Neural network-based radar chart generation system for age-friendly community health management

Abstract

The present application relates to the technical field of community data management, and in particular to a radar chart generation system for health management of an aging community based on a neural network. A data acquisition module collects facility equipment data, population structure data, and facility usage data during the activities of the elderly; a community clustering analysis module clusters communities according to population structure differences to improve the relevance of subsequent analysis processes; an activity behavior analysis module compares facility equipment data and facility usage data within the cluster to determine activity bias (commonality / individuality), and quantifies the individual activity influence indicators corresponding to each elderly person for communities with individual activity bias to reflect the association between individual preferences and facilities; and a radar chart adjustment module dynamically optimizes the radar chart based on the neural network and individual indicators to intuitively present the real activity characteristics of the elderly. The system can improve the accuracy and efficiency of community health management decisions and provide quantitative basis for optimization of aging services.

Description

Technical Field

[0001] This invention relates to the field of community data management technology, specifically to a radar chart generation system for age-friendly community health management based on neural networks. Background Technology

[0002] As physiological functions decline, older adults are more susceptible to various chronic diseases and sudden health problems, making their need for health management more urgent. With the deepening of the aging population and the increasing emphasis on health management, age-friendly community health management is showing a rapid growth trend. Neural networks, as an important branch of artificial intelligence, possess powerful capabilities in data processing, analysis, and prediction. They can perform deep learning on elderly health data, uncovering potential patterns within the data and providing technical support for generating comprehensive and accurate health management radar charts.

[0003] Existing technologies typically assess each age-friendly community's radar map based on static indicators such as facilities, location, seniors' health, and environment. However, this approach often results in poor integration of facilities and age-friendly services, failing to dynamically capture the correlation between seniors' personalized activity preferences and facility usage. Consequently, the generated community health management radar map may be inaccurate and could even negatively impact the development of age-friendly communities. Summary of the Invention

[0004] To address the aforementioned technical problems (specific problems need to be described), the purpose of this invention is to provide a radar chart generation system for age-friendly community health management based on neural networks. The specific technical solution adopted is as follows:

[0005] This invention proposes a radar chart generation system for age-friendly community health management based on neural networks, the system comprising:

[0006] The data acquisition module is used to acquire data on the facilities and population structure of each age-friendly community, and to regularly collect data on the use of facilities by each senior citizen during activities.

[0007] The community clustering analysis module is used to analyze the differences in population structure data among different age-friendly communities, and to perform cluster analysis on all age-friendly communities to obtain all clusters;

[0008] The activity behavior analysis module is used to compare the distribution of facility equipment data of age-friendly communities and the distribution characteristics of facility usage data of the elderly during activities in each cluster, so as to determine the activity bias of the elderly in each age-friendly community in each cluster; for age-friendly communities with an activity bias of personalized activities, the module determines the personalized activity impact index for each elderly person based on the similarity and change characteristics of facility usage data of the elderly during activities.

[0009] The radar chart adjustment module is used to adjust the radar chart of each age-friendly community with an activity bias towards individual activities, based on neural networks and individual activity impact indicators.

[0010] Furthermore, the method for obtaining the clusters includes:

[0011] The population structure data includes the number of elderly people, the number of elderly men, and the average age of the elderly, and the number of elderly people and the number of elderly men in the population structure data are combined into a population vector.

[0012] For any two age-friendly communities, calculate the absolute value of the difference between the average age of the elderly in the two age-friendly communities as the first difference factor, and calculate the Euclidean distance between the population vectors of the two age-friendly communities as the second difference factor.

[0013] The normalized value of the product of the first difference factor and the second difference factor is used as the difference feature value between the two age-friendly communities.

[0014] The K-means clustering algorithm and a preset K value are used to perform cluster analysis on all age-friendly communities to obtain all clusters. The distance metric of the clusters is the difference feature value between age-friendly communities.

[0015] Furthermore, determining the activity preferences of elderly people in each age-friendly community within each cluster includes:

[0016] Both facility configuration data and facility usage data are in the form of a collection of facility names;

[0017] Within each cluster, the similarity characteristics among the facility configuration data of all age-friendly communities are analyzed, thereby dividing the facilities in age-friendly communities in each cluster into two sets of facility configurations: common facility configuration sets and individual facility configuration sets.

[0018] Within each cluster, the union of all facility usage data for all elderly people in each age-friendly community is taken as the facility usage set for that age-friendly community.

[0019] The proportion of the number of facilities in the intersection of the facility usage set corresponding to each age-friendly community and the facility equipment set of each type is used as the first activity bias factor of each age-friendly community under each type of facility equipment set.

