Health behavior intervention and incentive method based on social network analysis

Abstract

The invention relates to the technical field of digital health management, in particular to a health behavior intervention and incentive method based on social network analysis, which comprises the following steps: acquiring health behavior data, social network data and interaction data of a user through a social platform and a health management application, preprocessing the acquired data, ensuring the accuracy and consistency of the data, and improving the user experience. Modeling by adopting a social network analysis technology, extracting social nodes and structural features, identifying a user behavior mode through clustering analysis, and evaluating health risks; on the basis of the analysis results, a personalized intervention strategy is generated, and through multi-dimensional incentive mechanisms such as material reward, social reward and psychological incentive, the user is pushed to continuously keep good health behaviors; according to the invention, accurate and long-term health management can be realized, and the problems of low efficiency and singleness in a traditional health intervention scheme are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital health management, and in particular to a health behavior intervention and motivation method based on social network analysis. Background Art

[0002] With the rapid development of information technology, digital health management has been widely used around the world, especially through social networking platforms and health management applications, users can track and manage their health status in real time. These applications not only provide health data monitoring and analysis functions, but also promote communication and motivation between users through social interaction, and promote the formation of healthy behaviors.

[0003] Although existing health management applications can provide personalized health advice to a certain extent, most of these advice is based on simple health data and preset models, lacking a deep understanding of user behavior and consideration of the social environment. At present, most technical solutions fail to fully integrate users' social network data, ignoring the potential impact of social relationships and social interactions on health behaviors. Factors such as interaction, influence, and information dissemination paths between users in social networks have not been fully utilized. In addition, existing health management systems also have limitations in terms of incentive mechanisms, mainly relying on simple points or rewards, and lacking multi-dimensional, personalized incentive mechanisms to maintain long-term user participation and behavior change. Summary of the invention

[0004] The present invention provides a health behavior intervention and motivation method based on social network analysis to solve the technical problems mentioned in the background technology section;

[0005] The health behavior intervention and motivation method based on social network analysis includes the following steps:

[0006] S1, data collection: by collecting users’ health behavior data, social network data, and interaction data on multiple social platforms and health management applications;

[0007] The health behavior data includes steps, diet, exercise frequency and rest time;

[0008] The social network data includes social relationship graph and social interaction frequency;

[0009] The interaction data includes emotional connections and recommendation behaviors between users;

[0010] S2, data preprocessing: cleaning, deduplication and standardization of the collected health behavior data, social network data and interaction data;

[0011] S3, social network analysis: Based on the preprocessed data, social network analysis technology is used to model the user's social network and extract key social nodes and structural features in the network, including social group structure, social influence index and information dissemination path;

[0012] S4, user behavior pattern identification: based on the social network analysis results, combined with health behavior data, a machine learning algorithm is used to identify the user's health behavior pattern, generate the user's health behavior feature profile, and identify potential health risk behaviors and health management goals;

[0013] S5, personalized intervention strategy generation: formulate personalized health behavior intervention strategies based on the user's health behavior profile and social relationships in the social network;

[0014] S6, Design of healthy behavior incentive mechanism: Design a healthy behavior incentive mechanism to encourage users to continue to maintain good health behaviors.

[0015] Optionally, the S1 includes:

[0016] S11, multi-platform data collection channels: real-time collection of users’ health behavior data, social network data, and interaction data on multiple social platforms (such as Weibo and WeChat) and health management applications (such as Keep and Xiaomi Sports Health);

[0017] S12, health behavior data collection: collect the user's health behavior data through the user's smart device, including steps, diet, exercise frequency and rest time;

[0018] S13, social network data collection: collect users’ social network data through social platform API interfaces, including social relationship graphs and social interaction frequencies;

[0019] S14, interactive data collection: by collecting users’ interactive behaviors in real time on social platforms and health management applications, capturing the interactive data between users and other social network members.

[0020] Optionally, S2 includes:

[0021] S21, data cleaning: clean the collected health behavior data, social network data and interaction data, and delete incomplete, missing or erroneous data records;

[0022] S22, data deduplication: deduplication of collected health behavior data, social network data and interaction data;

[0023] S23, data standardization: standardize health behavior data, social network data, and interaction data;

[0024] S24, time series data alignment: for health behavior data, social network data and interaction data collected across time periods, time series data alignment is performed;

[0025] S25, data conversion and feature construction: perform feature construction on the cleaned and standardized data.

[0026] Optionally, the S3 includes:

[0027] S31, Construction of social network modeling framework: Based on the preprocessed data, the user's social network model is constructed using graph theory methods;

[0028] S32, Topological structure analysis of social networks: Use social network analysis technology to perform topological structure analysis on the constructed social network model and extract the basic structural features of the social network.

[0029] Optionally, S3 further includes:

[0030] S33, social influence index calculation: through social network analysis, calculate the social influence index of each user to measure the user's communication ability in the social network;

[0031] S34, Information propagation path analysis: Based on social network analysis, identify the information propagation path in the network, determine the key nodes and efficient propagation paths of information propagation.

[0032] Optionally, the S4 includes:

[0033] S41, data fusion and feature extraction: fuse social network analysis data and health behavior data, and extract features related to health behavior identification;

[0034] S42, model construction: Based on the extracted features, a health behavior pattern recognition model is constructed using the K-means clustering algorithm to predict the user's behavior pattern and assess their health risks;

[0035] S43, health behavior characteristic profile generation: through the health behavior pattern recognition model, a health behavior characteristic profile is generated for each user, and health risk assessment is performed based on the results of behavior pattern recognition to provide a basis for personalized intervention;

[0036] S44, Health risk identification and health management goal setting: Based on the health behavior characteristic profile, identify the user's health risks and set health management goals.

