Health propaganda and education method and system based on intelligent system

Abstract

The invention relates to the field of health management, and discloses a health propaganda and education method and system based on an intelligent system, and the method comprises the steps: collecting environment data and multi-dimensional health data of a service object, and creating a scene perception layer; generating a causal relationship chain by using a preset Bayesian network optimization algorithm; determining an adaptive propaganda and education opportunity of the service object based on the health state and the scene type; constructing a self-management scoring device of the service object, and generating personalized propaganda and education contents of the service object; and creating a multi-scene collaborative propaganda and education platform of the service object and a corresponding effect evaluation system thereof. According to the invention, multi-scene collaborative propaganda and education can be realized, and targeted health propaganda and education contents are provided.

Description

Technical Field

[0001] The present invention relates to a health education method and system based on an intelligent system, belonging to the field of health management. Background Art

[0002] Health education refers to an educational activity that uses planned, organized and systematic educational activities to convey scientific health knowledge and skills to individuals or groups, guide individuals or groups to establish correct health concepts, help them consciously adopt and maintain behaviors and lifestyles that are beneficial to health, and aims to restore, maintain and promote the health of individuals or groups. With the acceleration of urbanization, people's lifestyles, eating habits and working environments have undergone tremendous changes. These changes are often accompanied by increased health risks and intensified urban air pollution, which has a negative impact on the health of residents' respiratory and cardiovascular systems. Therefore, strengthening health education to reduce health risks has become an extremely urgent need at present.

[0003] Traditional health education methods mainly spread health knowledge through distributing health brochures, posting posters, etc., and most of the education materials are universal. They cannot be adjusted according to the health status, cultural level, and acceptance ability of different groups of people, nor can they switch education strategies according to the scenarios in which the service recipients are located.

[0004] Therefore, there is an urgent need for a solution that can achieve collaborative education in multiple scenarios and provide targeted health education content. Summary of the Invention

[0005] The present invention provides a health education method and system based on an intelligent system, the main purpose of which is to achieve multi-scenario collaborative education and provide targeted health education content.

[0006] To achieve the above objectives, the present invention provides a health education method based on an intelligent system, comprising:

[0007] Acquire a service target for health education, collect environmental data and multi-dimensional health data of the service target, perform fuzzy processing on the multi-dimensional health data, and obtain a target health data set;

[0008] Creating a scene perception layer for the service object based on the environmental data and the target health data set, and identifying the scene type of the service object according to the scene perception layer;

[0009] generating a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyzing the health status of the service recipient based on the causal relationship chain, and determining an adaptive education opportunity for the service recipient based on the health status and the scenario type;

[0010] Identifying the intervention behavior of the service recipient based on the health status and the causal chain, constructing a self-management scorer for the service recipient based on the intervention behavior, and generating personalized education content for the service recipient based on the self-management scorer and the scenario perception layer;

[0011] Based on the adaptive education timing, the personalized education content, and the scenario type, a multi-scenario collaborative education platform is created for the service object; based on the multi-scenario collaborative education platform and the health status, the health empowerment level of the service object is identified; and based on the health empowerment level, an effect evaluation system of the multi-scenario collaborative education platform is generated;

[0012] In combination with the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system, health education processing of the service object is performed to obtain health education results.

[0013] Optionally, the creating a scene perception layer of the service object based on the environmental data and the target health data set includes:

[0014] Formatting the environmental data and the target health data set to obtain formatted data;

[0015] extracting key features from the environmental data and the target health data set based on the formatted data;

[0016] identifying scene semantics of the environmental data and the target health data set based on the key features;

[0017] Based on the scene semantics, generating a scene graph of the service object using the environmental data and the target health data set;

[0018] Calculating the similarity between the scene graphs, and extracting similar semantic labels of the scene graphs based on the similarity;

[0019] Based on the similar semantic tags, setting a scene graph index corresponding to the scene semantics;

[0020] The scene perception layer of the service object is created by combining the scene semantics, the scene graph and the scene graph index.

[0021] Optionally, generating the causal relationship chain in the target health data set by using a preset optimized Bayesian network algorithm includes:

[0022] performing semantic analysis on the target health data set to extract potential semantic relationships of the target health data set;

[0023] Determining key parameter variables of the target health data set according to the potential semantic relationship;

[0024] Calculating the mutual information factors between the key parameter variables, and constructing an initial Bayesian network structure of the key parameter variables based on the mutual information factors;

[0025] Setting probability distribution parameters of the initial Bayesian network structure, and defining a particle encoding method for the initial Bayesian network structure and the probability distribution parameters;

[0026] According to the particle encoding method, the initial Bayesian network structure and the probability distribution parameters are converted into particles;

[0027] Identifying a scoring function of the initial Bayesian network structure, and setting a fitness function of the particle based on the scoring function;

[0028] Analyzing the causal relationship of the key parameter variables using the optimized Bayesian network algorithm according to the fitness function;

[0029] Based on the causal relationship, a causal relationship chain in the target health data set is generated.

[0030] Optionally, determining the adaptive education opportunity for the service object based on the health status and the scenario type includes:

[0031] Extracting scene features corresponding to the scene types, and identifying conversion situations between the scene types based on the scene features;

[0032] Based on the conversion situation, determining the scene transition period of the service object, and extracting transition period features of the service object during the scene transition period;

[0033] According to the characteristics of the transition period, set up a cross-scenario education method for the service object;

[0034] Based on the health status, calculating the personalized timing decision weights of the service object at different time points;

[0035] Combine the cross-scenario education method and the personalized timing decision weight to determine the adaptive education timing of the service object

[0036] Optionally, constructing a self-management scorer for the service recipient based on the intervention behavior includes:

[0037] Analyzing changes in the health indicators of the service recipient based on the intervention behavior;

[0038] Collecting the behavior tracking data and self-management report of the service recipient, and identifying the self-management ability of the service recipient based on the behavior tracking data, the self-management report and changes in the health indicators;

[0039] Extracting the constituent elements of the self-management ability, and determining the scoring dimensions and scoring indicators of the service object based on the constituent elements;

[0040] Constructing a capability assessment matrix for the service object based on the scoring dimensions and the scoring indicators;

[0041] Analyze the behavioral triggering reasons of the intervention behavior and determine the behavioral duration of the intervention behavior;

[0042] Based on the behavior triggering cause, identifying the degree of initiative of the service object's behavior;

[0043] Calculating the service recipient's willingness score for the intervention behavior based on the triggering cause of the behavior, the duration of the behavior, and the initiative level of the behavior;

[0044] Setting an interpretation of the scoring result of the service object based on the capability assessment matrix and the willingness score;

[0045] A self-management scorer for the service object is constructed by combining the capability assessment matrix, the willingness score and the interpretation of the scoring result.

[0046] Optionally, generating personalized education content for the service object based on the self-management scorer and the scenario perception layer includes:

[0047] Collecting capability matrix data and real-time environment data of the service object according to the self-management scorer and the scenario perception layer;

[0048] Performing data fusion processing on the capability matrix data and the real-time environment data to generate a personalized data archive;

[0049] Based on the personalized data profile, identifying the health needs of the service recipient in different scenarios;

[0050] Creating a multimodal content representation of the service object according to the health needs and the different scenarios;

[0051] Extracting environmental characteristics of the service object based on the real-time environmental data;

[0052] According to the environmental characteristics, setting the missionary language style of the service object;

[0053] Constructing a mapping relationship between the environmental characteristics and the missionary language style, and defining an adaptive language style for the service object based on the mapping relationship;

[0054] The personalized data file, the multimodal content presentation form and the adaptive language style are combined to generate personalized education content for the service object.