[0020] In the data on the use of facilities by all elderly people in each age-friendly community, the proportion of the number of times a facility appears in each set of facilities to the total number and value of all elderly people's activities is used as the second activity bias factor for each age-friendly community under each set of facilities.

[0021] In each cluster, the normalized value of the product of the first activity bias factor and the second activity bias factor of each age-friendly community under each set of facilities is used as the activity bias index of each age-friendly community under each set of facilities. The activity bias index under the common set of facilities is the common activity bias index, and the activity bias index under the individual set of facilities is the individual activity bias index.

[0022] When the individual activity preference index of a certain age-friendly community is greater than or equal to the common activity preference index, the activity preference of the elderly in that age-friendly community is considered to be individual activity.

[0023] Furthermore, the methods for obtaining the common facility set and the individual facility set include:

[0024] Within each cluster, the data on the facilities of all age-friendly communities are intersected. The facilities in the intersection are taken as the common facilities set, and the facilities outside the intersection are taken as the individual facilities set.

[0025] Furthermore, the method for obtaining the personality activity impact indicators includes:

[0026] In any age-friendly community where activities are geared towards individualized activities, the facility usage data of each senior citizen during activities is arranged in the order of activity statistics to obtain a sorted sequence;

[0027] In the sorting sequence, based on the similarity between adjacent facility usage data, the first sexual activity influencing factor for each elderly person in the age-friendly community is determined;

[0028] In the sorting sequence, based on the variation characteristics between facility usage data, a second sexual activity influencing factor is determined for each elderly person in the age-friendly community;

[0029] The normalized value of the product of the first and second personal activity influencing factors for each elderly person in the age-friendly community is used as the personal activity influencing index for each elderly person in the age-friendly community.

[0030] Furthermore, the method for obtaining the first personality activity influencing factor includes:

[0031] In the sorting sequence, the Jaccard correlation coefficient between the facility usage data under each activity statistics and the facility usage data under the last activity statistics is calculated as the facility usage consistency factor corresponding to the number of activity statistics. The facility usage consistency factor corresponding to the number of activity statistics is weighted and fused using the ordinal value of the number of activity statistics in the sorting sequence, thereby obtaining the first age-friendly activity influence factor for each elderly person in the age-friendly community.

[0032] Furthermore, the method for obtaining the second personality activity influencing factor includes:

[0033] In the ranking sequence corresponding to the age-friendly community, the number of facilities in the intersection of each facility usage data and the individual facility configuration set of the cluster to which the age-friendly community belongs is used as a quantity factor. The first difference sequence of the quantity factors of all facility usage data in the ranking sequence is calculated, and the mean of all values ​​in the first difference sequence is normalized and used as the second sexual activity influence factor for each elderly person in the age-friendly community.

[0034] Furthermore, the radar chart assessment dimensions for each age-friendly community include geographical location score, senior health score, environmental score, service level score, hardware implementation score, and integrated medical and elderly care level score.

[0035] Furthermore, the radar chart of age-friendly communities, which adjusts each activity bias towards individual activities based on neural networks and individual activity impact indicators, includes:

[0036] For any age-friendly community with an activity bias towards personalized activities, the personalized activity impact index corresponding to each elderly person in the age-friendly community is used as a weight, and is input into a pre-trained neural network along with the facility usage data of each elderly person to obtain the updated medical and elderly care integration level score of the age-friendly community.

[0037] Based on the visualization engine, the geographical location score, health score of the elderly, environmental score, service level score, hardware implementation score, and the score of the level of upgrading of medical and elderly care integration of the age-friendly community are displayed on the radar chart, thereby obtaining the adjustment radar chart of the age-friendly community.

[0038] Furthermore, the neural network is a GNN.

[0039] The present invention has the following beneficial effects:

[0040] The data acquisition module obtains data on facility equipment and population structure for each age-friendly community, as well as regularly compiles data on facility usage by each senior citizen during activities. This helps address the issue of "poor integration of facilities and age-friendly services," making the analysis more aligned with the actual needs of the elderly. The community clustering analysis module analyzes and clusters the differences in population structure data among different age-friendly communities, grouping communities with similar population structure characteristics into the same cluster. This avoids interference from different types of communities (such as older and younger communities), improving the targeting and accuracy of subsequent activity behavior analysis. Furthermore, the activity behavior analysis module compares the distribution characteristics of facility equipment and facility usage data within each cluster to accurately determine the activity preferences of the elderly in the community. For communities dominated by personalized activities, the module analyzes the similarity and variation characteristics of facility usage data (such as a senior citizen's preference for a specific facility) to quantify the impact indicators of personalized activities. This reflects the correlation between each senior citizen's personalized activity preferences and facility usage, providing a quantitative basis for precise intervention. Finally, in the radar chart adjustment module, based on neural networks and individual activity impact indicators, the display of the radar chart is dynamically adjusted to make the visualization results more intuitively reflect the real activity characteristics of the elderly and improve the accuracy and efficiency of community health management decisions. Attached Figure Description

[0041] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a system block diagram of a radar chart generation system for age-friendly community health management based on neural networks, provided in one embodiment of the present invention.