[0037] Optionally, the S5 includes:

[0038] S51, health behavior profile analysis: Based on the generated health behavior profile, identify the user's strengths and potential risks in health management and develop personalized intervention strategies for the user;

[0039] S52, Social Relationship Analysis: Identify the user’s social support system and social influence by analyzing the user’s relationships and interaction patterns in social networks;

[0040] S53, intervention goal setting: setting personalized health behavior improvement goals based on the user's health behavior profile and social relationship analysis;

[0041] S54, Personalized intervention strategy development: Design intervention strategies based on health behavior improvement goals and social goals to help users achieve their health management goals.

[0042] Optionally, the S6 includes:

[0043] S61, Incentive Mechanism Goal Setting: Clarify the goals of the healthy behavior incentive mechanism and ensure that the incentive program meets the health needs of users;

[0044] S62, Incentive Reward Design: Design specific reward mechanisms to encourage users to continue to participate and maintain healthy behaviors;

[0045] S63, incentive triggering condition setting: setting incentive triggering conditions to ensure the timeliness and effectiveness of incentive measures.

[0046] Optionally, S6 further includes:

[0047] S64, Incentive feedback mechanism: Design an incentive feedback mechanism to enable users to perceive the results of healthy behaviors and maintain motivation;

[0048] S65, Incentive mechanism optimization: Continuously adjust and optimize the incentive mechanism based on user behavior feedback.

[0049] Beneficial effects of the present invention:

[0050] The present invention can accurately identify the health behavior characteristics of users and formulate personalized intervention strategies by comprehensively analyzing the health behaviors, social networks and interaction data of users. This personalized intervention helps to formulate actionable health goals based on the health needs of different users, and encourages users to form and maintain long-term healthy behaviors through social network influence and social incentives, thereby significantly improving the effectiveness and sustainability of health management.

[0051] The present invention uses social network analysis, machine learning and data preprocessing technology to efficiently process and analyze a large amount of health behavior data, social network data and interaction data. Through steps such as data cleaning, deduplication, standardization and cluster analysis, while ensuring data quality, potential health risk behaviors can be accurately identified through model building and feature extraction. The application of these technologies enables intervention strategies to be based on scientific analysis and data support, and to intervene in the health behavior of each user more accurately, avoiding the extensiveness and inefficiency of previous single intervention methods.

[0052] The present invention, by designing a multi-dimensional incentive mechanism, including material rewards, a points system, social rewards, and psychological incentives, can satisfy different needs of users while motivating them to engage in healthy behaviors, thereby maintaining their continued motivation to participate. In particular, with the support of social networks, users can further enhance the intrinsic driving force of their behaviors by participating in social interactions, completing health challenges, and obtaining social honors. The dynamic adjustment and personalized recommendation of the incentive mechanism can be continuously optimized based on the progress and feedback of users, thereby ensuring the long-term effectiveness and adaptability of the incentive measures, and promoting the long-term maintenance and continuous improvement of healthy behaviors. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A schematic diagram of a method flow of an embodiment of the present invention;

[0055] Figure 2 Schematic diagram of S4 process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0056] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0057] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0058] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0059] like Figure 1-Figure 2 As shown, the health behavior intervention and motivation method based on social network analysis includes the following steps:

[0060] S1, data collection: by collecting users’ health behavior data, social network data, and interaction data on multiple social platforms and health management applications;

[0061] Health behavior data include steps, diet, exercise frequency, and sleep schedule;

[0062] Social network data includes social relationship graphs and social interaction frequency;

[0063] Interaction data includes emotional connections and recommendation behaviors between users;

[0064] S2, data preprocessing: cleaning, deduplication and standardization of the collected health behavior data, social network data and interaction data to ensure data consistency and accuracy;

[0065] S3, social network analysis: Based on the preprocessed data, social network analysis technology is used to model the user's social network and extract key social nodes and structural features in the network, including social group structure, social influence index and information dissemination path;

[0066] S4, user behavior pattern recognition: Based on the results of social network analysis and combined with health behavior data, machine learning algorithms are used to identify users' health behavior patterns, generate health behavior feature profiles of users, and identify potential health risk behaviors and health management goals;

[0067] S5, personalized intervention strategy generation: formulate personalized health behavior intervention strategies based on the user's health behavior profile and social relationships in the social network;

[0068] S6, Design of healthy behavior incentive mechanism: Design a healthy behavior incentive mechanism to encourage users to continue to maintain good health behaviors.

[0069] S1 includes:

[0070] S11, multi-platform data collection channels: real-time collection of users’ health behavior data, social network data, and interaction data on multiple social platforms (such as Weibo and WeChat) and health management applications (such as Keep and Xiaomi Sports Health). The interfaces include API interfaces, data capture tools, and data sharing services provided by third-party platforms;

[0071] S12, health behavior data collection: collect the user's health behavior data through the user's smart device (such as Xiaomi bracelet, smart watch or mobile phone APP), including the number of steps, diet, exercise frequency and rest time, among which;

[0072] Step data: Collect the user's daily step data through smart devices (such as Xiaomi wristband) or WeChat step count;

[0073] Dietary data: Users record their daily diet through health management applications;

[0074] Exercise frequency: Use the motion sensors of smart devices to record the user's exercise type, exercise duration and frequency, and monitor the user's exercise habits;

[0075] Sleep and work schedule: Use sleep monitoring devices or health management applications to record the user's sleep time, work and work schedule in real time, including the time to fall asleep, the time to wake up, and the quality of sleep;

[0076] S13, social network data collection: collect users’ social network data through social platform API interfaces, including social relationship graphs and social interaction frequencies, among which;

[0077] The social relationship map is generated by calculating the connection strength between users. Each user is a node. The relationship between nodes is determined by the frequency of interaction, interaction type, etc. The social relationship map can be drawn by analyzing the social network structure of users such as friends and followers.