[0055] Optionally, the creating a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type includes:

[0056] Based on the adaptive education opportunity, collecting the interactive feedback data of the service object on the personalized education content, and collecting the real-time behavior data of the service object in the scenario type;

[0057] Analyze the knowledge mastery of the service recipients based on the interactive feedback data, and set the content push weight for the adaptive education opportunity;

[0058] Based on the knowledge mastery level and the content push weight, a dynamic collaborative education mode for the personalized education content in the scenario type is constructed;

[0059] Identifying the device usage status of the service object, and determining the interaction progress of the service object in the scenario type in combination with the device usage status, the knowledge mastery level and the real-time behavior data;

[0060] According to the interaction progress, a cross-scenario data synchronization mechanism of the service object is set;

[0061] Based on the cross-scenario data synchronization mechanism, construct a collaborative interaction interface for the service object under the scenario type;

[0062] By combining the dynamic collaborative education method, the cross-scenario data synchronization mechanism and the collaborative interaction interface, a multi-scenario collaborative education platform for the service object is created.

[0063] Optionally, identifying the health empowerment level of the service object according to the multi-scenario collaborative education platform and the health status includes:

[0064] Outputting multi-scenario health education data of the multi-scenario collaborative education platform, and extracting health empowerment features from the multi-scenario health education data;

[0065] Identify the information receiving preferences and information conversion rates of the service recipients on the multi-scenario collaborative education platform;

[0066] Determining the information interpretation level of the multi-scenario collaborative education platform based on the information reception preference and the information conversion rate;

[0067] Analyzing the type and urgency of the needs of the service recipient based on the health status;

[0068] Collecting psychological state data of the service recipient according to the type of demand and the urgency of the demand;

[0069] Based on the psychological state data, identifying the psychological acceptance degree of the service object to the multi-scenario health education data;

[0070] The health empowerment level of the service recipient is identified by combining the health empowerment characteristics, the information interpretation level and the psychological acceptance level.

[0071] Optionally, generating an effect evaluation system of the multi-scenario collaborative education platform based on the health empowerment level includes:

[0072] Extracting evaluation dimensions and evaluation indicators corresponding to the health empowerment level;

[0073] Identifying the project factors under the evaluation dimension, and setting the measurement method corresponding to the project factors according to the evaluation dimension and the evaluation indicator;

[0074] Outputting the measurement results of the project factors using the measurement method, and identifying the result correlation between the measurement results;

[0075] Determining the convergent validity of the multi-scenario collaborative education platform on the health empowerment level based on the measurement method and the result correlation;

[0076] Calculating the consistency level between the project factors according to the measurement results;

[0077] Based on the consistency level, generating a composite reliability of the multi-scenario collaborative education platform for the health empowerment level;

[0078] Combining the convergent validity and the composite reliability, an effectiveness evaluation system of the multi-scenario collaborative education platform is generated.

[0079] In order to solve the above problems, the present invention also provides a health education system based on an intelligent system, the system comprising:

[0080] A data acquisition module is used to obtain service objects for health education, collect environmental data and multi-dimensional health data of the service objects, and perform fuzzy processing on the multi-dimensional health data to obtain a target health data set;

[0081] A scene classification module is used to create a scene perception layer for the service object based on the environmental data and the target health data set, and identify the scene type of the service object according to the scene perception layer;

[0082] a teaching and education timing selection module, configured to generate a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyze the health status of the service object according to the causal relationship chain, and determine the adaptive teaching and education timing of the service object based on the health status and the scenario type;

[0083] a personalized output module, configured to identify the intervention behavior of the service recipient based on the health status and the causal chain, construct a self-management scorer for the service recipient based on the intervention behavior, and generate personalized education content for the service recipient based on the self-management scorer and the scenario perception layer;

[0084] a multi-scenario collaborative module, configured to create a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type; identify the health empowerment level of the service object based on the multi-scenario collaborative education platform and the health status; and generate an effect evaluation system for the multi-scenario collaborative education platform based on the health empowerment level;

[0085] The result output module is used to combine the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system to perform health education processing on the service object and obtain health education results.

[0086] Compared with the problems described in the background technology, the embodiment of the present invention creates a scene perception layer for the service object based on the environmental data and the target health data set, and can formulate a highly personalized health education plan for the unique situation of each service object, so that the education content is more in line with the actual needs of the service object, improve the accuracy and effectiveness of the education, and at the same time, can dynamically adjust the education content and method as the health status and environment of the service object change, to ensure that the education always matches the current situation of the service object, and realize continuous and dynamic health management support; further, the embodiment of the present invention generates the causal relationship in the target health data set by using a preset optimized Bayesian network algorithm Chain, can accurately find out the key factors that have a greater impact on health status, and avoid wasting energy on irrelevant factors. On the other hand, when the service objects clearly understand the causal relationship between their health status and certain behaviors or factors, they will be more willing to accept and actively cooperate with the education suggestions, thereby improving the effect of health education; the embodiment of the present invention determines the adaptive education timing of the service object based on the health status and the scenario type, and can carry out education when the service object needs relevant health knowledge most, thereby improving the effect of education. At the same time, by accurately grasping the education timing, the health education system can concentrate limited resources (such as manpower, material resources, time, etc.) where they are most needed, avoiding the waste of resources. Waste; further, the embodiment of the present invention generates personalized education content for the service object based on the self-management scorer and the scenario perception layer, and can accurately provide dynamic health education content according to the self-management level of the service object and the actual life scenario in which the service object is located, thereby promoting the service object to carry out long-term health management; the embodiment of the present invention creates a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content and the scenario type, thereby realizing highly customized, intelligent and dynamic health education services and improving the effectiveness and acceptance of health education; further, the embodiment of the present invention generates based on the health empowerment level The effect evaluation system of the multi-scenario collaborative education platform can help optimize the education strategy of the education platform and improve the pertinence and effectiveness of the education platform; finally, the embodiment of the present invention performs the health education processing of the service object by combining the scene perception layer, the multi-scenario collaborative education platform and the effect evaluation system to obtain the health education results. It can perceive the environment, physical condition, behavioral habits and other multi-dimensional information of the service object in real time, provide the service object with accurate and personalized health education content, realize cross-scenario and cross-platform education, help the service object improve health literacy and self-management ability, and optimize the resource allocation of health education to maximize the utilization of health education resources. Therefore, the health education method and system based on the intelligent system provided by the embodiment of the present invention can realize multi-scenario collaborative education and provide targeted health education content. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A flowchart of a health education method based on an intelligent system provided by one embodiment of the present invention;

[0088] Figure 2 A schematic diagram of modules for implementing the intelligent system-based health education system provided in one embodiment of the present invention.

[0089] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0090] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0091] An embodiment of the present application provides a health education method based on an intelligent system. The execution subject of the health education method based on an intelligent system includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the health education method based on an intelligent system can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0092] Example 1 Reference Figure 1 FIG. 1 is a flow chart of a health education method based on an intelligent system according to an embodiment of the present invention. In this embodiment, the health education method based on an intelligent system includes:

[0093] S1. Obtain service objects for health education, collect environmental data and multi-dimensional health data of the service objects, perform fuzzy processing on the multi-dimensional health data, and obtain a target health data set.

[0094] By obtaining the service objects of health education, the embodiments of the present invention can clarify the specific target population of health education, so as to formulate a more targeted education plan based on their characteristics. The health education refers to an educational activity that conveys scientific health knowledge and skills to individuals or groups through planned, organized and systematic educational activities, and guides service objects to establish correct health concepts. The service objects refer to individuals or groups who receive health education, such as hospital patients and community residents.

[0095] Furthermore, by collecting the environmental data and multi-dimensional health data of the service objects, the embodiments of the present invention can formulate highly personalized health education content and service plans for each service object. For example, if a service object is found to live in a highly polluted industrial area and its respiratory disease indicators are abnormal, then the health education can focus on emphasizing the harm of air pollution to respiratory health, and provide targeted protective measures and healthy living suggestions.

[0096] Optionally, the environmental data of the service object can be collected through geographic information system (GIS) technology, and the multi-dimensional health data of the service object can be collected using medical Internet of Things technology, such as a smart bracelet.