[0043] Figure 2 This is a flowchart of a method for obtaining indicators of the impact of individual activities according to an embodiment of the present invention;

[0044] Figure 3 This is a schematic diagram of the system structure of an age-friendly community health management radar system based on a neural network, provided in one embodiment of the present invention. Detailed Implementation

[0045] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a neural network-based radar chart generation system for age-friendly community health management proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0047] The following description, in conjunction with the accompanying drawings, details a specific solution for a neural network-based radar chart generation system for age-friendly community health management provided by this invention.

[0048] Please see Figure 1 The diagram illustrates a system block diagram of a radar chart generation system for age-friendly community health management based on neural networks, according to an embodiment of the present invention. The system includes: a data acquisition module 101, a community clustering analysis module 102, an activity behavior analysis module 103, and a radar chart adjustment module 104.

[0049] The data acquisition module 101 is used to acquire data on the facilities and population structure of each age-friendly community, and to regularly collect data on the use of facilities by each elderly person during activities.

[0050] As a primary living environment for the elderly, the efficiency of health management in age-friendly communities directly impacts the quality of life for this population. An age-friendly community health management radar chart can integrate multi-source data, including wearable devices, community health checkups, environmental monitoring, and questionnaires, to construct a comprehensive health assessment model that visually displays the health status of the elderly in age-friendly communities across multiple dimensions.

[0051] In this embodiment of the invention, in order to dynamically capture the correlation between the personalized activity preferences of the elderly and the use of facilities in age-friendly communities, the data acquisition module acquires multi-source data for each age-friendly community. This multi-source data includes: facility equipment data, which is a set of facility names, such as fitness equipment (horizontal bar, Tai Chi poles, etc.), lounge chairs, accessible pathways, medical stations, etc., which can be extracted from the facility management ledger of the community management department; population structure data, including the number of elderly people, the number of male elderly people, and the average age of the elderly, which can be extracted from the community's household registration system; and periodically collected data on the facility usage of each elderly person during activities, also in the form of a set of facility names. For example, if an elderly person used a horizontal bar and rested on a lounge chair during an activity, the facility usage data for that statistical period would be {horizontal bar, lounge chair}, which can be collected through periodic questionnaires distributed to the elderly.

[0052] In the embodiments of this invention, the collection and acquisition of various information data are all authorized by the relevant users, and the process does not violate relevant laws and regulations, nor does it violate public order and good morals.

[0053] The community cluster analysis module 102 is used to analyze the differences in population structure data among different age-friendly communities, and to perform cluster analysis on all age-friendly communities to obtain all clusters.

[0054] Demographic data is a core factor influencing the activity preferences and health needs of older adults. It directly relates to differences in activity levels and facility requirements. For example, older adults in more advanced communities may have mobility issues and rely more on accessible facilities, while older adults in younger communities are more active and prefer outdoor activities. Furthermore, demographic data is a "basic attribute" of age-friendly communities. Compared to facility data (which can be modified) and environmental data (which can be adjusted), it is more stable and easier to quantify, making it suitable as an initial basis for classification. Separating communities with significant demographic differences can avoid biases in subsequent activity behavior analysis caused by different community types (for example, if mixed analysis is used, it may misjudge "the activity preference of older adults in a certain community is group activities" when it is actually due to different community types).

[0055] Therefore, in this module, we analyze the differences in population structure data among different age-friendly communities, perform cluster analysis on all age-friendly communities, and obtain all clusters.

[0056] Preferably, in one embodiment of the present invention, the method for obtaining clusters includes:

[0057] Based on the data acquisition module, we know that the population structure data includes the number of elderly people, the number of male elderly people, and the average age of the elderly. Since not only the age of the elderly affects their activity preferences, gender information may also affect the type of activity. Therefore, the number of elderly people and the number of male elderly people in the population structure data are combined to form a population vector. For example, if the number of elderly people in a certain age-friendly community is 100, and the number of male elderly people is 48, then the population vector is [100, 48]. The population vector can represent the gender distribution characteristics of the elderly in each age-friendly community.

