Intelligent community resident health management method based on machine learning

Abstract

The invention relates to the technical field of smart communities, and discloses a smart community resident health management method based on machine learning, and the method comprises the steps: continuously collecting the health feedback data of residents, such as the physical condition change after outdoor activities, carrying out the correlation analysis of the health feedback data with the previous health data and environment data, dynamically updating and optimizing a correlation model, and obtaining the health feedback data of the residents; the prediction precision of the model and the effectiveness of suggestions are improved, the relationship between the use condition of the community public space and the resident health indexes is analyzed, key public facilities and areas influencing resident health are identified, data support and decision basis are provided for community planning and facility reconstruction, and the method is suitable for popularization and application. Community resident health and environment condition analysis reports are generated regularly, and resident health index change trends are visually displayed for reference of community managers and residents. According to the invention, intelligent analysis and application of environment and health data are realized, and technical support is provided for building a healthy community.

Description

Technical Field

[0001] The present invention relates to the technical field of smart communities, and in particular to a health management method for residents in smart communities based on machine learning. Background Art

[0002] When building a future community intelligent monitoring system based on the Internet of Things, we are faced with the technical challenge of how to achieve efficient integration and correlation analysis of residents' health data and environmental monitoring data. Since residents' health data is collected through smart bracelets, and environmental monitoring data comes from community environmental monitoring stations, the two types of data differ in terms of collection frequency, data format, timestamp, etc. In addition, the amount of health data and the time span of collection of different residents are also different, which further increases the difficulty of data integration.

[0003] When establishing a correlation model between environmental factors and health indicators, it is also necessary to consider how to select appropriate time series analysis and machine learning algorithms to accurately characterize the impact of environmental factors on residents' health. The model needs to be able to handle multi-dimensional, nonlinear complex relationships, and at the same time have a certain degree of robustness to noise and missing data. In addition, the model also needs to have a certain degree of interpretability in order to provide residents with credible health advice.

[0004] When analyzing the relationship between the use of community public spaces and residents' health indicators, how to accurately count the use of different public spaces is also a major challenge. Traditional manual statistical methods are difficult to achieve all-weather, full-coverage monitoring, and the intelligent monitoring system based on the Internet of Things needs to solve a series of technical problems such as sensor layout optimization, data transmission and storage, and privacy protection to achieve accurate correlation analysis between the use of public spaces and residents' health indicators. Summary of the invention

[0005] The purpose of the present invention is to overcome one or more of the above-mentioned existing technical problems and provide a smart community resident health management method based on machine learning.

[0006] To achieve the above objectives, the present invention provides a method for managing the health of residents in a smart community based on machine learning, comprising: Collect residents' health data based on smart bracelets, obtain environmental monitoring data from community environmental monitoring stations, clean and standardize these heterogeneous data, convert them into a unified data format, and store them in a time series database; Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Based on the results of time series analysis, a machine learning algorithm was selected to establish a correlation model between environmental factors and residents' health indicators, and the model was trained and optimized; Using the established correlation model, newly collected environmental monitoring data is predicted to determine the possible impact on residents' health under current environmental conditions, and personalized outdoor activity recommendations are generated based on residents' personal health characteristics; Collect residents’ health feedback data, correlate it with previous health and environmental data, and update and optimize the correlation model; Analyze the relationship between the use of community public spaces and residents’ health indicators, and identify key public facilities and areas that affect residents’ health; Generate community residents' health and environmental status analysis reports, visually display the changing trends of residents' health indicators and the impact of environmental factors, so that community managers and residents can make activity decisions based on the analysis reports.

[0007] According to one aspect of the present invention, the sensor in the smart bracelet is used to monitor and collect the user's heart rate, blood pressure and sleep quality data in real time; Collect information on pollutant concentration, temperature, humidity and ultraviolet intensity in the air through community environmental monitoring stations; Preprocess the collected health data and environmental data, including data cleaning and denoising; Standardize the preprocessed data to make it have a unified data format; Organize the standardized data according to time series and store it in a time series database; Use data mining algorithms to mine data in time series databases to extract information; Based on the mining results, provide users with personalized health advice and environmental protection measures.

[0008] According to one aspect of the present invention, based on the resident health data and environmental monitoring data in the time series database, a moving average algorithm is used to calculate the average value of the data in each time window to obtain smoothed time series data; Based on the smoothed time series data, the exponential smoothing algorithm is used to predict the trend of the data and obtain the changing trends of residents' health indicators and environmental factors in the future. According to the predicted trend of change, set the threshold of abnormal fluctuations. If the actual data exceeds the threshold range, it is judged as an abnormal situation and triggers the early warning mechanism; Obtain residents’ health data and environmental monitoring data when abnormal situations occur, analyze the differences between abnormal data and normal data, and identify the key factors that cause the abnormalities; Based on the key factors, the correlation analysis algorithm is used to calculate the correlation coefficient between residents' health indicators and environmental factors to obtain the quantitative relationship between the two; Based on the quantitative relationship, a prediction model of residents' health indicators and environmental factors is established, and the changes in residents' health indicators are predicted through environmental monitoring data; Based on the prediction results, formulate targeted health intervention measures to improve the living environment of residents.