[0097] The embodiment of the present invention obtains a target health data set by fuzzifying the multi-dimensional health data, which can eliminate differences in data formats, reduce the impact of noise, and improve the stability and reliability of the data. At the same time, it can anonymize information involving personal privacy, making it impossible to directly or indirectly identify personal identity and protect data privacy. The fuzzification processing refers to the process of converting precise numerical values ​​or clear concepts into fuzzy representations, which aims to better deal with problems such as uncertainty, imprecision and heterogeneity in the data.

[0098] Optionally, the fuzzy processing of the multidimensional health data can be implemented using fuzzy logic rules, and the basic process is: first, define the fuzzy set of multidimensional health data and determine the membership function; secondly, map the original multidimensional health data to the corresponding fuzzy set according to the membership function to obtain the membership value; then, use the fuzzy rule base to describe the dependency relationship between input data and the output result judgment; finally, fuse the output results of multiple rules through logical operations, and obtain the final fuzzy fusion output result in a clear manner.

[0099] S2. Based on the environmental data and the target health data set, create a scene perception layer for the service object, and identify the scene type of the service object according to the scene perception layer.

[0100] The embodiment of the present invention creates a scene perception layer for the service object based on the environmental data and the target health data set, and can formulate a highly personalized health education plan for the unique situation of each service object, so that the education content is more in line with the actual needs of the service object, and the accuracy and effectiveness of the education are improved. At the same time, it can dynamically adjust the education content and methods as the health status and environment of the service object change, ensuring that the education always matches the current situation of the service object, and realizing continuous and dynamic health management support. The scene perception layer refers to a system architecture that can comprehensively and real-time perceive the health status and environment of the service object.

[0101] As an embodiment of the present invention, the step of creating the scene perception layer of the service object based on the environmental data and the target health data set includes:

[0102] Formatting the environmental data and the target health data set to obtain formatted data;

[0103] extracting key features from the environmental data and the target health data set based on the formatted data;

[0104] identifying scene semantics of the environmental data and the target health data set based on the key features;

[0105] Based on the scene semantics, generating a scene graph of the service object using the environmental data and the target health data set;

[0106] Calculating the similarity between the scene graphs, and extracting similar semantic labels of the scene graphs based on the similarity;

[0107] Based on the similar semantic tags, setting a scene graph index corresponding to the scene semantics;

[0108] The scene perception layer of the service object is created by combining the scene semantics, the scene graph and the scene graph index.

[0109] The formatting process refers to the normalization and standardization of the original environmental data and target health data set, including operations such as unified data format, data type conversion, missing value filling, and outlier processing. The key features refer to representative information extracted from the environmental data and target health data set that can reflect the essential characteristics and inherent laws of the data, such as weather type (sunny, rainy, etc.) and seasonal information extracted from the environmental data, and health status descriptions (healthy, sick, etc.) extracted from the health data. The scene semantics refer to the semantic description of the actual scene represented by the environmental data and target health data. The scene graph refers to a data structure that represents the relationship between the environmental data and the target health data in a graphical structure. The nodes in the graph can represent various environmental factors, health indicators, and other data elements, and the edges represent the associations between these elements, such as causal relationships and correlations. The similarity refers to a quantitative indicator used to measure the similarity between two or more scene graphs. For example, when the similarity is calculated using the graph edit distance algorithm, the smaller the edit distance, the higher the similarity. The similarity semantic label refers to the common semantic label assigned to similar scene graphs based on the similarity between the scene graphs. The scene graph index refers to a data structure for quickly locating and retrieving scene graphs.

[0110] Optionally, based on the key features, the scene semantics of the environmental data and the target health data set can be identified by a convolutional neural network, such as a convolutional neural network that can automatically learn the hierarchical feature representation of the data, extract higher-level features through multi-layer convolution and pooling operations, and then identify different scene semantics. The specific steps of generating the scene graph of the service object based on the scene semantics using the environmental data and the target health data set are as follows: first, the key elements in the environmental data and the target health data set are used as nodes of the graph, for example, "temperature" and "air quality" in the environmental data, and "heart rate" and "blood pressure" in the target health data set can all be used as nodes. Then, the edges of the graph are defined according to the relationship between the key elements. For example, if "high temperature" causes "increased heart rate", a directed edge is established between the "temperature" node and the "heart rate" node. Finally, all nodes and edges are combined into a scene graph. Based on the similar semantic labels, the scene graph index corresponding to the scene semantics can be set using an inverted index table.

[0111] Furthermore, the embodiment of the present invention can accurately judge the service object's movement status and potential health risks by identifying the scene type of the service object based on the scene perception layer, thereby providing more targeted educational content. The scene type refers to different situational categories that are comprehensively defined based on multiple factors such as the service object's environment, behavior, physiological state, etc., such as hospital scenes and sports scenes.

[0112] S2. Generate a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyze the health status of the service recipient based on the causal relationship chain, and determine the timing of adaptive education for the service recipient based on the health status and the scenario type.

[0113] The embodiment of the present invention generates a causal chain in the target health data set by utilizing a preset optimized Bayesian network algorithm, which can accurately identify key factors that have a greater impact on health status and avoid wasting energy on irrelevant factors. On the other hand, when service recipients clearly understand the causal relationship between their health status and certain behaviors or factors, they will be more willing to accept and actively cooperate with education suggestions, thereby improving the effectiveness of health education. The causal chain refers to a series of interrelated causal relationship sequences constructed by analyzing the causal relationship between various variables in the target health data set. It describes the sequence and causal dependency between events or variables. For example, a long-term high-salt diet leads to an increase in sodium ion concentration in the body, and the increase in sodium ion concentration in the body causes an increase in extracellular fluid osmotic pressure, leading to an increase in blood pressure.

[0114] As an embodiment of the present invention, the step of generating the causal relationship chain in the target health data set by using a preset optimized Bayesian network algorithm includes:

[0115] performing semantic analysis on the target health data set to extract potential semantic relationships of the target health data set;

[0116] Determining key parameter variables of the target health data set according to the potential semantic relationship;

[0117] Calculating the mutual information factors between the key parameter variables, and constructing an initial Bayesian network structure of the key parameter variables based on the mutual information factors;

[0118] Setting probability distribution parameters of the initial Bayesian network structure, and defining a particle encoding method for the initial Bayesian network structure and the probability distribution parameters;

[0119] According to the particle encoding method, the initial Bayesian network structure and the probability distribution parameters are converted into particles;

[0120] Identifying a scoring function of the initial Bayesian network structure, and setting a fitness function of the particle based on the scoring function;

[0121] Analyzing the causal relationship of the key parameter variables using the optimized Bayesian network algorithm according to the fitness function;

[0122] Based on the causal relationship, a causal relationship chain in the target health data set is generated.

[0123] Among them, the semantic analysis refers to the process of analyzing text information in the health data set through natural language processing (NLP) technology. The potential semantic relationship refers to the semantic-level association relationship hidden in the data that is not easy to directly observe, such as the causal or association relationship between living habits and health status. The key parameter variable refers to the variable in the target health data set that plays a key role in studying health status, disease occurrence and development, etc., such as blood pressure, smoking, exercise frequency, etc. The mutual information factor refers to an indicator used to measure the degree of mutual dependence between two random variables. The initial Bayesian network structure refers to a directed acyclic graph formed by connecting variables with strong correlation with directed edges by calculating the mutual information between variables. The probability distribution parameter refers to the characteristic value that describes the probability distribution of random variables. In the Bayesian network, each node has its corresponding probability distribution parameter, which is used to represent the probability of the node under different values. The particle encoding method refers to a method of converting the structure and parameter information of the Bayesian network into a particle representation. Common particle encoding methods can be binary encoding, real number encoding or symbolic encoding, etc. The particle refers to the basic element in the particle swarm optimization algorithm. Each particle corresponds to a specific Bayesian network structure and parameter combination. The scoring function refers to a function used to evaluate the degree of fit of the Bayesian network structure and parameters to a given data set. Common scoring functions include Bayesian Information Criterion (BIC), Minimum Description Length (MDL), etc. The fitness function refers to a function used to evaluate the quality of particles in the particle swarm optimization algorithm. It is usually defined based on the scoring function and is used to measure the degree of adaptability of the Bayesian network structure and parameter combination represented by each particle to the target health data set. The causal relationship refers to the intrinsic connection between health factors (such as lifestyle habits, genetic factors, environmental factors, etc.) and health status or disease occurrence.