[0058] Then, for any two age-friendly communities, the absolute value of the difference between the average ages of the elderly in these two age-friendly communities is calculated as the first difference factor. The smaller the first difference factor, the more similar the age distribution characteristics of the elderly in the two age-friendly communities are, and the greater the probability that they will be classified into the same category in the subsequent clustering process. The Euclidean distance between the population vectors of the two age-friendly communities is calculated as the second difference factor. The smaller the second difference factor, the more similar the gender distribution characteristics of the elderly in the two age-friendly communities are, and the greater the probability that they will be classified into the same category in the subsequent process.

[0059] Therefore, the normalized value of the product of the first and second difference factors is used as the difference characteristic value between the two age-friendly communities. This difference characteristic value comprehensively considers multiple characteristics such as age, size, and gender. Based on the aforementioned analysis, the smaller the difference characteristic value, the greater the similarity between the two age-friendly communities in terms of overall characteristics, and the greater the likelihood that they will be grouped into the same cluster in subsequent processes. Normalization is a technique well-known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0060] Finally, the K-means clustering algorithm and a preset K value are used to perform cluster analysis on all age-friendly communities to obtain all clusters. The distance metric of the clusters is the difference feature value between age-friendly communities.

[0061] At this point, cluster analysis can be performed on all age-friendly communities to obtain all clusters. The activity levels, health needs, and facility usage habits of the elderly in each cluster may be more similar (e.g., in communities with an average age of 80, the elderly pay more attention to accessible pathways and medical stations; in communities with an average age of 65, they pay more attention to fitness equipment and leisure squares). Therefore, analyzing their activity preferences and the correlation of facility usage based on this will be more in line with reality.

[0062] It should be noted that in this embodiment of the invention, the optimal K value obtained by the elbow method is used as the preset K value, and the elbow method and the K-means clustering algorithm are well-known technologies, and the specific process will not be described in detail here; in other embodiments of the invention, the preset K value can also be set according to the implementation scenario, such as setting it to 4, and is not limited here.

[0063] The activity behavior analysis module 103 is used to compare the distribution of facility equipment data of age-friendly communities and the distribution characteristics of facility usage data of the elderly during activities in each cluster, so as to determine the activity bias of the elderly in each age-friendly community in each cluster; for age-friendly communities with an activity bias of personalized activities, the module determines the personalized activity impact index corresponding to each elderly person based on the similarity and change characteristics of facility usage data of the elderly during activities.

[0064] Within each cluster (a group of age-friendly communities with similar population structures), the activity preferences (individual activities / common activities) of the elderly in each age-friendly community within the cluster are determined by comparing the distribution characteristics of "facility equipment data" and "facility usage data." Given that in age-friendly communities biased towards common activities, the individual activities of the elderly typically do not significantly affect the radar chart generation results, this embodiment of the invention primarily focuses on analyzing age-friendly communities biased towards individual activities. By analyzing the similarity and variation characteristics of the elderly's facility usage data (such as a single elderly person's preference for a specific facility), an "influence index for individual activities" is determined. This index provides a quantitative basis for subsequent optimization of facility configuration and the development of personalized health plans.

[0065] The data on facilities reflects what an age-friendly community "has," while the data on facility usage reflects what the elderly "use." Therefore, firstly, by comparing the facilities and usage behaviors of the elderly in each cluster, we can reveal the gap between "whether the facilities exist" and "whether the facilities are used effectively" (e.g., a community is equipped with Tai Chi poles, but the elderly prefer to use horizontal bars), thereby determining the activity preferences of each elderly person.

[0066] Preferably, in one embodiment of the present invention, determining the activity bias of elderly people in each age-friendly community within each cluster includes:

[0067] Both facility configuration data and facility usage data are sets of facility names. Within each cluster, the intersection of facility configuration data from all age-friendly communities is taken. Facilities within the intersection are designated as the common facilities set, while facilities outside the intersection are designated as the individual facilities set. This allows us to categorize facilities in age-friendly communities within each cluster into two sets: the common facilities set and the individual facilities set. Facilities in the common facilities set can be considered as uniformly configured to meet the basic needs of the elderly in all age-friendly communities within the cluster, reflecting the general needs of this type of community based on its population structure. Facilities in the individual facilities set, on the other hand, are considered as additionally configured to meet the differentiated needs of some age-friendly communities, reflecting the specific needs of certain age-friendly communities.

[0068] Within each cluster, the union of all facility usage data for all elderly people in each age-friendly community is taken as the facility usage set for that age-friendly community. The facility usage set for each age-friendly community comprehensively reflects the scope of facility usage by all elderly people in that age-friendly community, and the union operation ensures that all facilities used in that age-friendly community can be recorded.

