User health intelligent management method and system based on AI

Abstract

The invention relates to the technical field of user health care management, and discloses an AI-based user health intelligent management method and system, and the system comprises a user living habit behavior stage prompt module, a user living habit behavior performance management module, and a user living habit behavior performance feedback management module. A robot carries a cloud lens to dynamically collect user living habit behavior fulfillment real-time image information and accurately analyze user living habit real-time fulfillment action information in combination with an internet search platform, so that user living habit behavior fulfillment dynamic intelligent monitoring is realized; according to the user living habit real-time performance action analysis parameters, an artificial intelligence identification algorithm and user living habit behavior performance content data are combined to carry out user living habit behavior performance real-time state intelligent evaluation, and the user living habit behavior real-time performance state intelligent supervision is realized. And the scientificity of user living habit behavior health management is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of user healthcare management, and specifically to an AI-based user health intelligent management method and system. Background Art

[0002] Health management is a medical behavior and process that uses modern health concepts, including physical, psychological, and social adaptability, as well as new medical models, including physical, psychological, and social, and is guided by traditional Chinese medicine. It uses the theories, technologies, methods, and means of modern medicine and modern management to comprehensively detect, evaluate, effectively intervene, and continuously track the overall health status of individuals and groups and the risk factors that affect their health. Its purpose is to obtain the greatest health benefits with the least investment. Health management aims to prevent and control the occurrence and development of diseases, reduce medical costs, and improve the quality of life. It provides health education for individuals and groups, improves self-management awareness and level, and addresses health risk factors related to their lifestyle. Health management includes not only medical management, but also user lifestyle management. Existing user lifestyle health management relies mostly on user consciousness, and cannot achieve intelligent supervision of user lifestyle behaviors, nor can it achieve punishment management for users who fail to fulfill their lifestyle behaviors.

[0003] The Chinese invention patent application with publication number CN114943629A discloses a health management and health care service system and a health management method thereof, which provides health intervention, health tracking guidance and health effect evaluation for users through health intervention plans, and issues health warnings when abnormal values ​​appear in the user's health indicator information; the health manager's mobile terminal is used for information maintenance of the health manager and information interaction with the customer's mobile terminal. The above technical solutions cannot realize intelligent supervision and management of the user's health management process. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In order to solve the problem that the existing user life habit health management relies more on user consciousness, cannot realize intelligent supervision of user life habit behavior, and cannot realize punishment management of users who fail to fulfill their life habit behavior, the above purpose of supervising user life habit behavior based on time characteristics, intelligently judging the status of user life habit behavior implementation, autonomously generating punishment plan for users who fail to fulfill their life habit behavior, intuitively feedback the results of user life habit behavior implementation, and realizing intelligent user health intelligent management is achieved.

[0006] (2) Technical solution

[0007] The present invention is implemented through the following technical solution: an AI-based user health intelligent management method, the method comprising the following steps:

[0008] S1. Collecting user life habit behavior plan text data and classifying the user life habit behavior plan time period and plan content information to generate user life habit behavior plan time period data and user life habit behavior plan content data respectively;

[0009] S2. Collect the data at the current time point and perform a judgment on the implementation stage of the user's lifestyle behavior with the data of the planned time period of the user's lifestyle behavior to generate data for judging the implementation stage of the user's lifestyle behavior. If the implementation stage has not been reached, repeat step S2 until the implementation stage is reached.

[0010] S3. When the user reaches the implementation stage, searching for implementation content of the specific implementation stage of the user's lifestyle behavior based on the user's lifestyle behavior implementation stage judgment data and the user's lifestyle behavior plan content data to generate user's lifestyle behavior implementation content data;

[0011] S4. Execute user life habit behavior execution prompts based on the execution time period and execution content information of the user's life habit behavior, and collect real-time image data of the user's life habit behavior execution;

[0012] S5. Analyze and process the real-time action object information of the user's life habit behavior based on the real-time image data of the user's life habit behavior performance to generate real-time action analysis data of the user's life habit performance. Analyze and process the real-time status of the user's life habit behavior performance together with the real-time status of the user's life habit behavior performance data to generate real-time status judgment data of the user's life habit behavior performance. If the performance is normal, continue to repeat steps S4 and S5 until the user fails to perform normally, thereby monitoring the user to complete the content of the life habit behavior plan in the current performance stage, and repeat steps S2, S3, S4, and S5 until the user fails to perform normally, or directly execute step S7 after the user has normally completed the content of the life habit behavior plan in all performance stages.

[0013] S6. When the user fails to fulfill the habitual behavior plan, a penalty plan search is performed based on the real-time performance analysis data of the user's living habits and the penalty plan data for failure to fulfill different types of living habits. The penalty plan data for failure to fulfill the user's living habits is generated, and steps S2, S3, S4, and S5 are repeated after the user completes the content of the living habits plan for the current fulfillment stage.

[0014] S7. Construct the user's life habit behavior execution result data and perform the user's life habit behavior execution result feedback task.

[0015] Preferably, the steps of collecting the user's life habit behavior plan text data and performing data classification processing on the user's life habit behavior plan time period and plan content information to generate the user's life habit behavior plan time period data and the user's life habit behavior plan content data are as follows:

[0016] S11, inputting the planned time period of the user's daily living habits and the text information of the living habits plan items corresponding to the planned time period online through the dialog box of the user health management platform, and generating the user's living habits plan text data Y;

[0017] S12, using the KD tree nearest neighbor search algorithm to perform classification search processing on the user's life habit behavior plan text data Y according to the time period and the life habit behavior plan content keywords corresponding to the time period, and generate the user's life habit behavior plan time period data set A and the user's life habit behavior plan content data A' respectively; where A = (a1, ..., a m ,…,a ε ), m=1,2,3,…,ε; where a m represents the user's life habit behavior plan time period data corresponding to the mth planned time period of the day, ε represents the maximum number of user's life habit behavior plan time periods; a m =[a m1 ,a m2 ], where a m1 and a m2 Respectively represent the user's life habit behavior plan time period data a m The minimum time data of the user's life habit behavior plan and the maximum time data of the user's life habit behavior plan, a m1 and a m2 The units are composed of year, month, day, hour and minute; A'=(a'1,…,a' m ,…,a' ε ), where a' m Represents the user's life habit behavior plan content data corresponding to the mth planned time period of the day, and the user's life habit behavior plan content data includes rest plan content information, diet plan content information, exercise plan content information, reading plan content information, dance plan content information, painting plan content information, calligraphy plan content information, photography plan content information and musical instrument plan content information.