[0078] Social interaction frequency: By analyzing the frequency of interaction between users and other social network members, the user's social activity is obtained. The interaction frequency includes the number and intensity of social behaviors such as likes, comments, shares, forwarding, and private messages. When collecting data, user privacy protection must be considered;

[0079] S14, interactive data collection: by collecting users' interactive behaviors in real time on social platforms and health management applications, capturing the interactive data between users and other social network members. The interactive data includes users' likes, comments, shares, forwarding, private messages, and following, etc.

[0080] Interaction frequency: counts the number of interactions between users and other social network members, including the frequency of likes, comments, private messages, etc.

[0081] Interaction type: Quantify the user's social activity based on different interactive behaviors (such as likes, comments, private messages, etc.) and record the user's social activity level;

[0082] User social influence: By analyzing interaction data, the influence index of each user on the social platform is calculated and evaluated based on factors such as interaction frequency, interaction breadth, and recommendation behavior;

[0083] Emotional connection analysis: Based on the social network data and interaction data collected in S13 and S14, the sentiment analysis algorithm is used to classify the sentiment of social interactions between users and identify the emotional connection between users. The sentiment analysis algorithm analyzes the sentiment of users' comments and private messages on social platforms and determines whether they are positive, negative or neutral, in order to quantify the social intimacy and interaction quality between users. These analysis results help identify users' influence in social networks, interaction intensity, and potential social cooperation possibilities;

[0084] Recommendation behavior analysis: By analyzing the recommendation behavior of users on social platforms and health management applications, we can identify the recommended content and its dissemination effect. Recommendation behavior includes users actively sharing health activities, diet suggestions, exercise plans, etc., as well as social dissemination activities in which users participate (such as forwarding, commenting, etc.). We analyze the frequency, dissemination breadth, influence, and audience feedback of these behaviors to identify users and social circles with high influence in social networks, and then conduct quantitative evaluation of users' social influence;

[0085] By collecting users' health behavior data, social network data and their interaction data in real time through multiple channels, a comprehensive user behavior profile is constructed. These data provide sufficient support for subsequent social network analysis, user behavior pattern recognition and the generation of personalized intervention strategies, thereby improving the intervention effect more accurately.

[0086] S2 includes:

[0087] S21, data cleaning: clean the collected health behavior data, social network data and interaction data, and delete incomplete, missing or erroneous data records. The specific steps include:

[0088] Outlier detection: Use the box plot method to identify outliers in the data and remove or correct them. For data such as steps, diet, and exercise, set a reasonable threshold range (for example, the number of steps should be between 0 and 20,000, and the exercise frequency should be within a reasonable range), detect and delete data points that exceed the threshold;

[0089] Missing value processing: For missing data caused by equipment failure or user failure to record, interpolation is used to fill in the missing data to ensure that each user's health behavior data and social network data are complete;

[0090] Redundant data removal: By comparing data from multiple sources, duplicate records are removed to avoid information noise caused by data redundancy. The uniqueness of each data record is ensured through data identifiers (such as user ID, timestamp, etc.), and duplicate entries are deleted;

[0091] S22, data deduplication: deduplication of collected health behavior data, social network data and interaction data, the steps are as follows:

[0092] Deduplication based on user ID and timestamp: When merging the health behavior data of the same user from different platforms, the user ID and timestamp are used to ensure the uniqueness of each record. If there are duplicate records in the same time period (for example, the user's step data is uploaded on different platforms at the same time), only the most accurate or latest record is retained and the duplicate records are deleted;

[0093] Deduplication based on interactive behavior: For social network and interactive data, ensure that each interactive behavior of a user with other users (such as comments, likes, private messages, etc.) is only counted once to avoid counting the same behavior multiple times. Use behavior ID and user ID for deduplication.

[0094] S23, data standardization: To ensure the consistency of data from different platforms and devices, health behavior data, social network data and interaction data are standardized. The specific steps include:

[0095] Numerical data standardization: Health behavior data (such as number of steps, exercise frequency, work and rest time, etc.) are standardized. Z-Score standardization is used to convert all numerical data to the same dimension for subsequent analysis. The Z-Score standardization formula is expressed as:

[0096]

[0097] Among them, X is the original data, μ is the mean of the data, and σ is the standard deviation of the data;

[0098] Categorical data standardization: Categorical data standardization is performed on social network data (such as users' social relationship graphs, social interaction frequency, etc.), and One-Hot encoding is used to convert categorical data into numerical data.

[0099] One-Hot encoding: convert each category (such as the interaction type on different social platforms) into a binary vector form to ensure that each category item is treated as an independent variable;

[0100] S24, time series data alignment: For health behavior data, social network data, and interaction data collected across time periods, time series data alignment is performed to ensure that all data are aligned in a unified time format for subsequent analysis and modeling. The specific steps include:

[0101] Unified time format: Unify the time data collected by different platforms into a standard format (such as ISO 8601 format) to facilitate time matching;

[0102] Time window division: according to the analysis requirements, set a unified time window (such as daily, weekly, monthly, etc.) to align data from different time periods for subsequent behavior pattern recognition and social network analysis;

[0103] Time filling: For missing time points in time series data, fill in the missing data through interpolation or the average value of the previous and next time periods to maintain data consistency;

[0104] S25, data conversion and feature construction: construct features for the cleaned and standardized data for subsequent data analysis and model training. The specific steps include:

[0105] Health behavior feature construction: construct the user's health behavior feature vector based on data such as step count, exercise frequency, and eating habits. These feature vectors are used to describe the user's health behavior pattern, such as average daily step count, weekly exercise frequency, daily diet type, etc.