[0069] Then, within each cluster, the facility usage set of each age-friendly community is compared with both the common facility allocation set and the individual facility allocation set to quantify the facility usage bias of each age-friendly community: the proportion of facilities in the intersection of the facility usage set corresponding to each age-friendly community and each type of facility allocation set (the common facility allocation set / individual facility allocation set corresponding to the cluster) is used as the first activity bias factor for each age-friendly community under each type of facility allocation set. At this time, the common facility allocation set corresponds to a first activity bias factor, which represents the degree to which the facility usage of each age-friendly community in each cluster is biased towards common facilities, and the larger the value, the greater the degree of bias towards common activities. Similarly, the individual facility allocation set also corresponds to a first activity bias factor, which represents the degree to which the facility usage of each age-friendly community in each cluster is biased towards individual facilities, and the larger the value, the greater the degree of bias towards individual activities.

[0070] Next, within each cluster, the frequency of occurrence of facilities in each type of facility set is calculated to quantify the bias in facility usage frequency for each age-friendly community. In the data on facility usage for all elderly residents in each age-friendly community, the proportion of the frequency of occurrence of facilities in each type of facility set (the common facility set / individual facility set corresponding to the cluster) in the total number of activity statistics for all elderly residents is used as the second activity bias factor for each age-friendly community under each type of facility set. Here, the common facility set corresponds to a second activity bias factor, which represents the frequency proportion of common facilities in the facility usage of each age-friendly community within each cluster. A larger value indicates a greater bias towards common activities. Similarly, the individual facility set also corresponds to a second activity bias factor, which represents the frequency proportion of individual facilities in the facility usage of each age-friendly community within each cluster. A larger value indicates a greater bias towards individual activities.

[0071] Within each cluster, the normalized value of the product of the first activity bias factor and the second activity bias factor for each age-friendly community under each set of facilities is used as the activity bias index for each age-friendly community under each set of facilities. Specifically, the activity bias index under common facilities is the common activity bias index, and the activity bias index under individual facilities is the individual activity bias index. Based on the foregoing analysis, a larger common activity bias index indicates that the activities of the elderly in the age-friendly community are more biased towards common activities; conversely, a larger individual activity bias index indicates that the activities of the elderly in the age-friendly community are more biased towards individual activities. Therefore, when the individual activity bias index of an age-friendly community is greater than or equal to the common activity bias index, the activities of the elderly in that age-friendly community are considered to be biased towards individual activities. Normalization is a technique well-known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0072] When the activities of the elderly in an age-friendly community tend to be individualized, in order to promote the shift of community health management from "facility-oriented" to "demand-oriented", the potential impact of each elderly person's activities on the community can be determined based on the similarity and changing characteristics of facility usage data during activities in the age-friendly community where activities tend to be individualized. This impact can be recorded as the individualized activity impact index.

[0073] Preferably, in one embodiment of the present invention, the method for obtaining the personality activity influence index includes:

[0074] Please see Figure 2 The diagram illustrates a method flowchart for obtaining a personality activity impact index according to an embodiment of the present invention. The method includes the following steps:

[0075] Step S301: In any age-friendly community where activities are geared towards individualized activities, arrange the facility usage data of each senior citizen during activities according to the order of activity statistics to obtain a sorted sequence.

[0076] Step S302: In the sorted sequence, based on the similarity between adjacent facility usage data, determine the first sexual activity influencing factor for each elderly person in the age-friendly community.

[0077] The core of personalized activities is whether older adults develop stable preferences for the use of differentiated facilities (personalized facilities). Therefore, in the ranking sequence, the facility usage data from the last activity is used as a benchmark. The Jaccard correlation coefficient between the facility usage data from each activity and the last activity is calculated as the facility usage consistency factor corresponding to each activity count. The larger the facility usage consistency factor, the higher the similarity between the set of facilities used in each activity and the set used in the last activity. Since the later the count, the closer the time is to the current moment, and therefore the higher the reference value, the facility usage consistency factor corresponding to the activity count is weighted and fused using the ordinal value of the activity count in the ranking sequence. That is, the ordinal value of the activity count in the ranking sequence is multiplied by the facility usage consistency factor corresponding to the activity count, and then the products of all activity counts are added together to obtain the first personalized activity influence factor for each older adult in the age-friendly community. The larger the first personalized activity influence factor, the less the older adult's use of facilities in the community changes in the age-friendly community that is already biased towards personalized activities. Therefore, the preference for personalized activities is more significant, and the degree of personalized activity influence is also greater.