[0124] Optionally, based on the potential semantic relationship, the key parameter variables of the target health data set can be determined using a correlation analysis method, the mutual information factor between the key parameter variables can be calculated using the Scikit-learn library in Python, and based on the particle encoding method, converting the initial Bayesian network structure and the probability distribution parameters into particles can be achieved using a particle swarm optimization algorithm.

[0125] It should be explained that the optimized Bayesian network algorithm refers to an improved algorithm for extracting causal chains from target health data sets. It measures the degree of dependence between variables by calculating the mutual information between variables, and constructs the initial Bayesian network structure based on this. By setting a particle encoding method suitable for the Bayesian network structure representation and combining the Bayesian network scoring function to define the particle fitness function, the algorithm can more effectively search for the globally optimal or approximately globally optimal Bayesian network structure, thereby accurately extracting the causal chain in the target health data set, providing strong support for applications such as health education.

[0126] Furthermore, the embodiments of the present invention can deeply explore the intrinsic connections between various factors in the health data by analyzing the health status of the service object based on the causal chain, accurately locate the key factors and causal paths that affect the health status of the service object, and help the service object better understand the relationship between health and disease. The health status refers to the comprehensive condition of the service object in terms of physiology, psychology and social adaptation, for example, the service object has normal physical functions, a healthy lifestyle, and harmonious interpersonal relationships.

[0127] Optionally, based on the causal chain, the health status of the service object can be analyzed by constructing a causal network. For example, the causal chain can be used to determine the key factors affecting the health status of the service object, and a causal network of the key factors can be constructed. Based on the causal network, the interactions and influence paths between different factors can be identified, thereby helping to understand the formation mechanism of the health status.

[0128] The embodiment of the present invention determines the adaptive education timing of the service object based on the health status and the scenario type, so that education can be carried out when the service object most needs relevant health knowledge, thereby improving the education effect. At the same time, by accurately grasping the education timing, the health education system can concentrate limited resources (such as manpower, material resources, time, etc.) where they are most needed, avoiding waste of resources. The adaptive education timing refers to the dynamic time point for recommending health education content based on the health status of the service object and the scenario type. For example, after the vital signs of hospitalized patients are stabilized and the pain is controlled after surgery, education is carried out during the time period when their emotions are relatively stable (such as 10:00-11:00 in the morning).

[0129] As an embodiment of the present invention, determining the adaptive education opportunity for the service object based on the health status and the scenario type includes:

[0130] Extracting scene features corresponding to the scene types, and identifying conversion situations between the scene types based on the scene features;

[0131] Based on the conversion situation, determining the scene transition period of the service object, and extracting transition period features of the service object during the scene transition period;

[0132] According to the characteristics of the transition period, set up a cross-scenario education method for the service object;

[0133] Based on the health status, calculating the personalized timing decision weights of the service object at different time points;

[0134] The adaptive teaching timing of the service object is determined by combining the cross-scenario teaching method and the personalized timing decision weight.

[0135] Among them, the scene characteristics refer to the attributes that can describe and distinguish different scene types. For example, the characteristics of the hospital scene include professional medical equipment, frequent exchanges of medical staff, and specific spatial layouts such as wards and treatment rooms. The conversion situation refers to the situation where the service object changes from one scene type to another. The scene transition period refers to the time period experienced by the service object from one scene type to another, such as the time period from the service object leaving the hospital to fully adapting to home rehabilitation life. The transition period characteristics refer to the characteristics exhibited by the service object during the scene transition period, including the duration of the transition period, the main behaviors and needs of the service object during the transition period, and the difficulties and problems encountered. The cross-scenario education method refers to the specific method and form of health education in the process of different scenario conversions, which is formulated according to the characteristics of the scenario transition period, so as to ensure the continuity, adaptability and gradual transmission of health knowledge. The personalized timing decision weight refers to the different weight values ​​​​given to health education at different time points based on the health status of the service recipients, which is used to measure the importance and urgency of education at various time points. For example, for service recipients with chronic diseases and unstable conditions, the weight of relevant treatment and nursing knowledge education when the condition fluctuates is higher; and for service recipients with stable conditions who are in the rehabilitation period, the weight of education at key nodes of rehabilitation training is greater.

[0136] Optionally, according to the scene characteristics, the conversion between the scene types can be identified by a hidden Markov model, and based on the conversion situation, the scene transition period of the service object can be determined using a dynamic time warping algorithm, and according to the transition period characteristics, the cross-scene education method of the service object can be set by a cluster analysis algorithm.

[0137] In an optional embodiment of the present invention, based on the health status, the personalized timing decision weights of the service object at different time points are calculated using the following formula:

[0138]

[0139] in, Represents the personalized timing decision weight of the service object at different time points, Indicates the weight of the e-th health indicator in the health status, which can be 0.2, represents the actual value of the e-th health indicator at time point t, represents the minimum value of the e-th health indicator at all time points, represents the maximum value of the e-th health indicator at all time points, m represents the number of health indicators in the health state, and e represents the health indicator index.

[0140] It should be noted that, in this application, by normalizing the value of each health indicator, the health status of the service object at different time points can be dynamically reflected. In particular, it should be noted that the formula Represents a normalization function, which is used to convert the value of the e-th health indicator at time point t Mapped to the interval [0, 1] to ensure comparability between different indicators.

[0141] S4. Identify the intervention behavior of the service object based on the health status and the causal chain, construct a self-management scorer for the service object based on the intervention behavior, and generate personalized education content for the service object based on the self-management scorer and the scenario perception layer.

[0142] The embodiment of the present invention can enhance the service object's understanding and application ability of health knowledge by identifying the service object's intervention behavior based on the health status and the causal chain. The intervention behavior refers to the specific actions taken by the service object to improve or maintain his or her own health status, for example, a service object with a chronic disease takes medicine on time and regularly monitors blood sugar or blood pressure.

[0143] Optionally, based on the health status and the causal chain, the intervention behavior of the service object can be identified through a decision tree model, such as using health status indicators (such as blood pressure, blood sugar levels, etc.), lifestyle data (such as eating frequency, exercise duration, etc.) and causal chain information (such as family medical history, environmental factors, etc.) as attributes to construct a decision tree model to determine whether the service object has taken specific intervention behavior.

[0144] Furthermore, the embodiment of the present invention constructs a self-management scorer for the service object based on the intervention behavior, which can accurately present the health self-management level of each service object and provide targeted health education content for the service object. The self-management scorer refers to a tool for quantitatively evaluating the performance of the service object in self-health management. For example, the self-management scorer will take into account the intervention behavior of multiple dimensions such as the service object's diet control, exercise frequency, medication compliance, execution of regular physical examinations, and stress management ability.

[0145] As an embodiment of the present invention, constructing the self-management scorer of the service object based on the intervention behavior includes:

[0146] Analyzing changes in the health indicators of the service recipient based on the intervention behavior;

[0147] Collecting the behavior tracking data and self-management report of the service recipient, and identifying the self-management ability of the service recipient based on the behavior tracking data, the self-management report and changes in the health indicators;

[0148] Extracting the constituent elements of the self-management ability, and determining the scoring dimensions and scoring indicators of the service object based on the constituent elements;

[0149] Constructing a capability assessment matrix for the service object based on the scoring dimensions and the scoring indicators;

[0150] Analyze the behavioral triggering reasons of the intervention behavior and determine the behavioral duration of the intervention behavior;

[0151] Based on the behavior triggering cause, identifying the degree of initiative of the service object's behavior;

[0152] Calculating the service recipient's willingness score for the intervention behavior based on the triggering cause of the behavior, the duration of the behavior, and the initiative level of the behavior;

[0153] Setting an interpretation of the scoring result of the service object based on the capability assessment matrix and the willingness score;

[0154] A self-management scorer for the service object is constructed by combining the capability assessment matrix, the willingness score and the interpretation of the scoring result.