[0009] According to one aspect of the present invention, historical records of environmental factors and residents' health indicators are obtained through time series data, and data cleaning is performed on the historical records; Based on the cleaned data, the support vector machine algorithm was used to establish the association model between environmental factors and health indicators, and the initial parameters of the model were determined by the cross-validation method; Based on the preliminary results of the association model, the model was optimized and trained based on the random forest algorithm; Based on the optimized correlation model, model evaluation is performed through error analysis and performance evaluation; Apply the optimized and trained model to actual environmental monitoring data and update the model parameters in real time.

[0010] According to one aspect of the present invention, key indicators are regularly collected through an environmental monitoring system, and data of the key indicators are transmitted to a central processing system; In the central processing system, machine learning algorithms are used to analyze the collected environmental data to predict future air quality and UV intensity; Based on the prediction results, assess the potential health impacts of current environmental conditions on residents with specific health characteristics; If the air quality index exceeds the preset threshold, personalized protective measures recommendations are generated based on the residents' health records, and the system pushes warnings of poor air quality and recommendations for outdoor activity restrictions to residents' mobile devices.

[0011] According to one aspect of the present invention, environmental monitoring equipment is deployed in the outdoor activity area of ​​residents to collect environmental data in real time, including temperature, humidity and air quality, and the collected environmental data is stored in a database; Design a health feedback questionnaire, push the questionnaire to residents through the mobile phone application, guide residents to fill in health feedback information on changes in their physical condition after outdoor activities, and upload the feedback data to the server; Read residents' historical health data from the database, including physical examination reports and medical records, clean and standardize the data, and store structured health data in a relational database; Adopt data fusion technology to associate environmental monitoring data, residents' health feedback data and historical health data to construct a multi-dimensional data set as input for association analysis; Use association rule mining algorithms to analyze multidimensional data sets, discover the association patterns between environmental factors and outdoor activities and changes in residents' physical conditions, and generate an association rule library; According to the rules in the association rule library, combined with the current environmental monitoring data and residents' outdoor activities, the decision tree algorithm is used to predict the changes in residents' physical conditions and obtain personalized health advice; Continue to collect new environmental monitoring data and residents’ health feedback data, and regularly update the association rule base and decision tree model.

[0012] According to one aspect of the present invention, by collecting and analyzing data on the use of community public spaces, data such as the frequency of use and length of stay of residents in different public facilities and areas are obtained; Based on the data of residents' health records, various health indicators of residents were extracted, including body mass index, cardiopulmonary function and incidence of chronic diseases; The correlation analysis algorithm is used to calculate the correlation coefficient matrix between the use of community public spaces and residents' health indicators, and to determine the correlation between each public facility and area and health indicators; Based on the results of the correlation analysis, key public facilities and areas that affect residents' health are identified and ranked in order of importance; Clustering algorithms are used to classify key public facilities and areas that affect residents' health, resulting in different types of public spaces, including sports, leisure, and social. Based on the actual situation of the community and the needs of residents, propose targeted planning and renovation plans for different types of key public spaces; The planning and renovation plans of key public spaces are compiled into decision-making reports to provide data support and decision-making basis for community management departments.

[0013] According to one aspect of the present invention, health data and environmental monitoring data of community residents are obtained, and the data are preprocessed, including data cleaning, missing value processing and outlier processing; For the pre-processed health data and environmental data, a correlation analysis algorithm is used to analyze the correlation between residents' health indicators and environmental factors, and the correlation strength between each health indicator and environmental factor is determined. If the correlation coefficient is greater than the preset threshold, it is determined that the health indicator and the environmental factor have a significant correlation. For health indicators and environmental factors with significant correlation, a time series analysis algorithm is used to establish a health indicator change trend prediction model; Generate a report on the health and environmental status of community residents based on the prediction results of health indicator change trends and correlation analysis results. The report includes the current level of each health indicator, its change trend and its correlation with environmental factors. Data visualization methods are used to visualize the key information in the analysis report.

[0014] To achieve the above objectives, the present invention provides a smart community resident health management system based on machine learning, comprising: Data collection module: collects residents' health data based on smart bracelets, obtains environmental monitoring data from community environmental monitoring stations, cleans and standardizes these heterogeneous data, converts them into a unified data format, and stores them in a time series database; Data analysis module: Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Model training module: Based on the results of time series analysis, select machine learning algorithms to establish the association model between environmental factors and residents' health indicators, and train and optimize the model; Activity suggestion generation module: Use the established association model to predict the newly collected environmental monitoring data, determine the possible impact on residents' health under current environmental conditions, and generate personalized outdoor activity suggestions based on residents' personal health characteristics; Model optimization module: collect residents' health feedback data, associate it with previous health data and environmental data, and update and optimize the association model; Healthy Area Identification Module: Analyzes the relationship between the use of community public spaces and residents’ health indicators, and identifies key public facilities and areas that affect residents’ health; Community management visualization module: Generates analysis reports on the health and environmental status of community residents, visually displays the changing trends of residents' health indicators and the impact of environmental factors, allowing community managers and residents to make activity decisions based on the analysis reports.