[0078] It should be noted that the method for obtaining the Jaccard correlation coefficient is a well-known technique, and the specific process will not be elaborated here.

[0079] Step S303: In the sorted sequence, based on the variation characteristics between facility usage data, determine the second sexual activity influencing factor for each elderly person in the age-friendly community.

[0080] In the ranking sequence corresponding to the age-friendly community, the number of facilities in the intersection of the data on the usage of each facility and the set of individual facilities in the cluster to which the age-friendly community belongs is used as a quantitative factor. The quantitative factor can quantify the quantitative characteristics of the individual facilities used by the elderly in each activity, so as to analyze the changes in this characteristic in the subsequent process, and can also be used to measure the degree of influence of individual activities.

[0081] The first-order difference sequence of the quantity factor for all facility usage data in the sorted sequence is calculated. This first-order difference sequence characterizes the change in the quantity factor; a positive value indicates an increase in the use of personalized facilities, while a negative value indicates a decrease. The mean of all values ​​in the first-order difference sequence is normalized and used as the second sexual activity influence factor for each elderly person in the age-friendly community. A larger second sexual activity influence factor indicates a greater frequency and magnitude of increases in the use of personalized facilities. This suggests that in age-friendly communities already biased towards personalized activities, the elderly's preference for personalized activities is more significant. Since the values ​​in the first-order difference sequence can be both positive and negative, the normalization method used here can be... function.

[0082] Step S304: Combine the first and second personality activity influencing factors for each elderly person in the age-friendly community that is biased towards individual activities to obtain the personality activity influencing index for each elderly person in the age-friendly community.

[0083] Based on the aforementioned steps, it is known that both the first and second personal activity influencing factors for each elderly person are positively correlated with the degree of influence of their personal activities. Therefore, the normalized value of the product of the first and second personal activity influencing factors for each elderly person in the age-friendly community is used as the personal activity influencing index for each elderly person in the age-friendly community. The larger the personal activity influencing index, the more significant the elderly person's personal activities are in the age-friendly community, and thus the greater the impact on the subsequent generation of the radar chart for the age-friendly community. Normalization is a technique well-known to those skilled in the art, and the choice of normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0084] The radar chart adjustment module 104 is used to adjust the radar chart of each age-friendly community with an activity bias towards individual activities based on neural networks and individual activity impact indicators.

[0085] Given that radar charts can efficiently integrate multi-dimensional data and intuitively reveal pattern characteristics, radar charts can be constructed for each age-friendly community to reflect the scoring status of each age-friendly community across multiple dimensions.

[0086] In this embodiment of the invention, the radar chart assessment dimensions for each age-friendly community include a geographical location score (measuring the geographical proximity of the community to external resources such as top-tier hospitals and parks), an elderly health score (assessing the overall health of the elderly in the community, which can be quantified through elderly physical examination data, etc.), an environmental score (assessing the quality of the community's physical environment, such as greening rate and noise control), a service level score (measuring the quality of non-hardware services provided by the community, such as daily care and cultural activities), a hardware implementation score (assessing the equipment and maintenance status of indoor facilities in the community, etc.), and a medical and elderly care integration score (used to measure the degree of integration of medical resources and elderly care services in the community). The scores of the above multiple dimensions can be converted into standard scores (e.g., 1-10 points) according to the calculation rules set by the user. Among them, the score of the level of medical and elderly care integration is a service-oriented dimension that can be dynamically optimized compared with the data of other dimensions. Furthermore, based on the aforementioned module, the personalized activity impact index corresponding to each elderly person in the age-friendly community with an activity bias towards personalized activities can be obtained. Therefore, in this module, the radar chart of the age-friendly community with an activity bias towards personalized activities can be adjusted based on this index, specifically adjusting the score of the level of medical and elderly care integration. This will enable the final adjusted radar chart to more accurately represent the real activity characteristics of the elderly in the age-friendly community and to represent a more realistic situation of medical and elderly care integration.

[0087] Preferably, in one embodiment of the present invention, adjusting the radar chart of age-friendly communities with each activity biased towards individual activities based on neural networks and individual activity impact indicators includes:

[0088] Data on facility usage among the elderly can predict or assess changes in their healthcare needs. Personalized activity impact indicators reflect the elderly’s preference for or intensity of using differentiated facilities. Therefore, for any age-friendly community with a preference for personalized activities, the personalized activity impact indicator corresponding to each elderly person in the age-friendly community is used as a weight and input together with the facility usage data of each elderly person into a pre-trained neural network (GNN can be used, and the training process is a well-known technique, which will not be elaborated here) to obtain the updated medical and elderly care integration level score of the age-friendly community, which is used to more accurately reflect the service supply capacity of the age-friendly community.

