset judgment rules and after-school care resources are allocated.

[0054] S4: Compile statistics on school and community resources to obtain a resource ledger, and allocate managed resources based on the resource ledger.

[0055] In this embodiment, let the matching dataset of enterprise employees be... Each matching dataset E contains job functions. Commuting routes and attendance hours The encoded subset of data, specifically, uses one-hot encoding to represent job functions. Encoded as discrete vectors commuting routes Encoded as a path coordinate sequence Encode attendance duration as a scalar , It is the set of positive real numbers.

[0056] Preferably, step S1 includes: setting a certain time period as the past six months, obtaining the job functions, commuting routes and attendance duration data of the company's employees in the past six months based on the human resources management system, and encoding the job functions, commuting routes and attendance duration in sequence using the first classification code, the second classification code and the third classification code, respectively, to obtain the first code data, the second code data and the third code data, wherein the code data includes the first code data, the second code data and the third code data.

[0057] In this embodiment, step S1 uses fuzzy matching to perform fuzzy matching between the encoded data and the hosting requirement data. The fuzzy matching function used is:

[0058] ;in, For text similarity conditions, For text similarity functions, The spatial distance threshold, For de-identified data of company employees, For managed data needs, Given the spatial distance condition, P r Let P be the coordinates of the commuting route. d The coordinates of the childcare service point are the school and / or community.

[0059] Preferably, the spatial distance threshold is used to ensure that service locations are within the employee's commuting range.

[0060] Preferably, the point-to-point topology of the heat map of the managed school and the residential community obtained based on the geographic information system includes:

[0061] Set up a point-to-point topology template that includes the first node, the second node, and the connection path;

[0062] Input the heat map data of the host school into the first node of the point-to-point topology template, and input the heat map of the residential community into the second node;

[0063] Set a dataset splitting rule, and use a data splitting method based on the splitting rule to split the matching dataset into multiple matching data units;

[0064] Based on the graph neural network, matching data units corresponding to the first node and the second node are input into the connection path to form an associated data string.

[0065] In this embodiment, the point-to-point topology graph structure defined based on the graph neural network is as follows: Where V is a node, E is a connection path, and the connection path is an edge of the topological graph structure.

[0066] Preferably, the dataset splitting rule is to set each matching data unit to include coded data and the corresponding managed requirement data.

[0067] Preferably, the urgency of the hosting needs is determined based on preset judgment rules, and hosting resources are allocated accordingly, including:

[0068] Based on the density distribution map, kernel radius, and bandwidth parameters, preset judgment rules are set;

[0069] Based on preset judgment rules, the urgency of the hosting needs is divided into low demand, medium demand, and high demand;

[0070] When the management demand corresponding to the commuting route is determined to be low demand, the nearest neighbor algorithm is used to merge the commuting route corresponding to low demand into the nearest medium demand commuting route.

[0071] When the demand for after-school care corresponding to the commuting route is determined to be medium demand, the school resources and community resources corresponding to medium demand are compared, and the corresponding school resources and / or community resources are matched based on the number of people seeking after-school care to obtain the after-school care resources for medium demand.

[0072] When the demand for after-school care is determined to be high along the commuting route, the school and community resources corresponding to the high demand will be designated as high-demand after-school care resources.

[0073] Preferably, based on the density distribution map, kernel radius, and bandwidth parameters, preset judgment rules are set, including:

[0074] The number of markers on the same commuting route on the statistical density distribution map is determined by using a data collector to obtain resident population data under kernel radius and bandwidth parameters. When the number of markers does not exceed one ten-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be low demand.

[0075] When the number of marked points exceeds one ten-thousandth but is less than one thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be medium demand.

[0076] When the number of marked points exceeds one-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be high.

[0077] Preferably, school and community resources are statistically analyzed to obtain a resource ledger, and managed resources are allocated based on the resource ledger, including:

[0078] The data includes the capacity of after-school care facilities in schools, the number of teachers and available time slots, as well as the area of ​​after-school care venues in communities, safety facilities and opening hours, to obtain a resource ledger.

[0079] The system uses a structured database to store the managed resource data in the resource ledger, and uses an optimized matching algorithm to dynamically allocate managed resources to the schools and communities with the highest matching degree based on the urgency of the managed needs.

[0080] Obtain meteorological data for a certain period in the future, determine whether to trigger an adjustment operation of the resource availability coefficient based on the meteorological data, and adjust the resource availability coefficient after the adjustment operation is triggered. The certain period is several days.

[0081] Preferably, meteorological data for a certain future period is acquired, and based on the meteorological data, it is determined whether to trigger an adjustment operation of the resource availability coefficient. If the adjustment operation is triggered, the resource availability coefficient is adjusted, including:

[0082] Meteorological data for the next few days is obtained from an online weather forecasting platform;

[0083] Compare meteorological data with preset extreme weather data;

[0084] When at least one meteorological data point meets the characteristics of extreme weather data, an adjustment operation is triggered and the resource availability coefficient is increased.