[0015] Based on this, the beneficial effects of the present invention are: the present invention can predict the impact of environmental conditions on residents' health, generate personalized outdoor activity recommendations, dynamically optimize model accuracy by continuously collecting feedback data, and provide decision-making basis for community planning by analyzing the relationship between community public space use and residents' health. It can also generate analysis reports regularly, visualize health indicators and environmental impacts, and improve community health management level and residents' health awareness. The present invention realizes intelligent analysis and application of environmental and health data, and provides technical support for building a healthy community. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of a method for health management of residents in a smart community based on machine learning according to an exemplary embodiment; Figure 2 It is a flow chart of a smart community resident health management system based on machine learning according to an exemplary embodiment. DETAILED DESCRIPTION

[0017] The present invention will now be discussed with reference to exemplary embodiments. It should be understood that the embodiments discussed are only for enabling those skilled in the art to better understand and thereby implement the present invention, rather than implying any limitation on the scope of the present invention.

[0018] As used herein, the term “including” and variations thereof are to be interpreted as open-ended terms meaning “including but not limited to.” The term “based on” is to be interpreted as “based, at least in part, on,” and the terms “one embodiment” and “an embodiment” are to be interpreted as “at least one embodiment.”

[0019] According to one embodiment of the present invention, Figure 1 is a flow chart of a method for health management of residents in a smart community based on machine learning according to an exemplary embodiment, such as Figure 1 As shown, to achieve the above purpose, the present invention provides a smart community resident health management method based on machine learning, comprising: Collect residents' health data based on smart bracelets, obtain environmental monitoring data from community environmental monitoring stations, clean and standardize these heterogeneous data, convert them into a unified data format, and store them in a time series database; Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Based on the results of time series analysis, a machine learning algorithm was selected to establish a correlation model between environmental factors and residents' health indicators, and the model was trained and optimized; Using the established correlation model, newly collected environmental monitoring data is predicted to determine the possible impact on residents' health under current environmental conditions, and personalized outdoor activity recommendations are generated based on residents' personal health characteristics; Collect residents’ health feedback data, correlate it with previous health and environmental data, and update and optimize the correlation model; Analyze the relationship between the use of community public spaces and residents’ health indicators, and identify key public facilities and areas that affect residents’ health; Generate community residents' health and environmental status analysis reports, visually display the changing trends of residents' health indicators and the impact of environmental factors, so that community managers and residents can make activity decisions based on the analysis reports.

[0020] According to one embodiment of the present invention, the sensor in the smart bracelet is used to monitor and collect the user's heart rate, blood pressure and sleep quality data in real time; Collect information on pollutant concentration, temperature, humidity and ultraviolet intensity in the air through community environmental monitoring stations; Preprocess the collected health data and environmental data, including data cleaning and denoising; Standardize the preprocessed data to make it have a unified data format; Organize the standardized data according to time series and store it in a time series database; Use data mining algorithms to mine data in time series databases to extract information; Based on the mining results, provide users with personalized health advice and environmental protection measures.

[0021] According to one embodiment of the present invention, based on the resident health data and environmental monitoring data in the time series database, a moving average algorithm is used to calculate the average value of the data in each time window to obtain smoothed time series data; Based on the smoothed time series data, the exponential smoothing algorithm is used to predict the trend of the data and obtain the changing trends of residents' health indicators and environmental factors in the future. According to the predicted trend of change, set the threshold of abnormal fluctuations. If the actual data exceeds the threshold range, it is judged as an abnormal situation and triggers the early warning mechanism; Obtain residents’ health data and environmental monitoring data when abnormal situations occur, analyze the differences between abnormal data and normal data, and identify the key factors that cause the abnormalities; Based on the key factors, the correlation analysis algorithm is used to calculate the correlation coefficient between residents' health indicators and environmental factors to obtain the quantitative relationship between the two; Based on the quantitative relationship, a prediction model of residents' health indicators and environmental factors is established, and the changes in residents' health indicators are predicted through environmental monitoring data; Based on the prediction results, formulate targeted health intervention measures to improve the living environment of residents.

[0022] According to one embodiment of the present invention, historical records of environmental factors and residents' health indicators are obtained through time series data, and data cleaning is performed on the historical records; Based on the cleaned data, the support vector machine algorithm was used to establish the association model between environmental factors and health indicators, and the initial parameters of the model were determined by the cross-validation method; Based on the preliminary results of the association model, the model was optimized and trained based on the random forest algorithm; Based on the optimized correlation model, model evaluation is performed through error analysis and performance evaluation; Apply the optimized and trained model to actual environmental monitoring data and update the model parameters in real time.