[0018] Preferably, the current time point data is collected and combined with the user's life habit behavior planned time period data to perform a judgment process on the user's life habit behavior implementation stage to generate user life habit behavior implementation stage judgment data. If the implementation stage has not been reached, step S2 is repeatedly executed until the implementation stage is reached as follows:

[0019] S21. Collect the current specific time point information online through the time timing module and generate the current time point data Z, where the unit of Z is composed of year, month, day, hour, and minute;

[0020] S22, compare the current time point data Z with the user life habit behavior plan time period data a in the user life habit behavior plan time period data set A. m Perform time value comparison and generate user life habit behavior execution stage judgment data A based on the time value comparison results jieduan ;

[0021] When Z∈a m When , it means that the user has reached the stage of fulfilling the habitual behavior, then the judgment data A of the stage of fulfilling the habitual behavior is output. jieduan To the implementation stage, the user's life habit behavior plan time period data a that matches the current time point data Z is output simultaneously m Corresponding planned time period information and planned time period number information;

[0022] when When , it means that the user has not yet reached the stage of fulfilling the habitual behavior, then the user's habitual behavior fulfillment stage judgment data A is output. jieduan If the user's living habit behavior has not reached the fulfillment stage, then step S2 is repeated until the user's living habit behavior fulfillment stage judgment data A is reached. jieduan To the implementation stage.

[0023] Preferably, when the implementation stage is reached, the implementation content search process of the specific implementation stage of the user's life habit behavior is performed based on the user's life habit behavior implementation stage judgment data and the user's life habit behavior plan content data, and the operation steps for generating the user's life habit behavior implementation content data are as follows:

[0024] S31, when the user's living habit behavior is performed, the judgment data A jieduan When it comes to the implementation stage, the iterative deepening search algorithm is used to judge the implementation stage of the user's living habit behavior data A jieduan The user's life habit behavior plan content data a' in the user's life habit behavior plan content data A' m Perform character matching on the planned time period number to search for the user's life habit behavior execution stage judgment data A jieduan The corresponding user life habit behavior plan content data a′ m , and generate user life habits and behavior content data through data identification

[0025] Preferably, the steps of performing the user's life habit behavior performance prompting task based on the performance time period and performance content information of the user's life habit behavior and collecting the user's life habit behavior performance real-time image data are as follows:

[0026] S41, through the mobile terminal, the user's life habit behavior execution stage judgment data A jieduan Corresponding life habit behavior plan execution time period information and the user's life habit behavior execution content data The corresponding life habit behavior plan fulfillment content information executes a user life habit behavior fulfillment prompt operation, and the mobile terminal includes any one of a smart bracelet, a smart phone and a tablet computer;

[0027] S42, during the user's life habit behavior execution operation, the robot equipped with a cloud camera is used to judge the data A according to the user's life habit behavior execution stage. jieduan The corresponding life habit behavior plan execution time period information collects the real-time action image information of the user performing the life habit behavior online, and generates the user's life habit behavior execution real-time image data O.

[0028] Preferably, the real-time action object information analysis and processing of the user's life habit behavior performance is performed based on the real-time image data of the user's life habit behavior performance, and the real-time performance analysis data of the user's life habit behavior is generated. The real-time status judgment processing of the user's life habit behavior performance is performed with the user's life habit behavior performance content data to generate the real-time performance status judgment data of the user's life habit behavior. When the performance is normal, the steps S4 and S5 are repeatedly executed until the user fails to perform normally, so as to supervise the user to complete the life habit behavior plan content of the current performance stage, and the steps S2, S3, S4, and S5 are repeated until the user fails to perform normally, or the user performs normally and completes the life habit behavior plan content of all performance stages and directly executes step S7. The operation steps are as follows:

[0029] S51. Importing the real-time image data O of the user's life habit behavior into a dialog box of an Internet search platform to perform graphic-text translation processing on the real-time action object feature information of the user's life habit behavior, thereby generating real-time action analysis data P of the user's life habit, wherein the real-time action analysis data of the user's life habit includes resting action types, eating action types, exercising action types, reading action types, dancing action types, painting action types, calligraphy action types, photography action types, and musical instrument action types, and the Internet search platform includes any one of a Google search platform, a Baidu search platform, and a 360 search platform;

[0030] S52, the user's life habit real-time performance action analysis data P and the user's life habit behavior performance content data Performing lifestyle behavior keyword matching, generating user lifestyle behavior real-time performance status judgment data B based on the lifestyle behavior keyword matching results, and executing the specific steps for generating the user lifestyle behavior real-time performance status judgment data B are as follows:

[0031] S521, initialization, update the maximum number of iterations T and update the performance status to identify the location of the pelican population. The formula for updating the performance status to identify the location of the pelican population is: where R i,j Indicates the position of the i-th performance status identification pelican in the j-th dimension, that is, the i-th performance status identification pelican in dimension j is the user's life habit behavior performance content data The position in the search space, Indicates a random integer for position adjustment, rand indicates a random number in the range [0,1], ψ j and ζ j Represents the user's life habit behavior fulfillment content data in dimension j Searching the search space for upper and lower boundaries of the life habit behavior keywords that match the user's life habit real-time performance action analysis data P;

[0032] S522, exploration stage, the implementation status identification pelican to determine the location of the prey, that is, in the user's life habit behavior implementation content data Search the search space to find the location of the life habit behavior keyword prey that matches the real-time performance analysis data P of the user's life habit, and model the performance state recognition pelican approaching prey strategy so that the algorithm can perform the content data of the user's life habit behavior The search space is scanned, and the life habit behavior keywords that match the user life habit real-time performance action analysis data P are searched in the algorithm and are found in the user life habit behavior performance content data The search space is randomly generated, and the formula for approximating the prey strategy is: where R′ i，j Indicates the i-th fulfillment status identification of the user's life habit behavior fulfillment content data in dimension j after the exploration phase update The position in the search space, M j The searched-out life habit behavior keyword prey that matches the user's life habit real-time performance action analysis data P is the user's life habit behavior performance content data in the jth dimension The position in the search space, Ξ M represents the objective function value of the searched life habit behavior keyword prey that matches the user's life habit real-time performance action analysis data P, Ξ irepresents the objective function value of the i-th fulfillment state identification pelican;