[0106] Social network feature construction: Based on social network data (such as social relationship graph, social interaction frequency, etc.), construct the user's social feature vector, including social activity, social influence, social circle size, etc.

[0107] Interaction data feature construction: By analyzing the user's interaction data, we construct interaction frequency features, interaction quality features (such as emotional connection analysis results), etc., which are used to describe the user's interaction behavior and emotional connection in the social network;

[0108] Through data cleaning, deduplication, standardization, time series alignment and feature construction, the collected health behavior data, social network data and interaction data are ensured to be consistent, accurate and available. These data preprocessing steps provide a reliable foundation for subsequent social network analysis, behavior pattern recognition and the generation of personalized intervention strategies, improving the overall system processing efficiency and intervention effect.

[0109] S3 includes:

[0110] S31, Construction of social network modeling framework: Based on the preprocessed data, a graph theory method is used to construct a user's social network model. The specific steps include:

[0111] Node definition: Each user is regarded as a node in the social network. The attributes of the node include the user's basic information (such as age, gender, region, etc.), health behavior data (such as number of steps, exercise frequency, etc.) and social interaction data (such as interaction frequency, interaction quality, etc.);

[0112] Edge definition: The edge in a social network represents the interactive relationship between users. The attributes of the edge include interaction frequency, interaction intensity, interaction emotion, etc., which can reflect the strength of the relationship between users. The existence and weight of the edge are determined by analyzing the interaction of users in the health management platform and the social platform (such as likes, comments, private messages, sharing, etc.);

[0113] Network structuring: Based on the edge connection relationship, the adjacency matrix or edge list is used to represent the social network. The operation is as follows:

[0114] Adjacency matrix: For each pair of nodes i and j, if there is a connection (i.e. interaction), the element value in the adjacency matrix is ​​the weight of the edge, otherwise it is zero;

[0115] Edge list: A social network is represented by a list of edges, which records the starting node, ending node and weight of each edge.

[0116] S32, topological structure analysis of social network: Use social network analysis technology to perform topological structure analysis on the constructed social network model and extract basic structural features in the social network. The specific steps include:

[0117] S321, Cluster Analysis: Identify the group structure in the social network through K-means clustering, divide the social network into multiple closely connected groups, and analyze the composition and behavior characteristics of each social group. The steps are as follows:

[0118] Data preparation: Use social network data represented by adjacency matrix or edge list as input to define the relationship strength (weight) between nodes;

[0119] Preliminary selection of the number of clusters k: Based on the characteristics of the user's social network, select an appropriate k value (the number of clusters) through the elbow rule or other statistical methods;

[0120] Cluster initialization: randomly select k initial center points to represent the centers of k social groups;

[0121] Iterative calculation: assign each node to the center node closest to it, and update the position of each center point at the same time, iterating until the center point no longer changes or the preset number of iterations is reached;

[0122] Clustering results: The social network nodes are divided into k social groups. By analyzing the health behavior characteristics and interaction patterns of these groups, the common characteristics of each social group are extracted;

[0123] The results of cluster analysis can help identify different groups in social networks and further analyze the interactions and influence between groups;

[0124] S322, degree centrality analysis: Calculate the degree centrality of each node, that is, the number of connections of the node, to evaluate the importance of each user in the social network. Nodes with higher degree centrality are core users in the social network and usually have a greater impact on information dissemination and social interaction. The steps are as follows:

[0125] Degree centrality calculation: Calculate the degree centrality of each user, that is, the number of edges of each node. For node i, the degree centrality calculation formula is:

[0126] C d (i) = deg(i) = ∑ j A ij ;

[0127] Where deg(i) is the degree of node i, A ij is the edge weight between node i and node j in the adjacency matrix;

[0128] Identification of core users: Nodes with higher degree centrality are core users in social networks. These users have greater influence in information dissemination and social interaction.

[0129] S323, Clustering coefficient analysis: The clustering coefficient is used to measure the closeness between nodes in a social network and reflect the aggregation of the network. The specific steps are as follows:

[0130] Local clustering coefficient calculation: For each node i, the local clustering coefficient C i The calculation formula is:

[0131]

[0132] Among them, e iis the actual number of edges of node i, k i is the degree of node i. The higher the local clustering coefficient, the tighter the social group of the node.

[0133] Global clustering coefficient: The global clustering coefficient is the average of the clustering coefficients of all nodes, reflecting the social aggregation of the entire network;

[0134]

[0135] Where N is the total number of nodes in the social network;

[0136] The analysis of cluster coefficient helps to identify the close connections between nodes in social networks and find user groups with strong interactive relationships;

[0137] Through social network modeling and topological structure analysis, an accurate social network model was established, and the network structure was analyzed in detail, which can clearly identify social groups, key nodes and interactive relationships between groups in the social network, providing a basis for subsequent analysis of social influence and information dissemination paths.

[0138] S3 also includes:

[0139] S33, social influence index calculation: Calculate the social influence index of each user through social network analysis to measure the user's communication ability in the social network. The specific steps include:

[0140] Definition of influence: The social influence index reflects the influence of a user on the behavior of other users. Influence can be comprehensively evaluated by factors such as the degree centrality of the node, the breadth of communication, and the activity of the user.