[0155] Among them, the changes in health indicators refer to the dynamic changes in various physiological and biochemical indicators of the service object before and after the implementation of the intervention behavior, such as the numerical changes of lung function indicators, immune-related indicators, etc., and the relief or aggravation of symptoms (such as pain level, dyspnea level, etc.); the behavioral tracking data refers to the health-related behavior information of the service object continuously collected with the help of wearable devices, mobile applications, sensors, medical record systems and other tools, such as sleep duration and quality data recorded by smart bracelets, number of visits and examination items recorded by medical systems, etc.; the self-management report refers to the subjective summary materials submitted by the service object in the form of text, questionnaires, interview records, etc. based on his or her own health management process, such as the experience of reading health information. Yes, the self-management ability refers to the comprehensive ability of an individual to independently carry out activities such as health knowledge learning, health behavior practice, health resource utilization, and health risk response in order to maintain and promote his or her own health. The constituent elements refer to the basic components of self-management ability, mainly including health knowledge reserves, health skill levels, behavior management capabilities, resource utilization capabilities, self-supervision and feedback capabilities, self-motivation and persistence capabilities, etc. The scoring dimensions refer to the different evaluation directions divided according to the constituent elements when conducting a quantitative evaluation of the self-management ability of the service recipients. The scoring indicators refer to the specific and measurable evaluation indicators under each scoring dimension. For example, under the health knowledge dimension, "correctness of disease etiology knowledge" and "awareness of healthy lifestyles" are set. and other indicators; under the health skills dimension, set indicators such as "blood pressure measurement operation standardization" and "insulin injection technology proficiency". The ability assessment matrix refers to a two-dimensional table constructed with scoring dimensions as rows and scoring indicators as columns. The specific scores or performance of the service objects on each scoring indicator can be filled in the corresponding cells to intuitively present their strengths and weaknesses in different ability dimensions. The behavior triggering reason refers to the internal motivation or external factor that prompts the service object to implement a specific intervention behavior. For example, the doctor's advice to the service object to continue taking the medicine can be regarded as an external reason, and the internal reason can be the service object's own health awareness improvement. The behavior duration refers to the length of time the service object continues to perform the behavior from the first implementation of the intervention behavior to the termination of the behavior. It can be counted in units such as days, weeks, and months. The degree of behavioral initiative refers to the degree of subjective initiative of the service object in the process of implementing the intervention behavior, reflecting whether its behavior is driven by its own will or relies on external supervision or pressure. Active behavior is manifested in independently formulating health plans, actively seeking health knowledge, and spontaneously adjusting behavioral strategies;Passive behaviors are often performed at the request of others. The willingness score is a numerical value used to quantify the client's subjective willingness to engage in intervention. The score interpretation is a detailed description of each score obtained by the client in the capability assessment matrix and the willingness score. For example, if the client's final self-management score is 70, the following explanation can be provided: The capability score is 35, which is mainly due to insufficient health knowledge and difficulties in accessing professional medical resources; the willingness score is 35, which indicates that the client is not proactive enough in implementing health management behaviors and the consistency of the behavior needs to be improved.

[0156] Optionally, based on the intervention behavior, the changes in the health indicators of the service object can be analyzed using a multivariate linear regression model. Based on the behavior tracking data, the self-management report and the changes in the health indicators, the self-management ability of the service object can be identified by establishing an evaluation index system, such as setting evaluation indicators for self-management ability, wherein the indicators include the stability of behavior, goal achievement, the degree of improvement of health indicators, and the effectiveness of self-management strategies. Corresponding weights are set for each indicator, and quantitative scores are performed based on actual data. Finally, a score that comprehensively reflects the self-management ability is comprehensively obtained. Based on the constituent elements, the scoring dimensions and scoring indicators of the service object can be determined using principal component analysis. Based on the ability assessment matrix and the willingness score, the interpretation of the service object's scoring results can be set through the SPSS Modeler evaluation tool.

[0157] In an optional embodiment of the present invention, the willingness score of the service subject in the intervention behavior is calculated using the following formula based on the triggering cause of the behavior, the duration of the behavior, and the initiative level of the behavior:

[0158]

[0159] Among them, R represents the willingness score of the service recipient in the intervention behavior, Indicates the weight of the triggering reason of the i-th behavior, which can be 0.2, Indicates the specific quantitative value of the triggering cause of the i-th behavior at time point T, which can be 5, where T represents the time point of the triggering cause of the behavior. represents the weight of the behavior duration, Indicates the duration of the behavior, The weight of the degree of initiative of the behavior, Indicates the degree of activeness of the behavior, which can be 0.8, n represents the total number of behavior triggering reasons, and i represents the index of the behavior triggering reason.

[0160] The embodiment of the present invention generates personalized education content for the service object based on the self-management scorer and the scenario perception layer, and can accurately provide dynamic health education content according to the self-management level of the service object and the actual life scenario in which the service object is located, thereby promoting the service object to carry out long-term health management. The personalized education content refers to customized health education information and suggestions generated based on the individual characteristics, health status, behavioral habits, psychological needs and specific scenarios of the service object. For example, for a service object with diabetes, the education content may include diabetes diet control principles, exercise precautions, blood sugar monitoring methods and prevention of complications.

[0161] As an embodiment of the present invention, generating personalized education content for the service object based on the self-management scorer and the scenario perception layer includes:

[0162] Collecting capability matrix data and real-time environment data of the service object according to the self-management scorer and the scenario perception layer;

[0163] Performing data fusion processing on the capability matrix data and the real-time environment data to generate a personalized data archive;

[0164] Based on the personalized data profile, identifying the health needs of the service recipient in different scenarios;

[0165] Creating a multimodal content representation of the service object according to the health needs and the different scenarios;

[0166] Extracting environmental characteristics of the service object based on the real-time environmental data;

[0167] According to the environmental characteristics, setting the missionary language style of the service object;

[0168] Constructing a mapping relationship between the environmental characteristics and the missionary language style, and defining an adaptive language style for the service object based on the mapping relationship;

[0169] The personalized data file, the multimodal content presentation form and the adaptive language style are combined to generate personalized education content for the service object.

[0170] Among them, the capability matrix data refers to the structured data generated by the quantitative assessment of the health management capability of the service object through the self-management scorer; the real-time environmental data refers to the dynamic data collected in real time by the scene perception layer, reflecting the environmental status of the service object, such as the current activity status of the service object and the situation of the surrounding personnel; the personalized data archive refers to the comprehensive data set for a specific service object generated after the capability matrix data and the real-time environmental data are fused through data cleaning, format conversion, feature extraction, etc.; the health needs refer to the specific needs of the service object for health management knowledge, skills or services in a specific scenario, identified by a data analysis algorithm (such as a data mining algorithm) based on the personalized data archive, such as the medication reminder needs of patients with hypertension in the home scenario; the multimodal content presentation refers to the use of text , images, audio, video, animation and other information carriers, a composite content presentation method generated by multimedia production technology according to the health needs and scene characteristics of the service objects, the environmental characteristics refer to the key information extracted from the real-time environmental data through the feature extraction algorithm, which can characterize the essential attributes of the environment in which the service objects are located, the educational language style refers to the set of language expression features used to present health education content, including vocabulary selection, sentence structure, tone and emotion, etc. For example, family scenes can correspond to a warm and friendly language style, and sports scenes can correspond to an energetic and encouraging language style, the mapping relationship refers to the correspondence model between environmental characteristics and educational language styles established by machine learning algorithms, and the adaptive language style refers to the educational language style that is automatically adjusted according to the real-time environmental characteristics of the service objects based on the mapping relationship.

[0171] Optionally, the mapping relationship between the environmental features and the missionary language style can be constructed using a neural network model.