[0033] S523, development stage, implementation status identification Pelican adopts the surface flying strategy to implement content data in the user's living habits behavior Hunt out the life habit behavior keyword prey that matches the user's life habit real-time performance action analysis data P in the search space, and model the hunting behavior process of the performance state recognition pelican, so that the algorithm can be used to identify the user's life habit behavior performance content data. Search the search space to find the life habit behavior keyword prey that best matches the user's life habit real-time performance action analysis data P, and the performance status recognition pelican hunting behavior calculation formula Where R i,j Indicates the i-th fulfillment status after the development phase update to identify the user's life habits and behavior fulfillment content data in dimension j of Pelican The position in the search space, rand is a random number in the range [0,1]; ω is a random integer of 0 or 2; t is the current number of iterations; T is the maximum number of iterations;

[0034] S524, when the algorithm meets the maximum number of iterations, the user's living habits are used to perform the action analysis data P in real time and the user's living habits behavior content data Performing lifestyle behavior keyword matching results to generate user lifestyle behavior real-time performance status judgment data B;

[0035] When P and If no matching of the life habit behavior keywords is successful, it means that the user has not properly fulfilled the current life habit behavior plan content, then the user's life habit behavior real-time fulfillment status judgment data B is outputted, indicating that the user's life habit behavior has not been properly fulfilled;

[0036] When P and If the keyword matching of the life habit behavior is successful, it means that the user is fulfilling the current life habit behavior plan content normally, and the real-time fulfillment status judgment data B of the user's life habit behavior is normally fulfilled is output. At this time, steps S4 and S5 are repeatedly executed until the real-time fulfillment status judgment data B of the user's life habit behavior is not fulfilled normally, so as to supervise the user to complete the content data of the user's life habit behavior fulfillment content. After the corresponding current stage life habit behavior plan content is implemented, steps S2, S3, S4, and S5 are repeatedly executed until the real-time implementation status judgment data B of the user's life habit behavior is not implemented normally, or the user has normally completed all the implementation stage life habit behavior plan contents in the user's life habit behavior plan content data A' and directly executes step S7.

[0037] Preferably, when the behavior is not performed normally, the penalty scheme search process for the user's habitual behavior failure is performed based on the real-time performance analysis data of the user's living habits and the penalty scheme data for failure to perform different types of habitual behaviors, the penalty scheme data for failure to perform the user's habitual behavior is generated, and the user is continuously supervised to complete the content of the habitual behavior plan in the current performance stage. The operation steps of repeating steps S2, S3, S4, and S5 are as follows:

[0038] S61, when the user's living habit behavior real-time performance status judgment data B is not performed normally, establish different types of living habit behavior non-performance penalty plan data set C = (c1, ..., c n ,…,c γ ), n=1,2,3,…,γ; where c n represents the penalty plan data for different types of lifestyle behavior failure corresponding to the nth lifestyle behavior type, γ represents the maximum number of lifestyle behavior types; lifestyle behavior types include resting, eating, exercising, reading, dancing, painting, calligraphy, photography, and musical instruments; the penalty plan data for different types of lifestyle behavior failure represents text information of standard penalty plans set for users who fail to perform different types of lifestyle behaviors normally; the penalty plan data for different types of lifestyle behavior failure include standing penalty plan information, running plan information, horse stance plan information, reading aloud plan information, and housework plan information;

[0039] S62, using a bidirectional search algorithm to compare the user's life habit real-time performance analysis data P with the different types of life habit behavior non-performance penalty solution data C in the different types of life habit behavior non-performance penalty solution data set C. n Perform character matching on life habits and behaviors, and search for the penalty plan data c corresponding to the real-time action analysis data P of the user's life habits. n , and generate data on user lifestyle behavior that fails to comply with the penalty plan through data identification At this time, continue to supervise the user to complete the user's life habit behavior fulfillment content data After the corresponding current stage of implementation of the life habit behavior plan content, repeat steps S2, S3, S4, and S5.

[0040] Preferably, the steps of constructing the user's life habit behavior fulfillment result data and performing the user's life habit behavior fulfillment result feedback operation are as follows:

[0041] S71, the user life habit behavior plan time period data set A, the user life habit behavior plan content data A', the user life habit behavior implementation stage judgment data A of the user life habit behavior implementation stage judgment result to the implementation stage jieduan , the user's living habits behavior real-time implementation status judgment data B and the user's living habits behavior non-implementation penalty plan data Combine the data to construct the user's life habit behavior performance result data K, where

[0042] S72: Transmit the user's life habit behavior fulfillment result data K to the user health management platform via the mobile communication network to perform the user's life habit behavior fulfillment result feedback operation.

[0043] An AI-based user health intelligent management system, used to implement the AI-based user health intelligent management method, the system includes a user life habit behavior stage prompt module, a user life habit behavior fulfillment management module, and a user life habit behavior fulfillment feedback management module;

[0044] The user life habit behavior stage prompt module includes a user life habit behavior plan text information collection unit, a user life habit behavior plan time period generation unit, a user life habit behavior plan content generation unit, a current time point information collection unit, a user life habit behavior fulfillment stage judgment unit, and a user life habit behavior fulfillment content search unit;

[0045] The user life habit behavior plan text information collection unit collects user life habit behavior plan text data through the user health management platform dialog box; the user life habit behavior plan time period generation unit searches for the user life habit behavior plan time period information based on the user life habit behavior plan text data to generate user life habit behavior plan time period data; the user life habit behavior plan content generation unit searches for the user life habit behavior plan content information based on the user life habit behavior plan text data to generate user life habit behavior plan content data; the current time point information collection unit collects current time point data through the time timing module; the user life habit behavior fulfillment stage judgment unit judges the fulfillment stage of the user life habit behavior based on the current time point data and the user life habit behavior plan time period data to generate user life habit behavior fulfillment stage judgment data; the user life habit behavior fulfillment content search unit searches for the fulfillment content of the specific fulfillment stage of the user life habit behavior based on the user life habit behavior fulfillment stage judgment data and the user life habit behavior plan content data to generate user life habit behavior fulfillment content data;