[0141] Influence index calculation method: Use PageRank algorithm to calculate the user's social influence index;

[0142] PageRank algorithm: It uses iterative calculation of the "importance" value of the node to reflect the influence of the user in the network. The PageRank calculation formula is:

[0143]

[0144] Where PR(A) represents the PageRank value of node A, M(A) is the set of nodes pointing to A, L(i) is the out-degree of node i, and d is the damping factor (usually 0.85);

[0145] S34, Information propagation path analysis: Based on social network analysis, identify the information propagation path in the network, determine the key nodes and efficient propagation paths of information propagation. The specific steps include:

[0146] S341, Shortest Path Analysis: In social networks, the propagation path of information from the source node to the target node is not unique, so it is necessary to identify the key path of information propagation by calculating the shortest path. The Dijkstra algorithm is used to calculate the shortest path. The specific steps are as follows:

[0147] Graph representation: each user in the social network is regarded as a node (V) in the graph, and the interaction relationship (edge) between users is regarded as an edge (E) in the graph. The weight of the edge represents the interaction intensity, influence or other related attributes between users;

[0148] Assume that the social network consists of a node set V = {v1,v2,...,v n} and edge set E = {(v i ,v j ,w ij )}, where w ij Represents node v i and node v j The weights between them (interaction intensity, social distance, etc.);

[0149] Implementation of Dijkstra's algorithm: Select a source node (user) v s As the starting point of information propagation, the distance values ​​of all nodes are initialized to infinity (indicating that they are not calculated), and the distance of the source node is set to 0;

[0150] For each node, calculate the shortest path through the adjacent nodes and update the shortest distance of the node. The specific calculation formula is:

[0151] d(v j )=min(d(v j ),d(v i )+w ij );

[0152] Among them, d(v j ) is the node v j The shortest distance to the source node, w i j is the node v i To node v j The weight of the edge;

[0153] Repeat the above steps until the shortest paths of all nodes are calculated;

[0154] S342, Optimization of communication paths: In social networks, the efficiency of information dissemination is affected by factors such as the closeness of social relationships, network structure, and user activity. Therefore, when optimizing the communication path, it is necessary to analyze the key nodes and communication channels in the social network to further improve the effect of information dissemination. The specific steps are as follows:

[0155] Information propagation efficiency evaluation: Based on the results of the shortest path analysis, the efficiency of information propagation from the source node to the target node is evaluated. The measurement criteria include: propagation time, number of propagation nodes, and information loss rate;

[0156] The effectiveness of the path can be evaluated by calculating the propagation speed and coverage of information dissemination. The specific formula is as follows:

[0157]

[0158] Among them, V 传播 represents the propagation speed, R 传播 Indicates coverage;

[0159] Identification of core nodes and social group structure: Through social network analysis (such as S31, S32), core nodes in social networks are identified, which play an important role in the information dissemination process.

[0160] Based on group structure analysis, we can find out the close social groups in the social network, and the information dissemination between these groups is faster and more efficient.

[0161] Optimize the transmission path: By using core nodes and social groups as the hub of information transmission, the transmission path of information from source nodes to target nodes is optimized. The optimization strategies include:

[0162] Select the starting point (source node) of information dissemination as a high centrality node to ensure that the information reaches multiple users quickly;

[0163] Find shortcuts for information dissemination in social network structures, reduce unnecessary dissemination steps through the shortest paths and core nodes, and increase the speed of information transmission;

[0164] Optimal communication strategy: formulate the best communication strategy based on the optimized communication path;

[0165] Design communication strategies between groups to ensure efficient flow of information between different social groups

[0166] S343, Information Diffusion Model: The process of information dissemination in social networks is often affected by factors such as the social influence and personal preferences of different users. A linear threshold model is used to simulate the effect of information diffusion. The specific steps are as follows:

[0167] Model definition: Let each node v in the social network i There is a threshold θ i ∈[0,1], represents the sensitivity of the node to receiving information. The smaller the threshold, the easier it is for the node to receive information.

[0168] The amount of information received by each node is the weighted sum of the amount of information transmitted by its neighboring nodes. The weight of information transmission is determined by the edge weight w ij express;

[0169] Information propagation process: Information is transmitted from source node v s Spread to its neighboring nodes, each node calculates its activation degree when receiving information from its neighboring nodes, expressed as:

[0170] a(v i )=∑ j∈N(i) w ij ·I(v j );

[0171] Among them, a(v i ) is the node v i The activation level, I(v j ) represents node v j Has it been activated (has the information been spread), w ij is the weight of the edge;

[0172] If the activation degree of the node exceeds the threshold θ i , then the node is activated and starts to spread information;

[0173] Simulation and prediction: By simulating the propagation process, the propagation potential of each node in the social network is evaluated and the contribution of different nodes to information diffusion is predicted;

[0174] Through multiple simulations, the communication capacity of each node is calculated to identify the most influential users in the network (i.e. the most influential nodes);

[0175] Evaluation of communication effect: Through information diffusion simulation, the communication effect of information in social networks is evaluated. The evaluation criteria include the scope of information dissemination, the depth of dissemination and the speed of dissemination;

[0176] By analyzing the dissemination potential of different nodes, the dissemination strategy is adjusted to improve the information diffusion effect.

[0177] S4 includes:

[0178] S41, data fusion and feature extraction: Fuse social network analysis data and health behavior data, and extract features related to health behavior identification, including:

[0179] Data fusion: The results obtained from social network analysis (such as social interaction data, social influence index, etc.) are integrated with the health behavior data obtained from health management applications (such as steps, diet, work and rest, etc.). In order to ensure the consistency and integrity of the data, a time alignment method is used to ensure that the timestamps of different data sources are consistent so that they can represent the user's health behavior in the same period.