[0085] Preferably, the process of compiling statistics on school and community resources to obtain a resource ledger, and allocating managed resources based on this ledger, also includes:

[0086] Build a correlation model between hosting needs and employee turnover rate to predict employee turnover rate.

[0087] The preferred correlation model is:

[0088] λ = β0 + β1 × η + β2 × α + ε, where λ is the employee turnover rate, β0 is a constant term, η is the hosting demand satisfaction rate, β1 is the hosting demand satisfaction rate coefficient, α is the historical turnover rate, β2 is the historical turnover rate coefficient, and ε is the random error term. The constant term is used to represent the baseline level of the turnover rate, the hosting demand satisfaction rate coefficient is used to represent the coefficient of the hosting demand satisfaction rate, and β1 is used to measure the degree of change of the employee turnover rate λ when the hosting demand satisfaction rate changes by one unit.

[0089] like Figure 2As shown, this invention also provides a system applied to the summer community care demand analysis method based on school-enterprise collaboration as described above. The system includes a data encoding module 1, a data matching module 2, a graph neural network module 3, a care demand judgment module 4, and a care resource allocation module 5. The data encoding module 1 is configured to acquire anonymized data of enterprise employees within a certain time period from a human resource management system, and encode the anonymized data to obtain encoded data. The system then uses a personnel process system to collect care demand data corresponding to the encoded data. The input of the data matching module 2 is connected to the output of the data encoding module 1. The data matching module 2 is configured to use fuzzy matching to perform fuzzy matching between the encoded data and the care demand data to obtain a matching dataset. The anonymized data includes job functions, commuting routes, and attendance duration, while the care demand data includes the number of caregivers and their ages. The input of the graph neural network module 3 is connected to the output of the data matching module 2. The graph neural network module 3 is configured to use geographic heatmap technology... The method acquires heat maps of the schools and residential communities corresponding to the matching dataset, obtains the point-to-point topology of the school and residential community heat maps based on the geographic information system, and uses a graph neural network to associate the matching dataset with the topology to obtain the associated data string. The input of the childcare demand judgment module 4 is connected to the output of the graph neural network module 3. The childcare demand judgment module 4 is configured to be based on an electronic map, and uses the annotation method to mark the associated data string as points on the commuting route of the electronic map to obtain the labeled data with the marked points. The density distribution map of the labeled data is obtained by using the kernel density analysis method. The kernel radius and bandwidth parameters are set and the number of schools and communities covered by the density distribution map is found. The urgency of the childcare demand is judged based on the preset judgment rules and the childcare resources are allocated. The input of the childcare resource allocation module 5 is connected to the output of the childcare demand judgment module 4. The childcare resource allocation module 5 is configured to count school resources and community resources to obtain a resource ledger, and allocate childcare resources based on the resource ledger.

[0090] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for analyzing the demand for summer community-based childcare services based on school-enterprise collaboration, characterized in that, include: Based on the human resources management system, the de-identified data of the company's employees within a certain period of time is obtained and encoded to obtain coded data. The human resources process system is used to collect the hosting demand data corresponding to the coded data. Fuzzy matching method is used to perform fuzzy matching between the coded data and the hosting demand data to obtain a matching dataset. The de-identified data includes job functions, commuting routes and attendance duration, and the hosting demand data includes the number of people to be hosted and the age of the people to be hosted. Geographic heatmap technology is used to obtain heatmaps of the after-school care schools and residential communities corresponding to the matching dataset. Based on the geographic information system, the point-to-point topology of the after-school care school heatmap and the residential community heatmap is obtained. A graph neural network is used to associate the matching dataset with the topology to obtain the associated data string. Based on the electronic map, the associated data is marked as points on the commuting route of the electronic map using the annotation method to obtain the marked data. The density distribution map of the marked data is obtained by the kernel density analysis method. The kernel radius and bandwidth parameters are set and the number of schools and communities covered by the density distribution map is found. The urgency of the after-school care demand is judged based on the preset judgment rules and after-school care resources are allocated. The school and community resources are statistically analyzed to obtain a resource ledger, and the managed resources are allocated based on the resource ledger.

2. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 1, characterized in that: The process involves acquiring anonymized data of company employees within a specific time period based on a human resources management system, and encoding the anonymized data to obtain coded data, including: Set a specific time period, within the past six months; Based on the data on the job functions, commuting routes and attendance duration of the company's employees over the past six months obtained from the human resources management system, the job functions, commuting routes and attendance duration are coded in sequence using the first category code, the second category code and the third category code, respectively, to obtain the first code data, the second code data and the third code data.

3. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 2, characterized in that: The point-to-point topology structure of the heat map of the managed school and the residential community obtained based on the geographic information system includes: Set up a point-to-point topology template that includes the first node, the second node, and the connection path; Input the heat map data of the host school into the first node of the point-to-point topology template, and input the heat map of the residential community into the second node; Set a dataset splitting rule, and use a data splitting method based on the splitting rule to split the matching dataset into multiple matching data units; Based on the graph neural network, matching data units corresponding to the first node and the second node are input into the connection path to form an associated data string.

4. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 3, characterized in that: The dataset splitting rule stipulates that each matching data unit includes coded data and the corresponding hosting requirement data.

5. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 4, characterized in that: The process of determining the urgency of hosting needs and allocating hosting resources based on preset judgment rules includes: Based on the density distribution map, kernel radius, and bandwidth parameters, the preset judgment rules are set; Based on preset judgment rules, the urgency of the hosting needs is divided into low demand, medium demand, and high demand; When the management demand corresponding to the commuting route is determined to be low demand, the nearest neighbor algorithm is used to merge the commuting route corresponding to low demand into the nearest medium demand commuting route. When the demand for after-school care corresponding to the commuting route is determined to be medium demand, the school resources and community resources corresponding to medium demand are compared, and the corresponding school resources and / or community resources are matched based on the number of people seeking after-school care to obtain the after-school care resources for medium demand. When the demand for after-school care is determined to be high along the commuting route, the school and community resources corresponding to the high demand will be designated as high-demand after-school care resources.

6. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 5, characterized in that: The preset judgment rules, based on the density distribution map, kernel radius, and bandwidth parameters, include: The number of markers on the same commuting route on the statistical density distribution map is determined by using a data collector to obtain resident population data under kernel radius and bandwidth parameters. When the number of markers does not exceed one ten-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be low demand. When the number of marked points exceeds one ten-thousandth but is less than one thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be medium demand. When the number of marked points exceeds one-thousandth of the resident population data, the hosting demand corresponding to the commuting route is judged to be high.

7. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 1, characterized in that: The statistics on school and community resources are compiled to obtain a resource ledger. Based on this resource ledger, managed resources are allocated, including: The resources are recorded by statistically analyzing the capacity of after-school care facilities, teacher allocation and available time periods of the schools, as well as the area of ​​after-school care venues, safety facilities and opening hours of the communities. The managed resource data in the resource ledger is stored in a structured database, and the managed resources are dynamically allocated to the schools and communities with the highest matching degree based on the urgency of the managed needs using an optimized matching algorithm. Obtain meteorological data for a certain period in the future, determine whether to trigger an adjustment operation of the resource availability coefficient based on the meteorological data, and adjust the resource availability coefficient after the adjustment operation is triggered. The certain period is several days.

8. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 1, characterized in that: The process of acquiring meteorological data for a certain future period, determining whether to trigger an adjustment operation for the resource availability coefficient based on the meteorological data, and adjusting the resource availability coefficient after the adjustment operation is triggered includes: Meteorological data for the next few days is obtained from an online weather forecasting platform; Compare meteorological data with preset extreme weather data; When at least one meteorological data point matches the characteristics of the extreme weather data, an adjustment operation is triggered and the resource availability coefficient is increased.

9. The method for analyzing the demand for summer community care based on school-enterprise collaboration as described in claim 1, characterized in that: The process of compiling statistics on school and community resources to obtain a resource ledger, and allocating managed resources based on this ledger, also includes: Build a correlation model between hosting needs and employee turnover rate to predict employee turnover rate; The association model is as follows: λ = β0 + β1 × η + β2 × α + ε, where λ is the employee turnover rate, β0 is a constant term, η is the hosting demand satisfaction rate, β1 is the hosting demand satisfaction rate coefficient, α is the historical turnover rate, β2 is the historical turnover rate coefficient, and ε is the random error term.

10. A system, characterized in that: The system is applied to the summer community care demand analysis method based on school-enterprise collaboration as described in any one of claims 1-9, and the system includes: The data encoding module (1) is configured to obtain desensitized data of enterprise employees within a certain period of time based on the human resources management system, encode the desensitized data to obtain encoded data, and use the human resources process system to collect the hosting demand data corresponding to the encoded data. The data matching module (2) is electrically connected to the output of the data encoding module (1). The data matching module (2) is configured to use fuzzy matching to perform fuzzy matching between the encoded data and the hosting demand data to obtain a matching dataset. The anonymized data includes job functions, commuting routes and attendance duration, and the hosting demand data includes the number of people being hosted and the age of the people being hosted. The graph neural network module (3) is electrically connected to the output of the data matching module (2). The graph neural network module (3) is configured to use geographic heat map technology to obtain the heat map of the school and the residential community corresponding to the matching dataset, obtain the point-to-point topology of the school and residential community heat map based on the geographic information system, and use graph neural network to associate the matching dataset with the topology to obtain the associated data string. The hostage demand judgment module (4) is input to the electrical connection graph neural network module (3). The hostage demand judgment module (4) is configured to be based on an electronic map, and the annotation method is used to mark the associated data string as points on the commuting route of the electronic map to obtain the marked data with marked points. The kernel density analysis method is used to obtain the density distribution map of the marked data, set the kernel radius and bandwidth parameters and find the number of schools and communities covered by the density distribution map, and judge the urgency of the hostage demand and allocate hostage resources based on the preset judgment rules. The managed resource allocation module (5) is connected to the output of the managed needs judgment module (4). The managed resource allocation module (5) is configured to count the school resources and community resources, obtain a resource ledger, and allocate managed resources based on the resource ledger.

Images