[0046] The user life habit behavior fulfillment management module includes a user life habit behavior fulfillment content prompt unit, a user life habit behavior fulfillment real-time image acquisition unit, a user life habit behavior real-time fulfillment action analysis unit, a user life habit behavior real-time fulfillment status judgment unit, a different type of life habit behavior non-fulfillment penalty plan storage unit, and a current life habit behavior non-fulfillment penalty plan search unit;

[0047] The user life habit behavior fulfillment content prompt unit performs the user life habit behavior fulfillment prompt operation based on the fulfillment time period and fulfillment content information of the user life habit behavior in combination with the mobile terminal; the user life habit behavior fulfillment real-time image acquisition unit collects the user life habit behavior fulfillment real-time image data through the robot equipped with a cloud lens; the user life habit behavior real-time fulfillment action analysis unit analyzes and processes the real-time action object information of the user life habit behavior fulfillment based on the user life habit behavior fulfillment real-time image data in combination with the Internet search platform to generate the user life habit real-time fulfillment action analysis data; the user life habit behavior real-time fulfillment status judgment unit determines the user life habit behavior based on the user life habit behavior fulfillment real-time image data. The real-time performance analysis data of the user's living habits and the content data of the user's living habits are used to judge the real-time performance status of the user's living habits, and generate the real-time performance status judgment data of the user's living habits; the different types of life habit behavior non-performance penalty scheme storage unit is used to store the different types of life habit behavior non-performance penalty scheme data; the current life habit behavior non-performance penalty scheme search unit searches for the penalty scheme for the non-performance status of the user's living habit behavior based on the real-time performance analysis data of the user's living habits and the different types of life habit behavior non-performance penalty scheme data, and generates the user's life habit behavior non-performance penalty scheme data;

[0048] The user life habit behavior fulfillment feedback management module includes a user life habit behavior fulfillment result construction unit and a user life habit behavior fulfillment result push unit;

[0049] The user life habit behavior fulfillment result construction unit constructs the user life habit behavior fulfillment result data based on the planned time period, planned content, fulfillment stage judgment results, fulfillment status analysis results and non-fulfillment penalty plan information of the user life habit behavior; the user life habit behavior fulfillment result push unit performs the user life habit behavior fulfillment result feedback operation based on the user life habit behavior fulfillment result data in combination with the user health management platform.

[0050] (3) Beneficial effects

[0051] The present invention provides an AI-based user health intelligent management method and system. It has the following beneficial effects:

[0052] 1. Accurately obtain the user's daily life habit behavior plan information online through the user health management platform dialog box, and at the same time, combine data analysis and processing to independently and efficiently separate the planned time period and planned items of the user's life habit behavior, so as to achieve scientific management efficiency of the user's life habit behavior in different time stages; use the time timing module to dynamically collect the current time point information and the user's life habit behavior planned time period parameters to accurately judge the user's life habit behavior performance stage, and realize accurate supervision of the user's life habit behavior performance time stage based on time characteristics, and improve the intelligence of the user's life habit behavior health management; accurately search and process the user's life habit behavior performance content information based on the user's life habit behavior performance stage judgment parameters combined with the intelligent search algorithm and the user's life habit behavior plan content parameters, and realize accurate search of the user's life habit behavior performance items in different time stages, so as to improve the convenience and applicability of the user's life habit behavior health management.

[0053] 2. By automatically and regularly executing reminders for user life habits based on the execution time period and content information of user life habits behaviors in combination with mobile terminals, the service quality of user life habits behavior health management can be improved; by dynamically collecting real-time image information of user life habits behavior execution through a robot equipped with a cloud camera and accurately analyzing the real-time action information of user life habits execution in combination with an Internet search platform, dynamic intelligent monitoring of user life habits behavior execution can be achieved; based on the analysis parameters of user life habits real-time execution action combined with artificial intelligence recognition algorithms and user life habits behavior execution content data, the real-time status of user life habits behavior execution can be intelligently evaluated to achieve intelligent supervision of the real-time execution status of user life habits behavior and improve the effect of user life habits behavior health management; based on the analysis parameters of user life habits real-time execution action combined with intelligent search algorithms and different types of life habits behavior non-execution penalty schemes based on big data storage, fine matching of user life habits behavior non-execution penalty schemes can be performed, which will help promote the quality of user life habits behavior execution and improve the scientific nature of user life habits behavior health management.

[0054] 3. By constructing the user's lifestyle behavior execution result parameters based on the planned time period, planned content, execution stage judgment results, execution status analysis results and non-execution penalty plan information of the user's lifestyle behavior, the digital collection of the user's lifestyle behavior execution results can be realized; based on the user's lifestyle behavior execution result parameters and combined with the user health management platform, the user's lifestyle behavior execution result feedback operation can be autonomously and visually executed to realize the intuitive and clear feedback of the lifestyle behavior execution results, thereby improving the quality of the user's lifestyle behavior health management. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of a module of an AI-based user health intelligent management system provided by the present invention;

[0056] Figure 2 This is a flowchart of an AI-based user health intelligent management method provided by the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0058] The embodiments of the AI-based user health intelligent management method and system are as follows:

[0059] Example 1:

[0060] See also Figure 1 - Figure 2 , an AI-based user health intelligent management method, the method comprises the following steps:

[0061] S1. Collecting user life habit behavior plan text data and classifying the user life habit behavior plan time period and plan content information to generate user life habit behavior plan time period data and user life habit behavior plan content data respectively;

[0062] S2. Collect the data at the current time point and compare it with the data of the user's life habit behavior plan time period to determine the implementation stage of the user's life habit behavior, and generate data for determining the implementation stage of the user's life habit behavior. If the implementation stage has not been reached, repeat step S2 until the implementation stage is reached.