[0023] According to one embodiment of the present invention, key indicators are regularly collected through an environmental monitoring system, and data of the key indicators are transmitted to a central processing system; In the central processing system, machine learning algorithms are used to analyze the collected environmental data to predict future air quality and UV intensity; Based on the prediction results, assess the potential health impacts of current environmental conditions on residents with specific health characteristics; If the air quality index exceeds the preset threshold, personalized protective measures recommendations are generated based on the residents' health records, and the system pushes warnings of poor air quality and recommendations for outdoor activity restrictions to residents' mobile devices.

[0024] According to one embodiment of the present invention, environmental monitoring equipment is deployed in the outdoor activity area of ​​residents to collect environmental data in real time, including temperature, humidity and air quality, and the collected environmental data is stored in a database; Design a health feedback questionnaire, push the questionnaire to residents through the mobile phone application, guide residents to fill in health feedback information on changes in their physical condition after outdoor activities, and upload the feedback data to the server; Read residents' historical health data from the database, including physical examination reports and medical records, clean and standardize the data, and store structured health data in a relational database; Adopt data fusion technology to associate environmental monitoring data, residents' health feedback data and historical health data to construct a multi-dimensional data set as input for association analysis; Use association rule mining algorithms to analyze multidimensional data sets, discover the association patterns between environmental factors and outdoor activities and changes in residents' physical conditions, and generate an association rule library; According to the rules in the association rule library, combined with the current environmental monitoring data and residents' outdoor activities, the decision tree algorithm is used to predict the changes in residents' physical conditions and obtain personalized health advice; Continue to collect new environmental monitoring data and residents’ health feedback data, and regularly update the association rule base and decision tree model.

[0025] According to one embodiment of the present invention, by collecting and analyzing data on the use of community public spaces, data such as the frequency of use and duration of stay of residents in different public facilities and areas are obtained; Based on the data of residents' health records, various health indicators of residents were extracted, including body mass index, cardiopulmonary function and incidence of chronic diseases; The correlation analysis algorithm is used to calculate the correlation coefficient matrix between the use of community public spaces and residents' health indicators, and to determine the correlation between each public facility and area and health indicators; Based on the results of the correlation analysis, key public facilities and areas that affect residents' health are identified and ranked in order of importance; Clustering algorithms are used to classify key public facilities and areas that affect residents' health, resulting in different types of public spaces, including sports, leisure, and social. Based on the actual situation of the community and the needs of residents, propose targeted planning and renovation plans for different types of key public spaces; The planning and renovation plans of key public spaces are compiled into decision-making reports to provide data support and decision-making basis for community management departments.

[0026] According to one embodiment of the present invention, health data and environmental monitoring data of community residents are obtained, and the data are preprocessed, including data cleaning, missing value processing and outlier processing; For the pre-processed health data and environmental data, a correlation analysis algorithm is used to analyze the correlation between residents' health indicators and environmental factors, and the correlation strength between each health indicator and environmental factor is determined. If the correlation coefficient is greater than the preset threshold, it is determined that the health indicator and the environmental factor have a significant correlation. For health indicators and environmental factors with significant correlation, a time series analysis algorithm is used to establish a health indicator change trend prediction model; Generate a report on the health and environmental status of community residents based on the prediction results of health indicator change trends and correlation analysis results. The report includes the current level of each health indicator, its change trend and its correlation with environmental factors. Data visualization methods are used to visualize the key information in the analysis report.

[0027] According to one embodiment of the present invention, the sensor in the smart bracelet collects the user's heart rate, blood pressure and sleep quality data every 5 minutes, and the community environmental monitoring station updates the air pollutant concentration, temperature and humidity, ultraviolet intensity and other information every 10 minutes. The system uses the Kalman filter algorithm to preprocess the collected health data and environmental data, remove outliers and noise interference, and ensure the accuracy and reliability of the data. Then, the preprocessed data is mapped to the [0,1] interval using the minimum-maximum normalization method to achieve data standardization. The standardized data is organized according to timestamps and stored in the InfluxDB time series database. Each data point contains attributes such as timestamps, data types and values. Based on the data in the time series database, the system uses the ARIMA model to perform time series prediction, and predicts changes in the future by analyzing the historical trends of user health data and environmental data. Based on the prediction results, the system provides users with personalized health advice. For example, when it is predicted that the air pollution index will be high in the next week, it is recommended that users minimize outdoor activities and wear masks when traveling; when it is predicted that the user's sleep quality will decline, it is recommended that users adjust their work and rest time to ensure adequate sleep. At the same time, the system can also make optimization suggestions for the community's environmental protection measures based on environmental monitoring data. For example, during periods of severe air pollution, it is recommended that the community increase the treatment of industrial waste gas and install air purification equipment in public places. By comprehensively analyzing user health data and environmental monitoring data, the system can provide scientific and effective decision-making support for user health management and community environmental protection.