[0089] Finally, based on a visualization engine (such as Vue3+ECharts), the geographical location score, health score of the elderly, environmental score, service level score, hardware implementation score, and the score of the level of upgrading of medical and elderly care integration of the age-friendly community are displayed on a radar chart, thereby obtaining the adjustment radar chart of the age-friendly community.

[0090] The core value of radar charts is to "guide management decisions" rather than simply show the current situation. Therefore, in this embodiment of the invention, by dynamically adjusting the score of the level of integration of medical care and elderly care, the service directions that need to be prioritized for improvement can be identified, and resources can be directed toward key needs.

[0091] It should be noted that, in this embodiment of the present invention, for age-friendly communities where activities tend to be common, it is assumed that the individual activities of the elderly will not affect the generated results when constructing the health management radar chart of such age-friendly communities. Therefore, the radar chart is not adjusted and can be constructed directly based on the scores of the aforementioned dimensions.

[0092] In summary, the data acquisition module obtains data on facility equipment and population structure for each age-friendly community, as well as regularly compiles data on facility usage by each senior citizen during activities. This helps address the issue of "poor integration of facilities and age-friendly services," making the analysis more aligned with the actual needs of the elderly. The community clustering analysis module analyzes and clusters the differences in population structure data among different age-friendly communities, grouping communities with similar population structure characteristics into the same cluster. This avoids interference from different types of communities (such as older and younger communities), improving the targeting and accuracy of subsequent activity behavior analysis. Furthermore, the activity behavior analysis module compares the distribution characteristics of facility equipment data and facility usage data within each cluster to accurately determine the activity preferences of the elderly in the community. For communities dominated by personalized activities, the analysis of the similarity and variation characteristics of facility usage data (such as a senior citizen's preference for a specific facility) quantifies the impact indicators of personalized activities, reflecting the correlation between each senior citizen's personalized activity preferences and facility usage, providing a quantitative basis for precise intervention. Finally, in the radar chart adjustment module, based on neural networks and individual activity impact indicators, the display of the radar chart is dynamically adjusted to make the visualization results more intuitively reflect the real activity characteristics of the elderly and improve the accuracy and efficiency of community health management decisions.

[0093] Please see Figure 3This document illustrates a schematic diagram of the system architecture of a neural network-based radar map generation system for age-friendly community health management, according to an embodiment of the present invention. The system includes a processor 400, a memory 401, a bus 402, and a communication interface 403. The processor 400, communication interface 403, and memory 401 are connected via the bus 402. The memory 401 may contain a high-speed random access memory, and the bus 402 may be an ISA bus, PCI bus, or EISA bus, etc. The processor 400 may be an integrated circuit chip with signal processing capabilities. The memory 401 stores at least one instruction, at least one program, code set, or instruction set. When the processor loads and executes the at least one instruction, at least one program, code set, or instruction set, it implements the steps of various modules in a neural network-based radar map system for age-friendly community health management.