[0063] S3. When the user reaches the implementation stage, searching and processing the implementation content of the specific implementation stage of the user's lifestyle behavior based on the user's lifestyle behavior implementation stage judgment data and the user's lifestyle behavior plan content data to generate the user's lifestyle behavior implementation content data;

[0064] S4. Execute user life habit behavior execution prompts based on the execution time period and execution content information of the user's life habit behavior, and collect real-time image data of the user's life habit behavior execution;

[0065] S5. Analyze and process the real-time action object information of the user's life habit behavior based on the real-time image data of the user's life habit behavior performance to generate real-time action analysis data of the user's life habit performance. Analyze and process the real-time status of the user's life habit behavior performance together with the user's life habit behavior performance content data to generate real-time status judgment data of the user's life habit behavior performance. If the performance is normal, continue to repeat steps S4 and S5 until the user fails to perform normally, thereby monitoring the user to complete the content of the life habit behavior plan in the current performance stage, and repeat steps S2, S3, S4, and S5 until the user fails to perform normally, or directly execute step S7 after the user has normally completed the content of the life habit behavior plan in all performance stages.

[0066] S6. When the behavior is not fulfilled normally, the user's habitual behavior is not fulfilled according to the real-time performance analysis data of the user's living habits and the penalty plan data of different types of habitual behavior failures. The penalty plan data of the user's habitual behavior failures is generated and the user is supervised to complete the content of the habitual behavior plan in the current performance stage. Repeat steps S2, S3, S4, and S5;

[0067] S7. Construct the user's life habit behavior execution result data and perform the user's life habit behavior execution result feedback task.

[0068] For further information, see Figure 1 - Figure 2 The steps for collecting the user's life habit behavior plan text data and classifying the user's life habit behavior plan time period and plan content information to generate the user's life habit behavior plan time period data and the user's life habit behavior plan content data are as follows:

[0069] S11, inputting the planned time period of the user's daily living habits and the text information of the living habits plan items corresponding to the planned time period online through the dialog box of the user health management platform, and generating the user's living habits plan text data Y;

[0070] S12, using the KD tree nearest neighbor search algorithm to perform classification search processing on the user's life habit behavior plan text data Y according to the time period and the life habit behavior plan content keywords corresponding to the time period, and generate the user's life habit behavior plan time period data set A and the user's life habit behavior plan content data A' respectively; where A = (a1, ..., a m ,…,a ε ), m=1,2,3,…,ε; where a mrepresents the user's life habit behavior plan time period data corresponding to the mth planned time period of the day, ε represents the maximum number of user's life habit behavior plan time periods; a m =[a m1 ,a m2 ], where a m1 and a m2 Respectively represent the user's life habit behavior plan time period data a m The minimum time data of the user's life habit behavior plan and the maximum time data of the user's life habit behavior plan, a m1 and a m2 The units are composed of year, month, day, hour and minute; A'=(a'1,…,a' m ,…,a' ε ), where a' m It represents the user's life habit behavior plan content data corresponding to the mth planned time period of the day. The user's life habit behavior plan content data includes rest plan content information, diet plan content information, exercise plan content information, reading plan content information, dance plan content information, painting plan content information, calligraphy plan content information, photography plan content information and musical instrument plan content information.

[0071] The current time point data is collected and combined with the user's life habit behavior planned time period data to determine the implementation stage of the user's life habit behavior, and the user's life habit behavior implementation stage judgment data is generated. If the implementation stage has not been reached, step S2 is repeated until the implementation stage is reached. The operation steps are as follows:

[0072] S21. Collect the current specific time point information online through the time timing module and generate the current time point data Z, where the unit of Z is composed of year, month, day, hour, and minute;

[0073] S22, compare the current time point data Z with the user life habit behavior plan time period data a in the user life habit behavior plan time period data set A. m Perform time value comparison and generate user life habit behavior execution stage judgment data A based on the time value comparison results jieduan ;

[0074] When Z∈a m When , it means that the user has reached the stage of fulfilling the habitual behavior, then the judgment data A of the stage of fulfilling the habitual behavior is output. jieduan To the implementation stage, output the user's life habit behavior plan time period data a that matches the current time point data Z at the same time m Corresponding planned time period information and planned time period number information;

[0075] when When , it means that the user has not yet reached the stage of fulfilling the habitual behavior, then the user's habitual behavior fulfillment stage judgment data A is output. jieduan The implementation stage has not yet been reached. At this time, step S2 is repeated until the user's living habit behavior implementation stage judgment data A jieduan To the implementation stage.

[0076] When it comes to the implementation stage, the implementation content of the specific implementation stage of the user's lifestyle behavior is searched and processed based on the user's lifestyle behavior implementation stage judgment data and the user's lifestyle behavior plan content data. The operation steps for generating the user's lifestyle behavior implementation content data are as follows:

[0077] S31, when the user's living habit behavior is performed, the judgment data A jieduan When it comes to the execution stage, the iterative deepening search algorithm is used to judge the user's living habits and behaviors during the execution stage. jieduan and the user's life habit behavior plan content data a' in the user's life habit behavior plan content data A' m Perform character matching on the planned time period number to search for the user's life habit behavior execution stage judgment data A jieduan Corresponding user lifestyle behavior plan content data a' m , and generate user life habits and behavior content data through data identification

[0078] Through the mutual cooperation between the user life habit behavior plan text information collection unit, the user life habit behavior plan time period generation unit and the user life habit behavior plan content generation unit, the user life habit behavior plan information of the day is accurately obtained online through the user health management platform dialog box. At the same time, combined with data analysis and processing, the user life habit behavior plan time period and plan item information are independently and efficiently separated, so as to realize the scientific management efficiency of user life habit behavior in different time stages; the current time point information collection unit and the user life habit behavior performance stage judgment unit cooperate with each other, and use the time timing module to dynamically collect the current time point information and the user life habit behavior plan time period parameters to accurately judge the user life habit behavior performance stage, so as to realize accurate supervision of the user life habit behavior performance time stage based on time characteristics, and improve the intelligence of user life habit behavior health management; the user life habit behavior performance content search unit, based on the user life habit behavior performance stage judgment parameters combined with the intelligent search algorithm and the user life habit behavior plan content parameters, accurately searches and processes the user life habit behavior performance content information, realizes accurate search of user life habit behavior performance items in different time stages, and improves the convenience and applicability of user life habit behavior health management.