[0028] According to one embodiment of the present invention, based on the resident health data and environmental monitoring data in the time series database, a moving average algorithm is used, the time window is set to 30 days, and the average value in each window is calculated to obtain smoothed time series data. For example, for the blood pressure data of residents, the original data is 130 / 80mmHg, 135 / 85mmHg, 140 / 90mmHg, etc. After moving average processing, the smoothed data is 132 / 82mmHg, 136 / 86mmHg, 138 / 88mmHg, etc. Then, an exponential smoothing algorithm is used, the smoothing coefficient is set to 6, and the trend of the smoothed data is predicted to predict the changing trend of the residents' blood pressure and ambient temperature in the next 30 days. According to the prediction results, the abnormal blood pressure fluctuation threshold is set to 150 / 95mmHg, and the abnormal temperature fluctuation threshold is set to 35°C. When the actual data exceeds the threshold, the early warning mechanism is triggered. Obtain data when abnormal conditions occur, such as blood pressure of 155 / 100mmHg and temperature of 38℃, analyze the difference between abnormal data and normal data, and determine that the key factor leading to abnormal blood pressure is high temperature. Using the Pearson correlation coefficient method, the correlation coefficient between residents' blood pressure and ambient temperature is calculated to be 8, and the quantitative relationship between the two is obtained. A prediction model for blood pressure and temperature is established. According to regression analysis, the model equation is: blood pressure = 5*temperature + 80. According to the model, when the temperature is predicted to be 30℃, 32℃, and 34℃ in the next three days, it is predicted that residents' blood pressure will reach 95 / 145mmHg, 96 / 148mmHg, and 97 / 151mmHg, which is beyond the normal range. According to the prediction results, cooling measures are formulated, such as increasing greening and installing air conditioners in residential areas to improve the living environment of residents. It is expected that the indoor temperature can be reduced by 5℃, so that blood pressure can be kept within the normal range, the health level of residents can be improved, and the coordinated development of residents' health and environmental factors can be achieved.

[0029] According to one embodiment of the present invention, in the study of the association analysis between environmental factors and residents' health indicators, firstly, by collecting time series data, such as the daily air quality index (AQI) and residents' health data such as the number of people registered for respiratory diseases in hospitals, data cleaning technology is used to remove missing values ​​and outliers, for example, the values ​​in AQI exceeding 500 are regarded as abnormal and deleted. Then, the linear interpolation method is used to process the remaining missing data to ensure the integrity and consistency of the data. After the data preprocessing is completed, the support vector machine (SVM) algorithm is used to build a preliminary model, and cross-validation is performed by setting different kernel functions such as radial basis function (RBF) and adjusting the penalty parameter C (such as C = 1, 10, 100) to select the best model parameter combination. After the preliminary model is established, the random forest algorithm is introduced for feature selection, and the environmental variables that contribute most to model prediction are screened out by calculating the weights of the influence of various environmental factors such as temperature, humidity, and AQI on health indicators. With the help of random forest, the parameters of the support vector machine model were adjusted again, and grid search technology was used to refine parameters such as kernel function type, C value and gamma parameter. After multiple rounds of training and verification, the optimal parameter settings were found. Finally, the optimized model was applied to new environmental monitoring data, and the model parameters were adjusted in real time to ensure that the model can accurately reflect the impact of environmental changes on residents' health, thereby achieving an accurate description and prediction of the relationship between environmental factors and health indicators.

[0030] According to one embodiment of the present invention, the environmental monitoring system collects key indicators such as PM5, ozone concentration and ultraviolet index every 15 minutes, and transmits the data to the central processing system through a wireless sensor network. The central processing system uses a support vector machine (SVM) algorithm to train environmental data from the past 30 days to establish an air quality and ultraviolet intensity prediction model. The model predicts the air quality index (AQI) and ultraviolet index (UVI) in the next 24 hours based on current data. If the AQI exceeds 100 or the UVI exceeds 7, the system automatically pushes warnings and suggestions to residents' mobile devices with specific health characteristics (such as respiratory diseases, photosensitive skin, etc.), such as "Tomorrow's AQI may reach 150, it is recommended to reduce outdoor activity time". At the same time, the system uses a decision tree algorithm to generate personalized protection measures based on the residents' skin type (I-VI type), allergy history and other health records, such as "Your skin belongs to type III, the current UVI is 8, it is recommended to use SPF30+ sunscreen and avoid going out from 10:00-14:00". The system updates health risk assessments and protection suggestions once a day to help residents manage health risks in real time.