[0172] S5. Based on the adaptive education timing, the personalized education content and the scenario type, a multi-scenario collaborative education platform is created for the service object; based on the multi-scenario collaborative education platform and the health status, the health empowerment level of the service object is identified; and based on the health empowerment level, an effect evaluation system of the multi-scenario collaborative education platform is generated.

[0173] The embodiment of the present invention creates a multi-scenario collaborative education platform for the service object based on the adaptive education timing, the personalized education content and the scenario type, thereby realizing highly customized, intelligent and dynamic health education services and improving the effectiveness and acceptance of health education. The multi-scenario collaborative education platform refers to an intelligent platform that can flexibly and accurately provide health education services according to the different scenarios of the service objects, as well as their personalized characteristics and needs.

[0174] As an embodiment of the present invention, the creating of the multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type includes:

[0175] Based on the adaptive education opportunity, collecting the interactive feedback data of the service object on the personalized education content, and collecting the real-time behavior data of the service object in the scenario type;

[0176] Analyze the knowledge mastery of the service recipients based on the interactive feedback data, and set the content push weight for the adaptive education opportunity;

[0177] Based on the knowledge mastery level and the content push weight, a dynamic collaborative education mode for the personalized education content in the scenario type is constructed;

[0178] Identifying the device usage status of the service object, and determining the interaction progress of the service object in the scenario type in combination with the device usage status, the knowledge mastery level and the real-time behavior data;

[0179] According to the interaction progress, a cross-scenario data synchronization mechanism of the service object is set;

[0180] Based on the cross-scenario data synchronization mechanism, construct a collaborative interaction interface for the service object under the scenario type;

[0181] By combining the dynamic collaborative education method, the cross-scenario data synchronization mechanism and the collaborative interaction interface, a multi-scenario collaborative education platform for the service object is created.

[0182] Among them, the interactive feedback data refers to the data generated by the service object through various interactive operations in the process of receiving personalized education content, such as the answer data of the service object in the health knowledge question and answer; the real-time behavior data refers to the behavioral information data generated in real time when the service object interacts with the education platform or related equipment under a specific scenario type, for example, the length and frequency of the service object browsing the education content; the knowledge mastery level refers to the quantitative or qualitative description of the knowledge mastered by the service object in the personalized education content; the content push weight refers to a value or level assigned to different parts or different types of personalized education content based on the interactive feedback data of the service object; the dynamic collaborative education method refers to a flexible organization of personalized education content based on the knowledge mastery level and content push weight of the service object. The method of combining, adjusting and pushing, the device usage status refers to the relevant operating information and usage of the device used by the service object to receive and interact with personalized education content, such as the device's battery level and network connection status; the interaction progress refers to the phased progress of the service object's interaction with personalized education content in a specific scenario type determined by comprehensively analyzing multiple aspects of information such as the service object's device usage status, knowledge mastery level, and real-time behavior data; the cross-scenario data synchronization mechanism refers to a technical means for ensuring that when the service object uses the education platform in different scenarios, its relevant data can be accurately and timely transmitted, updated and shared between various scenarios; the collaborative interaction interface refers to a user interface designed for the service object to interact with the multi-scenario collaborative education platform in a specific scenario type.

[0183] Optionally, based on the interaction feedback data, the knowledge mastery level of the service object can be analyzed using a comparative analysis method. Combined with the device usage status, the knowledge mastery level and the real-time behavior data, the interaction progress of the service object under the scenario type can be determined by an interaction progress evaluation model, for example, an interaction progress evaluation model established using a random forest algorithm.

[0184] Furthermore, the embodiments of the present invention can improve the service objects' participation and acceptance of health education and enhance their autonomy and ability in health management by identifying the health empowerment level of the service objects based on the multi-scenario collaborative education platform and the health status, thereby better achieving the goals of health education and improving the overall health level of the service objects. The health empowerment level refers to the ability and degree to which the service objects can understand their own health status, acquire health knowledge and resources, make health decisions and adopt positive health behaviors.

[0185] As an embodiment of the present invention, identifying the health empowerment level of the service object based on the multi-scenario collaborative education platform and the health status includes:

[0186] Outputting multi-scenario health education data of the multi-scenario collaborative education platform, and extracting health empowerment features from the multi-scenario health education data;

[0187] Identify the information receiving preferences and information conversion rates of the service recipients on the multi-scenario collaborative education platform;

[0188] Determining the information interpretation level of the multi-scenario collaborative education platform based on the information reception preference and the information conversion rate;

[0189] Analyzing the type and urgency of the needs of the service recipient based on the health status;

[0190] Collecting psychological state data of the service recipient according to the type of demand and the urgency of the demand;

[0191] Based on the psychological state data, identifying the psychological acceptance degree of the service object to the multi-scenario health education data;

[0192] The health empowerment level of the service recipient is identified by combining the health empowerment characteristics, the information interpretation level and the psychological acceptance level.

[0193] Among them, the multi-scenario health education data refers to the structured and unstructured data sets collected through the multi-scenario collaborative education platform for the dissemination of health knowledge and behavioral intervention, such as the health course videos, graphic materials and text materials released by the platform. The health empowerment characteristics refer to the quantitative indicators extracted from the multi-scenario health education data that can reflect the user's health self-management ability, such as the degree of knowledge mastery. The information reception preference refers to the user's tendency to choose different types of health information in the multi-scenario collaborative education platform, such as the health topics that users visit frequently. The information conversion rate refers to the measurement of the user's conversion of the received health education information into actual health behavior. The information interpretation level refers to the degree to which the multi-scenario collaborative education platform interprets, analyzes and understands the service recipients' data; the demand type refers to the different demands of the service recipients based on their own health status, such as the service recipients' demand for relevant knowledge and treatment methods of their own diseases; the urgency of the demand refers to the time urgency for the service recipients to obtain health education content; the psychological state data refers to data that can reflect the user's psychological state, such as the service recipients' health management self-efficacy score; and the psychological acceptance level refers to the service recipients' subjective feelings of closeness or alienation from the health education content.

[0194] Optionally, the extraction of health empowerment features from the multi-scenario health education data can be achieved using the chi-square test method, such as extracting key features from the multi-scenario health education data, and then calculating the correlation between each key feature and the health empowerment label (such as the degree of change in the user's health behavior, the level of health knowledge mastery, etc.) to obtain health empowerment features. The information conversion rate of the service object in the multi-scenario collaborative education platform can be identified by the ratio of the sum of the number of times the knowledge assessment is passed and the number of times the health behavior is changed to the total number of times information is received. Based on the psychological state data, the psychological acceptance of the service object to the multi-scenario health education data can be identified by the behavioral state of the service object, such as whether the service object's behavior in diet, exercise, work and rest has changed in the direction of health education.

[0195] The embodiment of the present invention generates an effect evaluation system for the multi-scenario collaborative education platform based on the health empowerment level, which can help optimize the education strategy of the education platform and improve the pertinence and effectiveness of the education platform. The effect evaluation system refers to a system for evaluating the education effect and impact of the multi-scenario collaborative education platform.

[0196] As an embodiment of the present invention, generating an effect evaluation system of the multi-scenario collaborative education platform based on the health empowerment level includes:

[0197] Extracting evaluation dimensions and evaluation indicators corresponding to the health empowerment level;

[0198] Identifying the project factors under the evaluation dimension, and setting the measurement method corresponding to the project factors according to the evaluation dimension and the evaluation indicator;

[0199] Outputting the measurement results of the project factors using the measurement method, and identifying the result correlation between the measurement results;

[0200] Determining the convergent validity of the multi-scenario collaborative education platform on the health empowerment level based on the measurement method and the result correlation;

[0201] Calculating the consistency level between the project factors according to the measurement results;

[0202] Based on the consistency level, generating a composite reliability of the multi-scenario collaborative education platform for the health empowerment level;

[0203] Combining the convergent validity and the composite reliability, an effectiveness evaluation system of the multi-scenario collaborative education platform is generated.