[0180] Feature extraction: Extract features from the fused data to extract the most representative features for identifying the user's health behavior patterns, including the following categories:

[0181] Lifestyle characteristics: including work and rest time, exercise frequency, dietary preferences, etc.;

[0182] Social influence characteristics: including interaction frequency, information transmission path, social influence, etc. in social networks;

[0183] Behavioral trend features: such as periodic fluctuations in step count or exercise frequency, which can then extract user behavior patterns;

[0184] S42, model construction: Based on the extracted features, a health behavior pattern recognition model is constructed using the K-means clustering algorithm to predict the user's behavior pattern and assess their health risks. The specific steps include:

[0185] Data preparation and standardization: The features extracted from social network analysis and health behavior data are used as input data to ensure that the scales of all features are consistent to avoid the dimensional differences between different features affecting the clustering results;

[0186] Determine the number of clusters (K value): Use the elbow rule to determine the number of clusters K in the K-means algorithm;

[0187] The elbow rule plots the sum of squared errors (SSE) at different K values ​​to find the "elbow" point as the optimal K value;

[0188] Apply K-means clustering algorithm: Use the selected K value and adopt K-means clustering algorithm to perform cluster analysis on the feature data;

[0189] The K-means clustering algorithm assigns each user to the nearest cluster center through iterative optimization, thus forming K different health behavior pattern categories, each category representing a user group with similar health behavior characteristics;

[0190] The specific algorithm steps include:

[0191] Initialize K cluster centers (you can randomly select K sample points, or use the k-means++ algorithm to initialize the center points);

[0192] Based on user data, calculate the distance between each user and K cluster centers and assign them to the nearest cluster center;

[0193] Update each cluster center to the mean of the sample points to which it belongs;

[0194] Repeat the above steps until the cluster center no longer changes or the preset number of iterations is reached;

[0195] Model evaluation and adjustment: Evaluate the clustering results, analyze the matching degree between the clustering effect and the user behavior pattern, use the silhouette coefficient to evaluate the quality of clustering, ensure the stability and interpretability of the clustering results, and adjust the K value or the parameters of the clustering algorithm according to the evaluation results to obtain a more appropriate behavior pattern division;

[0196] Health risk assessment: Based on the user behavior pattern of each cluster, combined with health behavior data (such as number of steps, exercise frequency, diet, etc.), the health risk corresponding to each health behavior pattern is assessed;

[0197] A health risk scoring model is used to assign a risk level to each user's behavior pattern based on the characteristics of health behaviors (such as low physical activity, high unhealthy diet, etc.), which are divided into three categories: high risk, medium risk and low risk.

[0198] Calculation formula example:

[0199] Risk score R i =w1·X1+w2·X2+…+w n ·X n , where X i are features (such as steps, diet, sleep, etc.), w i is the corresponding weight, indicating the impact of the characteristic on health risk;

[0200] S43, Health behavior characteristic profile generation: Generate a health behavior characteristic profile for each user through the health behavior pattern recognition model, and conduct health risk assessment based on the results of behavior pattern recognition to provide a basis for personalized intervention, including:

[0201] Health behavior feature profile generation: Based on the results of K-means clustering, each user is divided into the corresponding health behavior pattern group, and the user's health behavior feature profile is generated based on the group label. The profile contains:

[0202] Behavior category label: Users are divided into different health behavior groups based on clustering results, such as healthy behavior type, potential risk type, deterioration type, etc.;

[0203] Risk assessment results: Based on the analysis of health behaviors within the group, a health risk assessment report is provided for each user, such as low risk, medium risk, and high risk;

[0204] Personalized suggestions: Based on the user's group characteristics, provide personalized health intervention suggestions, such as increasing exercise, adjusting diet, improving work and rest, etc.;

[0205] S44, Health Risk Identification and Health Management Goal Setting: Based on the health behavior profile, identify the user's health risks and set health management goals to ensure the effectiveness of intervention measures, including:

[0206] Health risk identification: Identify high-risk users based on their health behavior profiles and assess their likelihood of illness. For example, users with long-term low physical activity and irregular diet are identified as high-risk users.

[0207] Health management goal setting: Based on the results of health risk assessment, personalized health management goals are set for users. These goals include improving diet, increasing exercise, adjusting work and rest schedules, etc., to ensure that reasonable intervention measures are taken to help users reduce health risks.

[0208] S5 includes:

[0209] S51, Health Behavior Profile Analysis: Based on the generated health behavior profile, identify the user's strengths and potential risks in health management and develop personalized intervention strategies for the user, including:

[0210] Health behavior pattern analysis: Analyze the user's steps, diet, exercise frequency, and rest time and other health behavior data to identify the user's health behavior pattern, especially potential risk behaviors (such as long-term sitting, uneven diet, etc.);

[0211] Health risk assessment: Based on the analysis results, the user's current health risk is assessed. For example, a user may have a higher risk of obesity or cardiovascular disease.

[0212] Improvement point extraction: Combine user behavior data to identify specific areas that require intervention and improvement, such as increasing exercise, improving eating habits, etc.

[0213] S52, Social Relationship Analysis: Identify the user's social support system and social influence by analyzing the user's relationships and interaction patterns in social networks, including:

[0214] Social relationship graph extraction: extract the user's social network data, build the user's social relationship graph, analyze the user's social circle and interaction frequency, and identify key figures or social influencers in the social network;

[0215] Social influence assessment: By analyzing the user's interactive behavior in social networks, the social influence index is calculated to assess the user's status in the social circle and his ability to influence the behavior of others;

[0216] Social support identification: Identify social support groups related to user health management, analyze the influence of these groups, and their positive or negative effects on user health behaviors;

[0217] S53, Intervention goal setting: Based on the user's health behavior profile and social relationship analysis, set personalized health behavior improvement goals to ensure the pertinence and effectiveness of intervention measures, including:

[0218] Health behavior improvement goals: Set specific health improvement goals for users, such as increasing daily steps, improving diet quality, increasing exercise, etc.