[0031] According to one embodiment of the present invention, environmental monitoring equipment deployed in the outdoor activity area of ​​residents collects temperature, humidity, air quality and other indicators in real time, such as temperature range from -10°C to 50°C, humidity from 10% to 90%, and PM5 concentration from 0μg / m³ to 200μg / m³. These data are updated every minute and transmitted to the central database for storage via a wireless network. Through a mobile phone application, residents receive a health feedback questionnaire after outdoor activities and fill in their health conditions, such as dyspnea, allergic reactions, etc., which are then uploaded to the server. The server extracts the residents' historical health data from the database, such as blood pressure and blood sugar values ​​in the physical examination report, and removes any abnormal values ​​or missing data through data cleaning, such as replacing missing blood sugar values ​​with average values. Then, using data fusion technology, such as weighted average method, real-time environmental monitoring data is combined with residents' health feedback and historical health data to construct a multidimensional data set. Association rule mining algorithms, such as Apriori algorithm, are applied to this data set to analyze the association pattern between environmental factors and changes in residents' health status, thereby generating an association rule base. For example, it was found that when PM5 exceeds 150μg / m³, the probability of residents experiencing respiratory discomfort increases by 30%. Based on these rules, combined with current environmental data and residents' outdoor activity records, a decision tree algorithm is used to predict possible changes in residents' health status and provide personalized health recommendations, such as recommending reducing outdoor exercise when air quality is poor. As new data continues to be collected, the system regularly updates the association rule base and decision tree model, and uses incremental learning methods to adjust model parameters, such as adjusting the depth of the decision tree, to improve the accuracy of predictions and the effectiveness of recommendations.

[0032] According to one embodiment of the present invention, in the study of the relationship between community public space and residents' health, data is first collected by installing sensors and using mobile applications, for example, by using infrared sensors and GPS tracking technology to record the frequency of residents' activities and the length of stay in parks, fitness areas and rest areas. For example, the data shows that the average length of stay of residents in the fitness area is 30 minutes each time, and the average number of visits per week is 4 times. Then, using the data in the residents' health records, such as the average body mass index (BMI) of 25, cardiopulmonary function indicators, etc., the Pearson correlation coefficient algorithm is used in the Python programming language to calculate the correlation between the frequency of use, the length of stay and the health indicators. The analysis results may show that the correlation coefficient between the frequency of use of the fitness area and the body mass index is -6, indicating that a higher frequency of use is associated with a lower body mass index. Based on these analysis results, public facilities can be ranked in importance to identify the facilities that have the greatest impact on residents' health. Then, the K-means clustering algorithm is used to classify public spaces into different types, such as classifying areas with frequent activities and complete fitness equipment as sports-type public spaces. Finally, a specific transformation plan is proposed based on the specific conditions of the community, such as the age structure and health needs of residents. For example, for sports public spaces, smart fitness equipment will be added, which can record the user's exercise data and provide health advice, while optimizing paths and lighting, and adding more rest areas to meet the needs of the elderly and children. These planning and transformation plans will be integrated into a detailed decision-making report to provide scientific data support for community management departments, helping them make more targeted decisions, thereby effectively improving the health level and quality of life of residents.

[0033] According to one embodiment of the present invention, after obtaining the health data and environmental monitoring data of community residents, the data is preprocessed using Python's Pandas library. Through data cleaning, duplicate records and invalid data are removed, missing values ​​are processed by mean filling or KNN interpolation, and outliers are identified and eliminated by box plot method or Z-score method to ensure that the accuracy and completeness of the data reach more than 95%. Then the Pearson correlation coefficient is used to calculate the correlation between health indicators and environmental factors to obtain a correlation coefficient matrix. If the correlation coefficient is greater than 6, it is determined that the health indicator has a significant correlation with the environmental factor. For the PM5 concentration and the incidence of respiratory diseases among residents with significant correlation, the ARIMA time series model is used to establish a health indicator change trend prediction model for the next 3 months, and the model accuracy rate reaches more than 85%. According to the prediction results and correlation analysis results, a health and environmental status analysis report is generated, and data visualization libraries such as Matplotlib and Seaborn are used to generate a line chart to show the trend of health indicator changes, a scatter plot to show the correlation analysis results, and a radar chart to show the comprehensive evaluation results of multiple indicators, so that the report content is more intuitive and easy to understand. Finally, a Web application was built using the Django framework to regularly push analysis reports to community managers and residents, and to provide functions such as intelligent recommendations for health knowledge and online consultations, further improving the level of community health management and residents' health awareness.