[0079] For further information, see Figure 1 - Figure 2 Based on the execution time period and execution content information of the user's life habit behavior, the user's life habit behavior execution prompt operation is executed, and the operation steps for collecting the user's life habit behavior execution real-time image data are as follows:

[0080] S41, through the mobile terminal to the user's life habit behavior performance stage judgment data A jieduan Corresponding lifestyle behavior plan execution time period information and user lifestyle behavior execution content data The corresponding lifestyle behavior plan fulfillment content information executes a user lifestyle behavior fulfillment prompt operation, and the mobile terminal includes any one of a smart bracelet, a smart phone, and a tablet computer;

[0081] S42, when the user performs the life habit behavior execution operation, the robot is simultaneously used to carry out the cloud camera to judge the data A according to the user's life habit behavior execution stage. jieduan The corresponding life habit behavior plan execution time period information collects the real-time action image information of the user performing the life habit behavior online, and generates the user's life habit behavior execution real-time image data O.

[0082] Based on the real-time image data of the user's life habit behavior, the real-time action object information of the user's life habit behavior is analyzed and processed to generate the real-time analysis data of the user's life habit behavior. The real-time status judgment processing of the user's life habit behavior is performed together with the user's life habit behavior content data to generate the real-time judgment data of the user's life habit behavior. When the user performs normally, steps S4 and S5 are repeated until the user fails to perform normally, so as to supervise the user to complete the life habit behavior plan content of the current performance stage, and steps S2, S3, S4, and S5 are repeated until the user fails to perform normally, or the user completes the life habit behavior plan content of all performance stages normally and directly executes step S7. The operation steps are as follows:

[0083] S51. Importing the real-time image data O of the user's life habit behavior into a dialog box of an internet search platform to perform graphic-text translation processing on the real-time action object feature information of the user's life habit behavior, thereby generating real-time action analysis data P of the user's life habit, where the real-time action analysis data of the user's life habit includes rest action types, eating action types, exercise action types, reading action types, dancing action types, painting action types, calligraphy action types, photography action types, and musical instrument action types. The internet search platform includes any one of the Google search platform, the Baidu search platform, and the 360 ​​search platform.

[0084] S52, the user's living habits real-time performance action analysis data P and the user's living habits behavior performance content data Perform lifestyle behavior keyword matching, and generate user lifestyle behavior real-time performance status judgment data B based on the lifestyle behavior keyword matching results. The specific steps for generating user lifestyle behavior real-time performance status judgment data B are as follows:

[0085] S521, initialization, update the maximum number of iterations T and update the performance status to identify the location of the pelican population. The formula for updating the performance status to identify the location of the pelican population is: where R i,j Indicates the position of the i-th fulfillment status identification pelican in the j-th dimension, that is, the i-th fulfillment status identification pelican in dimension j is the user's life habit behavior fulfillment content data The position in the search space, Indicates a random integer for position adjustment, rand indicates a random number in the range [0,1], ψ j and ζ j Represents the user's life habits and behavior fulfillment content data in dimension j Search the search space for the upper and lower boundaries of the life habit behavior keywords that match the user's real-time life habit action analysis data P;

[0086] S522, exploration phase, the implementation of state identification pelican to determine the location of prey, that is, in the user's living habits behavior implementation content data Search the search space to find the location of the life habit behavior keyword prey that matches the user's life habit real-time execution action analysis data P, and model the execution state recognition pelican approaching prey strategy so that the algorithm can perform the user's life habit behavior content data The search space is scanned, and the life habit behavior keywords that match the user's life habit real-time execution action analysis data P are searched in the algorithm. The search space is randomly generated, and the formula for approximating the prey strategy is: where R′ i,j Indicates the i-th fulfillment status after the exploration phase update to identify the user's life habits and behavior fulfillment content data in dimension j of Pelican The position in the search space, M j Represents the searched life habit behavior keyword that matches the user life habit real-time execution action analysis data P, and the user life habit behavior execution content data in the jth dimension The position in the search space, Ξ M represents the objective function value of the searched life habit behavior keyword prey that matches the user's life habit real-time action analysis data P, Ξ i represents the objective function value of the i-th fulfillment state identification pelican;

[0087] S523, Development stage, Implementation status recognition Pelican uses the surface flying strategy to implement content data in user's living habits and behaviors Hunt out the life habit behavior keyword prey that matches the user's life habit real-time execution action analysis data P in the search space, and model the hunting behavior process of the execution state recognition pelican, so that the algorithm can identify the user's life habit behavior execution content data in the user's life habit behavior execution content data. Search the search space to find the most matching life habit behavior keyword prey with the user's life habit real-time execution action analysis data P, and perform state recognition. The calculation formula of the pelican's hunting behavior Where R i,j Indicates the i-th fulfillment status after the development phase update to identify the user's life habits and behavior fulfillment content data in dimension j of Pelican The position in the search space, rand is a random number in the range [0,1]; ω is a random integer of 0 or 2; t is the current number of iterations; T is the maximum number of iterations;

[0088] S524, when the algorithm meets the maximum number of iterations, the action analysis data P and the content data of the user's living habits are executed in real time according to the user's living habits. Performing lifestyle behavior keyword matching results to generate user lifestyle behavior real-time performance status judgment data B;

[0089] When P and If no matching of the lifestyle behavior keywords is successful, it means that the user has not properly fulfilled the current lifestyle behavior plan, and the user's lifestyle behavior real-time fulfillment status judgment data B is outputted, indicating that the user has not properly fulfilled the plan.

[0090] When P and If the matching of the life habit behavior keywords is successful, it means that the user is fulfilling the current life habit behavior plan content normally, and the user's life habit behavior real-time fulfillment status judgment data B is output to indicate normal fulfillment. At this time, steps S4 and S5 are repeated until the user's life habit behavior real-time fulfillment status judgment data B is not fulfilled normally, so as to supervise the user to complete the user's life habit behavior fulfillment content data. After the corresponding current stage of life habit behavior plan content is completed, repeat steps S2, S3, S4, and S5 until the user's life habit behavior real-time execution status judgment data B is that it is not performed normally, or the user has normally completed all the life habit behavior plan contents in the user's life habit behavior plan content data A′ and directly executes step S7.