[0094] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0096] 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 principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A radar chart generation system for age-friendly community health management based on neural networks, characterized in that, The system includes: The data acquisition module is used to acquire data on the facilities and population structure of each age-friendly community, and to regularly collect data on the use of facilities by each senior citizen during activities. The community clustering analysis module is used to analyze the differences in population structure data among different age-friendly communities, and to perform cluster analysis on all age-friendly communities to obtain all clusters; The activity behavior analysis module is used to compare the distribution of facility equipment data of age-friendly communities and the distribution characteristics of facility usage data of the elderly during activities in each cluster, so as to determine the activity bias of the elderly in each age-friendly community in each cluster; for age-friendly communities with an activity bias of personalized activities, the module determines the personalized activity impact index for each elderly person based on the similarity and change characteristics of facility usage data of the elderly during activities. The radar chart adjustment module is used to adjust the radar chart of each age-friendly community with an activity bias towards individual activities, based on neural networks and individual activity impact indicators. The determination of the activity preferences of the elderly in each age-friendly community within each cluster includes: Both facility configuration data and facility usage data are in the form of a collection of facility names; Within each cluster, the similarity characteristics among the facility configuration data of all age-friendly communities are analyzed, thereby dividing the facilities in age-friendly communities in each cluster into two sets of facility configurations: common facility configuration sets and individual facility configuration sets. Within each cluster, the union of all facility usage data for all elderly people in each age-friendly community is taken as the facility usage set for that age-friendly community. The proportion of the number of facilities in the intersection of the facility usage set corresponding to each age-friendly community and the facility equipment set of each type is used as the first activity bias factor of each age-friendly community under each type of facility equipment set. In the data on the use of facilities by all elderly people in each age-friendly community, the proportion of the number of times a facility appears in each set of facilities to the total number and value of all elderly people's activities is used as the second activity bias factor for each age-friendly community under each set of facilities. In each cluster, the normalized value of the product of the first activity bias factor and the second activity bias factor of each age-friendly community under each set of facilities is used as the activity bias index of each age-friendly community under each set of facilities. The activity bias index under the common set of facilities is the common activity bias index, and the activity bias index under the individual set of facilities is the individual activity bias index. When the individual activity preference index of a certain age-friendly community is greater than or equal to the common activity preference index, the activity preference of the elderly in that age-friendly community is considered to be individual activity. The methods for obtaining the indicators of the impact of individual activities include: In any age-friendly community where activities are geared towards individualized activities, the facility usage data of each senior citizen during activities is arranged in the order of activity statistics to obtain a sorted sequence; In the sorting sequence, based on the similarity between adjacent facility usage data, the first sexual activity influencing factor for each elderly person in the age-friendly community is determined; In the sorting sequence, based on the variation characteristics between facility usage data, a second sexual activity influencing factor is determined for each elderly person in the age-friendly community; The normalized value of the product of the first and second personal activity influencing factors for each elderly person in the age-friendly community is used as the personal activity influencing index for each elderly person in the age-friendly community. The radar chart assessment dimensions for each age-friendly community include geographical location score, senior health score, environment score, service level score, hardware implementation score, and integration of medical and elderly care level score. The radar chart of age-friendly communities, which adjusts the activity bias towards individual activities based on neural networks and personalized activity impact indicators, includes: For any age-friendly community with an activity bias towards personalized activities, the personalized activity impact index corresponding to each elderly person in the age-friendly community is used as a weight, and is input into a pre-trained neural network along with the facility usage data of each elderly person to obtain the updated medical and elderly care integration level score of the age-friendly community. Based on the visualization engine, the geographical location score, health score of the elderly, environmental score, service level score, hardware implementation score, and the score of the level of upgrading of medical and elderly care integration of the age-friendly community are displayed on the radar chart, thereby obtaining the adjustment radar chart of the age-friendly community.

2. The radar chart generation system for age-friendly community health management based on neural networks according to claim 1, characterized in that, The method for obtaining the clusters includes: The population structure data includes the number of elderly people, the number of elderly men, and the average age of the elderly, and the number of elderly people and the number of elderly men in the population structure data are combined into a population vector. For any two age-friendly communities, calculate the absolute value of the difference between the average age of the elderly in the two age-friendly communities as the first difference factor, and calculate the Euclidean distance between the population vectors of the two age-friendly communities as the second difference factor. The normalized value of the product of the first difference factor and the second difference factor is used as the difference feature value between the two age-friendly communities. The K-means clustering algorithm and a preset K value are used to perform cluster analysis on all age-friendly communities to obtain all clusters. The distance metric of the clusters is the difference feature value between age-friendly communities.

3. The radar chart generation system for age-friendly community health management based on neural networks according to claim 1, characterized in that, The methods for obtaining the common facility equipment set and the personalized facility equipment set include: Within each cluster, the intersection of facility configuration data for all age-friendly communities is taken. The facilities in the intersection are used as the common facility configuration set, and the facilities outside the intersection are used as the individual facility configuration set.

4. The radar chart generation system for age-friendly community health management based on neural networks according to claim 1, characterized in that, The methods for obtaining the first personality activity influence factor include: In the sorting sequence, the Jaccard correlation coefficient between the facility usage data under each activity statistics and the facility usage data under the last activity statistics is calculated as the facility usage consistency factor corresponding to the number of activity statistics. The facility usage consistency factor corresponding to the number of activity statistics is weighted and fused using the ordinal value of the number of activity statistics in the sorting sequence, thereby obtaining the first age-friendly activity influence factor for each elderly person in the age-friendly community.

5. The radar chart generation system for age-friendly community health management based on neural networks according to claim 1, characterized in that, The methods for obtaining the second personality activity influencing factor include: In the ranking sequence corresponding to the age-friendly community, the number of facilities in the intersection of each facility usage data and the individual facility configuration set of the cluster to which the age-friendly community belongs is used as a quantity factor. The first difference sequence of the quantity factors of all facility usage data in the ranking sequence is calculated, and the mean of all values ​​in the first difference sequence is normalized and used as the second sexual activity influence factor for each elderly person in the age-friendly community.

6. The radar chart generation system for age-friendly community health management based on neural networks according to claim 1, characterized in that, The neural network is a GNN.

Images