[0204] Among them, the evaluation dimension refers to the different angles or aspects of evaluating the level of health empowerment, and the evaluation indicator refers to the attribute parameter specifically used to measure each evaluation dimension. For example, in the knowledge dimension, the evaluation indicator can be the score of the health knowledge test; in the behavior dimension, the evaluation indicator can be the frequency of occurrence of a certain health behavior. The measurement method refers to the specific method and means of obtaining project factors, for example, through questionnaires to understand the user's health knowledge mastery, and through practical operation assessments to evaluate the user's health skill level. The project factor refers to a more specific subdivision factor under the evaluation dimension. For example, in the knowledge dimension, the project factor can be disease prevention knowledge, nutritional knowledge, The measurement results refer to the data or information obtained after measuring the project factors using a specific measurement method. The result correlation refers to the degree of association between the measurement results of different project factors. For example, if there is a high positive correlation between the disease prevention knowledge score and the nutrition knowledge score, it means that the measurement results of these two project factors have strong consistency. The convergent validity refers to the consistency of the measurement results of the health empowerment levels of different project factors responded by different measurement methods. The consistency level refers to the degree of coordination of project factors in measuring health empowerment levels. The composite reliability refers to a reliability indicator generated based on the consistency level between project factors within the measurement method.

[0205] Optionally, the correlation between the measurement results can be identified using the Pearson correlation coefficient method. According to the evaluation dimensions and the evaluation indicators, the measurement method corresponding to the project factors can be set using the properties of the evaluation dimensions and evaluation indicators. For example, for knowledge dimensions and indicators, a questionnaire survey method can be used, and for skill dimensions and indicators, it can be implemented using actual operation assessment methods.

[0206] S6. Combine the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system to perform health education processing for the service object and obtain health education results.

[0207] The embodiment of the present invention combines the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system to perform health education processing on the service object and obtain health education results. It can perceive the service object's environment, physical condition, behavioral habits and other multi-dimensional information in real time, provide the service object with accurate and personalized health education content, realize cross-scene and cross-platform education, help the service object improve health literacy and self-management ability, and optimize the resource allocation of health education to maximize the utilization of health education resources. The health education result refers to the result obtained after health education processing of the service object by combining the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system, such as the service object's attention to a healthy lifestyle has been improved.

[0208] Compared with the problems described in the background technology, the embodiment of the present invention creates a scene perception layer for the service object based on the environmental data and the target health data set, and can formulate a highly personalized health education plan for the unique situation of each service object, so that the education content is more in line with the actual needs of the service object, improve the accuracy and effectiveness of the education, and at the same time, can dynamically adjust the education content and method as the health status and environment of the service object change, to ensure that the education always matches the current situation of the service object, and realize continuous and dynamic health management support; further, the embodiment of the present invention generates the causal relationship in the target health data set by using a preset optimized Bayesian network algorithm Chain, can accurately find out the key factors that have a greater impact on health status, and avoid wasting energy on irrelevant factors. On the other hand, when the service objects clearly understand the causal relationship between their health status and certain behaviors or factors, they will be more willing to accept and actively cooperate with the education suggestions, thereby improving the effect of health education; the embodiment of the present invention determines the adaptive education timing of the service object based on the health status and the scenario type, and can carry out education when the service object needs relevant health knowledge most, thereby improving the effect of education. At the same time, by accurately grasping the education timing, the health education system can concentrate limited resources (such as manpower, material resources, time, etc.) where they are most needed, avoiding the waste of resources. Waste; further, the embodiment of the present invention generates personalized education content for the service object based on the self-management scorer and the scenario perception layer, and can accurately provide dynamic health education content according to the self-management level of the service object and the actual life scenario in which the service object is located, thereby promoting the service object to carry out long-term health management; the embodiment of the present invention creates a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content and the scenario type, thereby realizing highly customized, intelligent and dynamic health education services and improving the effectiveness and acceptance of health education; further, the embodiment of the present invention generates based on the health empowerment level The effect evaluation system of the multi-scenario collaborative education platform can help optimize the education strategy of the education platform and improve the pertinence and effectiveness of the education platform; finally, the embodiment of the present invention performs the health education processing of the service object by combining the scene perception layer, the multi-scenario collaborative education platform and the effect evaluation system to obtain the health education results. It can perceive the environment, physical condition, behavioral habits and other multi-dimensional information of the service object in real time, provide the service object with accurate and personalized health education content, realize cross-scenario and cross-platform education, help the service object improve health literacy and self-management ability, and optimize the resource allocation of health education to maximize the utilization of health education resources. Therefore, the health education method and system based on the intelligent system provided by the embodiment of the present invention can realize multi-scenario collaborative education and provide targeted health education content.

[0209] Example 2 like Figure 2 FIG. 1 is a functional module diagram of a health education system based on an intelligent system according to the present invention.

[0210] The health education system 200 based on an intelligent system described in the present invention can be installed in an electronic device. According to the functions implemented, the health education system based on an intelligent system can include a data acquisition module 201, a scene classification module 202, an education timing selection module 203, a personalized output module 204, a multi-scene collaboration module 205, and a result output module 206. The module described in the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can perform fixed functions, which are stored in the memory of the electronic device.

[0211] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0212] The data acquisition module 201 is used to obtain service objects for health education, collect environmental data and multi-dimensional health data of the service objects, and perform fuzzy processing on the multi-dimensional health data to obtain a target health data set;

[0213] The scene classification module 202 is used to create a scene perception layer for the service object based on the environmental data and the target health data set, and identify the scene type of the service object according to the scene perception layer;

[0214] The education timing selection module 203 is configured to generate a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyze the health status of the service object according to the causal relationship chain, and determine the adaptive education timing of the service object based on the health status and the scenario type;

[0215] The personalized output module 204 is configured to identify the intervention behavior of the service recipient based on the health status and the causal relationship chain, construct a self-management scorer for the service recipient based on the intervention behavior, and generate personalized education content for the service recipient based on the self-management scorer and the scenario perception layer;

[0216] The multi-scenario collaboration module 205 is configured to create a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type; identify the health empowerment level of the service object based on the multi-scenario collaborative education platform and the health status; and generate an effect evaluation system for the multi-scenario collaborative education platform based on the health empowerment level;

[0217] The result output module 206 is used to combine the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system to perform health education processing on the service object and obtain health education results.

[0218] In detail, the modules in the intelligent system-based health education system 200 described in the embodiment of the present invention are used in the same manner as above. Figure 1 The technical means are the same as the health education method based on intelligent system described in , and can produce the same technical effects, so I will not go into details here.

[0219] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0220] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A health education method based on an intelligent system, characterized in that: The method comprises: Acquire a service target for health education, collect environmental data and multi-dimensional health data of the service target, perform fuzzy processing on the multi-dimensional health data, and obtain a target health data set; Creating a scene perception layer for the service object based on the environmental data and the target health data set, and identifying the scene type of the service object according to the scene perception layer; generating a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyzing the health status of the service recipient based on the causal relationship chain, and determining an adaptive education opportunity for the service recipient based on the health status and the scenario type; Identifying the intervention behavior of the service recipient based on the health status and the causal chain, constructing a self-management scorer for the service recipient based on the intervention behavior, and generating personalized education content for the service recipient based on the self-management scorer and the scenario perception layer; Based on the adaptive education timing, the personalized education content, and the scenario type, a multi-scenario collaborative education platform is created for the service object; based on the multi-scenario collaborative education platform and the health status, the health empowerment level of the service object is identified; and based on the health empowerment level, an effect evaluation system of the multi-scenario collaborative education platform is generated; In combination with the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system, health education processing of the service object is performed to obtain health education results.