[0219] Social goal setting: Based on the results of social network analysis, social goals are set, such as enhancing social interaction, improving the influence of users’ healthy behaviors in social networks, or guiding users to participate in healthy social activities;

[0220] Health risk reduction goals: Set health risk reduction goals based on the user's current health risk status, such as reducing sedentary time, increasing fruit and vegetable intake, etc.

[0221] S54, Personalized intervention strategy development: Design intervention strategies based on health behavior improvement goals and social goals to help users achieve their health management goals, including:

[0222] Provide users with personalized health behavior recommendations based on their health behavior data. For example, for users who do not exercise enough, develop a plan to gradually increase their exercise amount; for users with an unbalanced diet, provide a reasonable diet adjustment plan.

[0223] S6 includes:

[0224] S61, Incentive Mechanism Goal Setting: Clarify the goals of the healthy behavior incentive mechanism and ensure that the incentive plan meets the health needs of users, including:

[0225] Long-term health goal setting: Set personalized long-term health goals based on the user's health assessment data, including weight loss, improving cardiovascular health, lowering blood sugar levels, and increasing physical strength.

[0226] For example: losing 10 kg, maintaining normal blood sugar levels, achieving a healthy body mass index (BMI) target, etc.

[0227] Short-term behavioral goal setting: Based on long-term health goals, break them down into short-term behavioral goals to ensure that users can achieve healthy behaviors every day or every week. For example: walking 8,000 steps a day, doing aerobic exercise 3 times a week, maintaining a healthy diet, etc.

[0228] Health behavior assessment standards: Based on the quantification of data, set specific behavior standards, such as number of steps, exercise duration, calorie consumption, etc., to form measurable goals;

[0229] S62, Incentive Reward Design: Design a specific reward mechanism to encourage users to continue to participate and maintain healthy behaviors, including:

[0230] Material rewards: motivate users to achieve health goals through material rewards, such as:

[0231] Health products: such as fitness equipment, sports shoes, smart health bracelets, etc.;

[0232] Service rewards: such as free health checks, gym memberships, healthy food packs, etc.

[0233] Discount coupons: After users complete their set health goals, they will be rewarded with discount coupons for health products or services;

[0234] Points rewards: Points are awarded based on the user's health behavior (number of steps, exercise frequency, diet control, etc.). Points can be exchanged for items, services, or become vouchers for specific prizes;

[0235] For example: 100 points will be awarded for every 10,000 steps completed, 50 points will be awarded for each healthy diet record submission, etc.

[0236] Social Rewards: Motivate users through social interactions and increase participation in their social network, such as:

[0237] Social honors: such as virtual medals, health rankings, etc., where users display their health achievements on social platforms;

[0238] Social Challenge: Create a health challenge to compare healthy behaviors with friends or other users, and motivate healthy behaviors through competition results;

[0239] Psychological motivation: Motivate users through positive psychological feedback and stimulate their internal motivation. For example:

[0240] Daily encouragement messages: remind users through APP or SMS to stick to healthy behaviors and provide positive psychological feedback;

[0241] Virtual achievements: For example, after completing the "30-day health challenge", you will be awarded the title of "Health Expert";

[0242] S63, incentive triggering condition setting: setting incentive triggering conditions to ensure the timeliness and effectiveness of incentive measures, including:

[0243] Behavior completion trigger: When the user reaches the set health behavior goal, the reward is triggered. For example:

[0244] Daily step count reaches the target: walking more than 8,000 steps a day will trigger the points reward;

[0245] Healthy eating record: Keep a weekly eating record, reach healthy eating goals, and receive virtual prizes;

[0246] Periodic incentives: Set periodic goals and evaluate the achievement of healthy behaviors based on weeks and months, for example:

[0247] When the weekly step count reaches 56,000 (weekly step count target), a material reward will be triggered;

[0248] If the monthly healthy eating goal is achieved by 80%, a healthy food package will be awarded;

[0249] Social interaction incentives: Incentives are triggered by user interactions on social platforms, such as:

[0250] Friends’ participation: Users can invite friends to participate in health challenges, and both themselves and their friends can receive rewards.

[0251] Social sharing: Users share their health achievements on social platforms and receive social rewards, such as points, virtual honors, etc.

[0252] The S6 also includes:

[0253] S64, Incentive feedback mechanism: Design an incentive feedback mechanism to enable users to perceive the results of healthy behaviors and maintain motivation, including:

[0254] Instant feedback: After users complete their health goals, they are given reward feedback immediately. For example:

[0255] The app shows “Congratulations on completing today’s goal and getting 100 points!”

[0256] Publish users’ achievements and awards on social platforms;

[0257] Regular feedback reports: Generate weekly or monthly health behavior feedback reports that detail the user's progress in setting goals and provide specific improvement suggestions. For example:

[0258] "You have successfully lost 3 kg this month. If you keep up the trend, you can achieve your annual weight loss goal."

[0259] Healthy behavior trend chart: Through data visualization, it shows the user's healthy behavior progress trend, including steps, calories consumed, exercise duration, etc., so that users can see the results of long-term efforts and motivate them to continue to participate;

[0260] S65, Incentive Mechanism Optimization: Based on user behavior feedback, continuously adjust and optimize the incentive mechanism to ensure its long-term effectiveness, including:

[0261] Data analysis optimization: Optimize the reward form based on user participation data and feedback, for example:

[0262] Analyze which rewards are most popular with users and which rewards have weaker incentive effects, and adjust the reward plan accordingly;

[0263] Adjust the form of social incentives, add more social interaction incentive elements, and improve users' social interaction participation;

[0264] Personalized reward recommendations: Through data mining and user preference analysis, personalized reward recommendations are provided, such as:

[0265] When users prefer health products or services, we recommend redemption of related health products.