[0091] When the behavior is not performed normally, the penalty scheme for the user's non-performance of habitual behavior is searched and processed based on the real-time performance analysis data of the user's living habits and the penalty scheme data for non-performance of different types of habitual behavior. The penalty scheme data for non-performance of the user's habitual behavior is generated, and the user is continuously supervised to complete the content of the habitual behavior plan for the current performance stage. The operation steps of steps S2, S3, S4, and S5 are repeated as follows:

[0092] S61, when the user's living habit behavior real-time performance status judgment data B is not performed normally, establish different types of living habit behavior non-performance penalty plan data set C = (c1, ..., c n ,…,c γ ), n=1,2,3,…,γ; where c n represents the penalty plan data for different types of lifestyle behaviors not fulfilled corresponding to the nth lifestyle behavior type, and γ represents the maximum number of lifestyle behavior types; lifestyle behavior types include resting, eating, exercising, reading, dancing, painting, calligraphy, photography, and musical instruments. The penalty plan data for different types of lifestyle behaviors not fulfilled represents the standard penalty plan text information set for users who fail to fulfill different types of lifestyle behaviors normally; the penalty plan data for different types of lifestyle behaviors not fulfilled include information on standing penalty plan, running plan, horse stance plan, reading aloud plan, and housework plan;

[0093] S62, using a bidirectional search algorithm to compare the user's life habit real-time performance analysis data P with the different types of life habit behavior non-performance penalty plan data c in the different types of life habit behavior non-performance penalty plan data set C. n Perform character matching on life habits and search for different types of life habits and behaviors that do not comply with penalty plan data c corresponding to the real-time action analysis data P of the user's life habits n , and generate data on user lifestyle behavior that fails to comply with the penalty plan through data identification At this time, continue to monitor the user's completion of user lifestyle behavior content data After the corresponding current stage of implementation of the life habit behavior plan content, repeat steps S2, S3, S4, and S5.

[0094] The user lifestyle behavior fulfillment content prompt unit automatically and regularly executes user lifestyle behavior fulfillment prompts based on the fulfillment time period and fulfillment content information of the user's lifestyle behavior in conjunction with the mobile terminal, thereby improving the service quality of user lifestyle behavior health management. The user lifestyle behavior fulfillment real-time image acquisition unit and the user lifestyle behavior fulfillment real-time action analysis unit cooperate with each other to dynamically collect real-time image information of the user's lifestyle behavior fulfillment through a robot-mounted cloud camera and accurately analyze the user's lifestyle behavior fulfillment real-time action information in conjunction with the Internet search platform, thereby realizing dynamic intelligent monitoring of the user's lifestyle behavior fulfillment. The user lifestyle behavior real-time fulfillment status judgment unit intelligently evaluates the user's lifestyle behavior fulfillment real-time status based on the user's lifestyle behavior real-time action analysis parameters combined with artificial intelligence recognition algorithms and user lifestyle behavior fulfillment content data, thereby realizing intelligent supervision of the user's lifestyle behavior real-time fulfillment status and improving the effectiveness of user lifestyle behavior health management. The current lifestyle behavior non-fulfillment penalty plan search unit finely matches the user's lifestyle behavior non-fulfillment status penalty plan based on the user's lifestyle behavior real-time action analysis parameters combined with intelligent search algorithms and different types of lifestyle behavior non-fulfillment penalty plans based on big data storage, which helps to promote the quality of user lifestyle behavior fulfillment and improve the scientific nature of user lifestyle behavior health management.

[0095] For further information, see Figure 1 - Figure 2 The steps for constructing user lifestyle behavior fulfillment result data and performing user lifestyle behavior fulfillment result feedback are as follows:

[0096] S71, the user's life habit behavior plan time period data set A, the user's life habit behavior plan content data A', the user's life habit behavior implementation stage judgment data A that the user's life habit behavior implementation stage judgment result is in the implementation stage jieduan , user life habit behavior real-time implementation status judgment data B and user life habit behavior non-implementation penalty plan data Combine the data to construct the user's life habit behavior performance result data K, where

[0097] S72. Transmit the user's lifestyle behavior fulfillment result data K to the user health management platform via the mobile communication network to perform the user's lifestyle behavior fulfillment result feedback operation.

[0098] Through the user life habit behavior execution result construction unit, the user life habit behavior execution result parameters are constructed based on the planned time period, planned content, execution stage judgment results, execution status analysis results and non-execution penalty plan information of the user's life habit behavior, thereby realizing the digital collection of the user's life habit behavior execution results; the user life habit behavior execution result push unit, based on the user life habit behavior execution result parameters and the user health management platform, autonomously and visually executes the user life habit behavior execution result feedback operation, thereby realizing intuitive and clear feedback on the life habit behavior execution results, and improving the quality of user life habit behavior health management.

[0099] Example 2:

[0100] See also Figure 1 - Figure 2 , an AI-based user health intelligent management system, used to implement an AI-based user health intelligent management method, the system includes a user life habit behavior stage prompt module, a user life habit behavior fulfillment management module, and a user life habit behavior fulfillment feedback management module;

[0101] The user life habit behavior stage prompt module includes a user life habit behavior plan text information collection unit, a user life habit behavior plan time period generation unit, a user life habit behavior plan content generation unit, a current time point information collection unit, a user life habit behavior fulfillment stage judgment unit, and a user life habit behavior fulfillment content search unit;

[0102] a user life habit behavior plan text information collection unit, which collects user life habit behavior plan text data through the user health management platform dialog box; a user life habit behavior plan time period generation unit, which searches and processes the user life habit behavior plan time period information based on the user life habit behavior plan text data, and generates user life habit behavior plan time period data; a user life habit behavior plan content generation unit, which searches and processes the user life habit behavior plan content information based on the user life habit behavior plan text data, and generates user life habit behavior plan content data; a current time point information collection unit, which collects current time point data through a time timing module; a user life habit behavior performance stage judgment unit, which judges and processes the performance stage of the user life habit behavior based on the current time point data and the user life habit behavior plan time period data, and generates user life habit behavior performance stage judgment data; a user life habit behavior performance content search unit, which searches and processes the performance content of the specific performance stage of the user life habit behavior based on the user life habit behavior performance stage judgment data and the user life habit behavior plan content data, and generates user life habit behavior performance content data;