[0034] Furthermore, in order to achieve the above-mentioned purpose of the invention, the present invention also provides a smart community resident health management system based on machine learning. Figure 2 is a flow chart of a smart community resident health management system based on machine learning according to an exemplary embodiment. Figure 2 As shown, a smart community resident health management system based on machine learning in the present invention includes: Data collection module: collects residents' health data based on smart bracelets, obtains environmental monitoring data from community environmental monitoring stations, cleans and standardizes these heterogeneous data, converts them into a unified data format, and stores them in a time series database; Data analysis module: Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Model training module: Based on the results of time series analysis, select machine learning algorithms to establish the association model between environmental factors and residents' health indicators, and train and optimize the model; Activity suggestion generation module: Use the established association model to predict the newly collected environmental monitoring data, determine the possible impact on residents' health under current environmental conditions, and generate personalized outdoor activity suggestions based on residents' personal health characteristics; Model optimization module: collect residents' health feedback data, associate it with previous health data and environmental data, and update and optimize the association model; Healthy Area Identification Module: Analyzes the relationship between the use of community public spaces and residents’ health indicators, and identifies key public facilities and areas that affect residents’ health; Community management visualization module: Generates analysis reports on the health and environmental status of community residents, visually displays the changing trends of residents' health indicators and the impact of environmental factors, allowing community managers and residents to make activity decisions based on the analysis reports.

[0035] Based on this, the beneficial effect of the present invention is that the present invention can predict the impact of environmental conditions on residents' health, generate personalized outdoor activity recommendations, dynamically optimize model accuracy by continuously collecting feedback data, and provide decision-making basis for community planning by analyzing the relationship between community public space use and residents' health. It can also generate analysis reports regularly, visualize health indicators and environmental impacts, and improve community health management level and residents' health awareness. The present invention realizes intelligent analysis and application of environmental and health data, and provides technical support for building a healthy community.

[0036] Those of ordinary skill in the art will appreciate that the modules and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0037] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and equipment can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0038] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0039] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0040] In addition, each functional module in the embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0041] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the energy-saving signal sending / receiving method of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0042] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

[0043] It should be understood that the size of the serial numbers of each step in the content of the invention and the embodiments of the present invention does not absolutely mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

Claims

1. A method for health management of residents in a smart community based on machine learning, characterized in that: include: Collect residents' health data based on smart bracelets, obtain environmental monitoring data from community environmental monitoring stations, clean and standardize these heterogeneous data, convert them into a unified data format, and store them in a time series database; Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Based on the results of time series analysis, a machine learning algorithm was selected to establish a correlation model between environmental factors and residents' health indicators, and the model was trained and optimized; Using the established correlation model, newly collected environmental monitoring data is predicted to determine the possible impact on residents' health under current environmental conditions, and personalized outdoor activity recommendations are generated based on residents' personal health characteristics; Collect residents’ health feedback data, correlate it with previous health and environmental data, and update and optimize the correlation model; Analyze the relationship between the use of community public spaces and residents’ health indicators, and identify key public facilities and areas that affect residents’ health; Generate community residents' health and environmental status analysis reports, visually display the changing trends of residents' health indicators and the impact of environmental factors, so that community managers and residents can make activity decisions based on the analysis reports.

2. A method for health management of smart community residents based on machine learning as claimed in claim 1, characterized in that: Using sensors in smart bracelets, users’ heart rate, blood pressure and sleep quality data can be monitored and collected in real time; Collect information on pollutant concentration, temperature, humidity and ultraviolet intensity in the air through community environmental monitoring stations; Preprocess the collected health data and environmental data, including data cleaning and denoising; Standardize the preprocessed data to make it have a unified data format; Organize the standardized data according to time series and store it in a time series database; Use data mining algorithms to mine data in time series databases to extract information; Based on the mining results, provide users with personalized health advice and environmental protection measures.

3. A method for health management of smart community residents based on machine learning as claimed in claim 2, characterized in that: Based on the resident health data and environmental monitoring data in the time series database, the moving average algorithm is used to calculate the average value of the data in each time window to obtain the smoothed time series data; Based on the smoothed time series data, the exponential smoothing algorithm is used to predict the trend of the data and obtain the changing trends of residents' health indicators and environmental factors in the future. According to the predicted trend of change, set the threshold of abnormal fluctuations. If the actual data exceeds the threshold range, it is judged as an abnormal situation and triggers the early warning mechanism; Obtain residents’ health data and environmental monitoring data when abnormal situations occur, analyze the differences between abnormal data and normal data, and identify the key factors that cause the abnormalities; Based on the key factors, the correlation analysis algorithm is used to calculate the correlation coefficient between residents' health indicators and environmental factors to obtain the quantitative relationship between the two; Based on the quantitative relationship, a prediction model of residents' health indicators and environmental factors is established, and the changes in residents' health indicators are predicted through environmental monitoring data; Based on the prediction results, formulate targeted health intervention measures to improve the living environment of residents.

4. A method for health management of smart community residents based on machine learning as claimed in claim 3, characterized in that: Obtain historical records of environmental factors and residents’ health indicators through time series data, and clean the historical records; Based on the cleaned data, the support vector machine algorithm was used to establish the association model between environmental factors and health indicators, and the initial parameters of the model were determined by the cross-validation method; Based on the preliminary results of the association model, the model was optimized and trained based on the random forest algorithm; Based on the optimized correlation model, model evaluation is performed through error analysis and performance evaluation; Apply the optimized and trained model to actual environmental monitoring data and update the model parameters in real time.