[0266] For users who have social interaction needs, add more social challenges and virtual honors.

[0267] Dynamically adjust reward standards: Adjust reward standards based on the user's health progress to avoid incentive programs that are too simple or lack challenges, for example:

[0268] When users get used to the goal of walking 8,000 steps a day, they can raise the goal to 10,000 steps to maintain a moderate level of challenge.

[0269] Adjust rewards based on changes in the user's health to encourage continued effort in the new phase.

[0270] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0271] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A health behavior intervention and motivation method based on social network analysis, characterized in that: The following steps are involved: S1, data collection: by collecting users’ health behavior data, social network data, and interaction data on multiple social platforms and health management applications; The health behavior data includes steps, diet, exercise frequency and rest time; The social network data includes social relationship graph and social interaction frequency; The interaction data includes emotional connections and recommendation behaviors between users; S2, data preprocessing: cleaning, deduplication and standardization of the collected health behavior data, social network data and interaction data; S3, social network analysis: Based on the preprocessed data, social network analysis technology is used to model the user's social network and extract the structural features of social nodes and networks, including social group structure, social influence index and information dissemination path; S4, user behavior pattern identification: based on the social network analysis results, combined with health behavior data, a machine learning algorithm is used to identify the user's health behavior pattern, generate the user's health behavior feature profile, and identify potential health risk behaviors and health management goals; S5, personalized intervention strategy generation: formulate personalized health behavior intervention strategies based on the user's health behavior profile and social relationships in the social network; S6, Design of healthy behavior incentive mechanism: Design a healthy behavior incentive mechanism to encourage users to continue to maintain good health behaviors.

2. The method for health behavior intervention and motivation based on social network analysis according to claim 1, characterized in that: The S1 includes: S11, multi-platform data collection channels: real-time collection of users’ health behavior data, social network data, and interaction data on multiple social platforms and health management applications; S12, health behavior data collection: collect the user's health behavior data through the user's smart device, including steps, diet, exercise frequency and rest time; S13, social network data collection: collect users’ social network data through social platform API interfaces, including social relationship graphs and social interaction frequencies; S14, interactive data collection: by collecting users’ interactive behaviors in real time on social platforms and health management applications, capturing the interactive data between users and other social network members.

3. The health behavior intervention and motivation method based on social network analysis according to claim 2 is characterized in that: The S2 includes: S21, data cleaning: clean the collected health behavior data, social network data and interaction data, and delete incomplete, missing or erroneous data records; S22, data deduplication: deduplication of collected health behavior data, social network data and interaction data; S23, data standardization: standardize health behavior data, social network data, and interaction data; S24, time series data alignment: for health behavior data, social network data and interaction data collected across time periods, time series data alignment is performed; S25, data conversion and feature construction: perform feature construction on the cleaned and standardized data.

4. The health behavior intervention and motivation method based on social network analysis according to claim 3 is characterized in that: The S3 includes: S31, Construction of social network modeling framework: Based on the preprocessed data, the user's social network model is constructed using graph theory methods; S32, Topological structure analysis of social networks: Use social network analysis technology to perform topological structure analysis on the constructed social network model and extract the basic structural features of the social network.

5. The method for health behavior intervention and motivation based on social network analysis according to claim 4, characterized in that: The S3 further includes: S33, social influence index calculation: through social network analysis, calculate the social influence index of each user to measure the user's communication ability in the social network; S34, Information propagation path analysis: Based on social network analysis, identify the information propagation path in the network and determine the key nodes and propagation paths of information propagation.

6. The method for health behavior intervention and motivation based on social network analysis according to claim 5, characterized in that: The S4 includes: S41, data fusion and feature extraction: fuse social network analysis data and health behavior data, and extract features related to health behavior identification; S42, model construction: Based on the extracted features, a health behavior pattern recognition model is constructed using the K-means clustering algorithm to predict the user's behavior pattern and assess their health risks; S43, health behavior characteristic profile generation: through the health behavior pattern recognition model, a health behavior characteristic profile is generated for each user, and health risk assessment is performed based on the results of behavior pattern recognition to provide a basis for personalized intervention; S44, Health risk identification and health management goal setting: Based on the health behavior characteristic profile, identify the user's health risks and set health management goals.

7. The method for health behavior intervention and motivation based on social network analysis according to claim 6, characterized in that: The S5 includes: S51, health behavior profile analysis: Based on the generated health behavior profile, identify the user's strengths and potential risks in health management and develop personalized intervention strategies for the user; S52, Social Relationship Analysis: Identify the user’s social support system and social influence by analyzing the user’s relationships and interaction patterns in social networks; S53, intervention goal setting: setting personalized health behavior improvement goals based on the user's health behavior profile and social relationship analysis; S54, Personalized intervention strategy development: Design intervention strategies based on health behavior improvement goals and social goals to help users achieve their health management goals.

8. The health behavior intervention and motivation method based on social network analysis according to claim 1 is characterized in that: The S6 includes: S61, Incentive Mechanism Goal Setting: Clarify the goals of the healthy behavior incentive mechanism and ensure that the incentive program meets the health needs of users; S62, Incentive Reward Design: Design reward mechanisms to motivate users to continue to participate and maintain healthy behaviors; S63, incentive triggering condition setting: setting incentive triggering conditions to ensure the timeliness and effectiveness of incentive measures.

9. The health behavior intervention and motivation method based on social network analysis according to claim 8, characterized in that: The S6 further includes: S64, Incentive feedback mechanism: Design an incentive feedback mechanism to enable users to perceive the results of healthy behaviors and maintain motivation; S65, Incentive mechanism optimization: Continuously adjust and optimize the incentive mechanism based on user behavior feedback.

Images