2. A health education method based on an intelligent system as claimed in claim 1, characterized in that: The step of creating a scene perception layer for the service object based on the environmental data and the target health data set includes: Formatting the environmental data and the target health data set to obtain formatted data; extracting key features from the environmental data and the target health data set based on the formatted data; identifying scene semantics of the environmental data and the target health data set based on the key features; Based on the scene semantics, generating a scene graph of the service object using the environmental data and the target health data set; Calculating the similarity between the scene graphs, and extracting similar semantic labels of the scene graphs based on the similarity; Based on the similar semantic tags, setting a scene graph index corresponding to the scene semantics; The scene perception layer of the service object is created by combining the scene semantics, the scene graph and the scene graph index.

3. The health education method based on an intelligent system according to claim 1, characterized in that: The generating of the causal relationship chain in the target health data set by using a preset optimized Bayesian network algorithm includes: performing semantic analysis on the target health data set to extract potential semantic relationships of the target health data set; Determining key parameter variables of the target health data set according to the potential semantic relationship; Calculating the mutual information factors between the key parameter variables, and constructing an initial Bayesian network structure of the key parameter variables based on the mutual information factors; Setting probability distribution parameters of the initial Bayesian network structure, and defining a particle encoding method for the initial Bayesian network structure and the probability distribution parameters; According to the particle encoding method, the initial Bayesian network structure and the probability distribution parameters are converted into particles; Identifying a scoring function of the initial Bayesian network structure, and setting a fitness function of the particle based on the scoring function; Analyzing the causal relationship of the key parameter variables using the optimized Bayesian network algorithm according to the fitness function; Based on the causal relationship, a causal relationship chain in the target health data set is generated.

4. The health education method based on an intelligent system according to claim 1, characterized in that: The determining of the adaptive education opportunity for the service object based on the health status and the scenario type includes: Extracting scene features corresponding to the scene types, and identifying conversion situations between the scene types based on the scene features; Based on the conversion situation, determining the scene transition period of the service object, and extracting transition period features of the service object during the scene transition period; According to the characteristics of the transition period, set up a cross-scenario education method for the service object; Based on the health status, calculating the personalized timing decision weights of the service object at different time points; The adaptive teaching timing of the service object is determined by combining the cross-scenario teaching method and the personalized timing decision weight.

5. The health education method based on an intelligent system according to claim 1, characterized in that: The step of constructing a self-management scorer for the service recipient based on the intervention behavior includes: Analyzing changes in the health indicators of the service recipient based on the intervention behavior; Collecting the behavior tracking data and self-management report of the service recipient, and identifying the self-management ability of the service recipient based on the behavior tracking data, the self-management report and changes in the health indicators; Extracting the constituent elements of the self-management ability, and determining the scoring dimensions and scoring indicators of the service object based on the constituent elements; Constructing a capability assessment matrix for the service object based on the scoring dimensions and the scoring indicators; Analyze the behavioral triggering reasons of the intervention behavior and determine the behavioral duration of the intervention behavior; Based on the behavior triggering cause, identifying the degree of behavioral initiative of the service object; Calculating the service recipient's willingness score for the intervention behavior based on the triggering cause of the behavior, the duration of the behavior, and the initiative level of the behavior; Setting an interpretation of the scoring result of the service object based on the capability assessment matrix and the willingness score; Combining the capability assessment matrix, the willingness score and the interpretation of the scoring result, a self-management scorer for the service object is constructed.

6. The health education method based on an intelligent system according to claim 1, characterized in that: Generating personalized education content for the service object based on the self-management scorer and the scenario perception layer includes: Collecting capability matrix data and real-time environment data of the service object according to the self-management scorer and the scenario perception layer; Performing data fusion processing on the capability matrix data and the real-time environment data to generate a personalized data archive; Based on the personalized data profile, identifying the health needs of the service recipient in different scenarios; Creating a multimodal content representation of the service object according to the health needs and the different scenarios; Extracting environmental characteristics of the service object based on the real-time environmental data; According to the environmental characteristics, setting the missionary language style of the service object; Constructing a mapping relationship between the environmental characteristics and the missionary language style, and defining an adaptive language style for the service object based on the mapping relationship; The personalized data file, the multimodal content presentation form and the adaptive language style are combined to generate personalized education content for the service object.

7. The health education method based on an intelligent system according to claim 1, characterized in that: The step of creating a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type includes: Based on the adaptive education opportunity, collecting the interactive feedback data of the service object on the personalized education content, and collecting the real-time behavior data of the service object in the scenario type; Analyze the knowledge mastery of the service recipients based on the interactive feedback data, and set the content push weight for the adaptive education opportunity; Based on the knowledge mastery level and the content push weight, a dynamic collaborative education mode for the personalized education content in the scenario type is constructed; Identifying the device usage status of the service object, and determining the interaction progress of the service object in the scenario type in combination with the device usage status, the knowledge mastery level and the real-time behavior data; According to the interaction progress, a cross-scenario data synchronization mechanism of the service object is set; Based on the cross-scenario data synchronization mechanism, construct a collaborative interaction interface for the service object under the scenario type; By combining the dynamic collaborative education method, the cross-scenario data synchronization mechanism and the collaborative interaction interface, a multi-scenario collaborative education platform for the service object is created.

8. The health education method based on an intelligent system according to claim 1, characterized in that: The identifying the health empowerment level of the service object according to the multi-scenario collaborative education platform and the health status includes: Outputting multi-scenario health education data of the multi-scenario collaborative education platform, and extracting health empowerment features from the multi-scenario health education data; Identify the information receiving preferences and information conversion rates of the service recipients on the multi-scenario collaborative education platform; Determining the information interpretation level of the multi-scenario collaborative education platform based on the information reception preference and the information conversion rate; Analyzing the type and urgency of the needs of the service recipient based on the health status; Collecting psychological state data of the service recipient according to the type of demand and the urgency of the demand; Based on the psychological state data, identifying the psychological acceptance degree of the service object to the multi-scenario health education data; The health empowerment level of the service recipient is identified by combining the health empowerment characteristics, the information interpretation level and the psychological acceptance level.

9. The health education method based on an intelligent system according to claim 1, characterized in that: The effect evaluation system of the multi-scenario collaborative education platform is generated based on the health empowerment level, including: Extracting evaluation dimensions and evaluation indicators corresponding to the health empowerment level; Identifying the project factors under the evaluation dimension, and setting the measurement method corresponding to the project factors according to the evaluation dimension and the evaluation indicator; Outputting the measurement results of the project factors using the measurement method, and identifying the result correlation between the measurement results; Determining the convergent validity of the multi-scenario collaborative education platform on the health empowerment level based on the measurement method and the result correlation; Calculating the consistency level between the project factors according to the measurement results; Based on the consistency level, generating a composite reliability of the multi-scenario collaborative education platform for the health empowerment level; Combining the convergent validity and the composite reliability, an effectiveness evaluation system of the multi-scenario collaborative education platform is generated.

10. A health education system based on an intelligent system, characterized in that: The system comprises: A data acquisition module is used to obtain service objects for health education, collect environmental data and multi-dimensional health data of the service objects, and perform fuzzy processing on the multi-dimensional health data to obtain a target health data set; A scene classification module is used to create a scene perception layer for the service object based on the environmental data and the target health data set, and identify the scene type of the service object according to the scene perception layer; a teaching and education timing selection module, configured to generate a causal relationship chain in the target health data set using a preset optimized Bayesian network algorithm, analyze the health status of the service object according to the causal relationship chain, and determine the adaptive teaching and education timing of the service object based on the health status and the scenario type; a personalized output module, configured to identify the intervention behavior of the service recipient based on the health status and the causal chain, construct a self-management scorer for the service recipient based on the intervention behavior, and generate personalized education content for the service recipient based on the self-management scorer and the scenario perception layer; a multi-scenario collaborative module, configured to create a multi-scenario collaborative education platform for the service object based on the adaptive education opportunity, the personalized education content, and the scenario type; identify the health empowerment level of the service object based on the multi-scenario collaborative education platform and the health status; and generate an effect evaluation system for the multi-scenario collaborative education platform based on the health empowerment level; The result output module is used to combine the scene perception layer, the multi-scene collaborative education platform and the effect evaluation system to perform health education processing on the service object and obtain health education results.

Images