5. A method for health management of smart community residents based on machine learning as claimed in claim 4, characterized in that: Regularly collect key indicators through the environmental monitoring system and transmit the data of key indicators to the central processing system; In the central processing system, machine learning algorithms are used to analyze the collected environmental data to predict future air quality and UV intensity; Based on the prediction results, assess the potential health impacts of current environmental conditions on residents with specific health characteristics; If the air quality index exceeds the preset threshold, personalized protective measures recommendations are generated based on the residents' health records, and the system pushes warnings of poor air quality and recommendations for outdoor activity restrictions to residents' mobile devices.

6. A method for health management of smart community residents based on machine learning as claimed in claim 5, characterized in that: By deploying environmental monitoring equipment in residents' outdoor activity areas, real-time environmental data, including temperature, humidity and air quality, is collected and stored in a database; Design a health feedback questionnaire, push the questionnaire to residents through the mobile phone application, guide residents to fill in health feedback information on changes in their physical condition after outdoor activities, and upload the feedback data to the server; Read residents' historical health data from the database, including physical examination reports and medical records, clean and standardize the data, and store structured health data in a relational database; Adopt data fusion technology to associate environmental monitoring data, residents' health feedback data and historical health data to construct a multi-dimensional data set as input for association analysis; Use association rule mining algorithms to analyze multidimensional data sets, discover the association patterns between environmental factors and outdoor activities and changes in residents' physical conditions, and generate an association rule library; According to the rules in the association rule library, combined with the current environmental monitoring data and residents' outdoor activities, the decision tree algorithm is used to predict the changes in residents' physical conditions and obtain personalized health advice; Continue to collect new environmental monitoring data and residents’ health feedback data, and regularly update the association rule base and decision tree model.

7. A method for health management of smart community residents based on machine learning as claimed in claim 6, characterized in that: By collecting and analyzing data on the use of community public spaces, we can obtain data such as the frequency of use and length of stay of residents in different public facilities and areas; Based on the data of residents' health records, various health indicators of residents were extracted, including body mass index, cardiopulmonary function and incidence of chronic diseases; The correlation analysis algorithm is used to calculate the correlation coefficient matrix between the use of community public spaces and residents' health indicators, and to determine the correlation between each public facility and area and health indicators; Based on the results of the correlation analysis, key public facilities and areas that affect residents' health are identified and ranked in order of importance; Clustering algorithms are used to classify key public facilities and areas that affect residents' health, resulting in different types of public spaces, including sports, leisure, and social. Based on the actual situation of the community and the needs of residents, propose targeted planning and renovation plans for different types of key public spaces; The planning and renovation plans of key public spaces are compiled into decision-making reports to provide data support and decision-making basis for community management departments.

8. A method for health management of smart community residents based on machine learning as claimed in claim 7, characterized in that: Obtain health data and environmental monitoring data of community residents, and pre-process the data, including data cleaning, missing value processing, and outlier processing; For the pre-processed health data and environmental data, a correlation analysis algorithm is used to analyze the correlation between residents' health indicators and environmental factors, and the correlation strength between each health indicator and environmental factor is determined. If the correlation coefficient is greater than the preset threshold, it is determined that the health indicator and the environmental factor have a significant correlation. For health indicators and environmental factors with significant correlation, a time series analysis algorithm is used to establish a health indicator change trend prediction model; Generate a report on the health and environmental status of community residents based on the prediction results of health indicator change trends and correlation analysis results. The report includes the current level of each health indicator, its change trend and its correlation with environmental factors. Data visualization methods are used to visualize the key information in the analysis report.

9. A smart community resident health management system based on machine learning, characterized in that: include: Data collection module: collects residents' health data based on smart bracelets, obtains environmental monitoring data from community environmental monitoring stations, cleans and standardizes these heterogeneous data, converts them into a unified data format, and stores them in a time series database; Data analysis module: Based on the resident health data and environmental monitoring data in the time series database, the time series analysis algorithm is used to perform trend analysis and anomaly detection on the data to identify the changing trends and anomalies of resident health indicators and environmental factors; Model training module: Based on the results of time series analysis, select machine learning algorithms to establish the association model between environmental factors and residents' health indicators, and train and optimize the model; Activity suggestion generation module: Use the established association model to predict the newly collected environmental monitoring data, determine the possible impact on residents' health under current environmental conditions, and generate personalized outdoor activity suggestions based on residents' personal health characteristics; Model optimization module: collect residents’ health feedback data, associate it with previous health data and environmental data, and update and optimize the association model; Healthy Area Identification Module: Analyzes the relationship between the use of community public spaces and residents’ health indicators, and identifies key public facilities and areas that affect residents’ health; Community management visualization module: Generates analysis reports on the health and environmental status of community residents, visually displays the changing trends of residents' health indicators and the impact of environmental factors, allowing community managers and residents to make activity decisions based on the analysis reports.

Images