[0103] The user life habit behavior fulfillment management module includes a user life habit behavior fulfillment content prompt unit, a user life habit behavior fulfillment real-time image acquisition unit, a user life habit behavior real-time fulfillment action analysis unit, a user life habit behavior real-time fulfillment status judgment unit, a different type of life habit behavior non-fulfillment penalty plan storage unit, and a current life habit behavior non-fulfillment penalty plan search unit;

[0104] a user life habit behavior fulfillment content prompting unit, which performs user life habit behavior fulfillment prompting operations based on the fulfillment time period and fulfillment content information of the user life habit behavior in combination with the mobile terminal; a user life habit behavior fulfillment real-time image acquisition unit, which collects user life habit behavior fulfillment real-time image data through a robot-mounted cloud camera; a user life habit behavior real-time fulfillment action analysis unit, which analyzes and processes the real-time action object information of the user life habit behavior fulfillment based on the user life habit behavior fulfillment real-time image data in combination with an Internet search platform, and generates user life habit real-time fulfillment action analysis data; a user life habit behavior real-time fulfillment status judgment unit, which judges and processes the real-time fulfillment status of the user life habit behavior based on the user life habit real-time fulfillment action analysis data and the user life habit behavior fulfillment content data, and generates user life habit behavior real-time fulfillment status judgment data; a different type of life habit behavior non-fulfillment penalty plan storage unit, which is used to store different types of life habit behavior non-fulfillment penalty plan data; a current life habit behavior non-fulfillment penalty plan search unit, which searches and processes the penalty plan for the user life habit behavior non-fulfillment status based on the user life habit real-time fulfillment action analysis data and different types of life habit behavior non-fulfillment penalty plan data, and generates user life habit behavior non-fulfillment penalty plan data;

[0105] The user life habit behavior fulfillment feedback management module includes a user life habit behavior fulfillment result construction unit and a user life habit behavior fulfillment result push unit;

[0106] The user life habit behavior execution result construction unit constructs the user life habit behavior execution result data based on the planned time period, planned content, execution stage judgment results, execution status analysis results and non-execution penalty plan information of the user life habit behavior; the user life habit behavior execution result push unit performs the user life habit behavior execution result feedback task based on the user life habit behavior execution result data and the user health management platform.

[0107] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An AI-based user health intelligent management method, characterized in that: The method comprises the following steps: S1. Collecting text data of the user's life habit behavior plan and generating the user's life habit behavior plan time period data and the user's life habit behavior plan content data; S2. Collect data at the current time point to generate data to determine the user's lifestyle behavior implementation stage. If the implementation stage has not yet been reached, repeat step S2 until the implementation stage is reached. S3. When the implementation stage is reached, the implementation content of the specific implementation stage of the user's lifestyle behavior is searched and processed to generate the implementation content data of the user's lifestyle behavior; S4, executing the user's life habit behavior execution prompt operation, collecting the user's life habit behavior execution real-time image data; S5. Analyze and process the real-time action object information of the user's life habit behavior performance to generate real-time analysis data of the user's life habit performance, and perform real-time status judgment processing on the user's life habit behavior performance to generate real-time status judgment data on the user's life habit behavior performance. If the performance is normal, continue to repeat steps S4 and S5 until the user fails to perform normally, thereby monitoring the user to complete the content of the life habit behavior plan in the current performance stage, and repeat steps S2, S3, S4, and S5 until the user fails to perform normally, or directly execute step S7 after the user has normally completed the content of the life habit behavior plan in all performance stages; S6. When the user fails to fulfill the habitual behavior, a penalty plan search is performed to generate penalty plan data for the user's habitual behavior failure, and the user is supervised to complete the content of the habitual behavior plan in the current fulfillment phase, and then steps S2, S3, S4, and S5 are repeated. S7. Construct the user's life habit behavior execution result data and perform the user's life habit behavior execution result feedback task.

2. The AI-based user health intelligent management method according to claim 1, characterized in that: Said S1 comprises the following steps: S11, inputting the planned time period of the user's daily living habits and the text information of the living habits plan items corresponding to the planned time period online through the dialog box of the user health management platform, and generating the user's living habits plan text data Y; S12, using the KD tree nearest neighbor search algorithm to perform classification search processing on the user's life habit behavior plan text data Y according to the time period and the life habit behavior plan content keywords corresponding to the time period, and generate the user's life habit behavior plan time period data set A and the user's life habit behavior plan content data A' respectively; where A = (a1, ..., a m ,…,a ε ), m=1,2,3,…,ε; where a m represents the user's life habit behavior plan time period data corresponding to the mth planned time period of the day, ε represents the maximum number of user's life habit behavior plan time periods; a m =[a m1 ,a m2 ], where a m1 and a m2 Respectively represent the user's life habit behavior plan time period data a m The minimum time data of the user's life habit behavior plan and the maximum time data of the user's life habit behavior plan, a m1 and a m2 The units are composed of year, month, day, hour and minute; A'=(a'1,…,a' m ,…,a' ε ), where a' m Indicates the user's life habit behavior plan content data corresponding to the mth planned time period of the day.

3. The AI-based user health intelligent management method according to claim 2, characterized in that: The S2 comprises the following steps: S21. Collect the current specific time point information online through the time timing module and generate the current time point data Z, where the unit of Z is composed of year, month, day, hour, and minute; S22, the Z and the a in the A m Perform time value comparison and generate user life habit behavior execution stage judgment data A based on the time value comparison results jieduan ; When Z∈a m When the table outputs the A jieduan To the execution stage, output the a that matches the Z at the same time m Corresponding planned time period information and planned time period number information; when When the A jieduan If the execution stage has not yet been reached, then step S2 is repeated until the A jieduan To the implementation stage.

4. The AI-based user health intelligent management method according to claim 3, characterized in that: The S3 includes the following steps: S31, when the A jieduan When it comes to the execution stage, the iterative deepening search algorithm is used to jieduan and the a' in the m Perform character matching on the planned time period number and search for the A jieduan The corresponding a' m , and generate user life habits and behavior content data through data identification 5. The AI-based user health intelligent management method according to claim 4, characterized in that: The S4 comprises the following steps: S41, through the mobile terminal, the A jieduan The corresponding life habit behavior plan implementation time period information and the The corresponding lifestyle behavior plan execution content information executes the user's lifestyle behavior execution prompt task; S42, when the user performs the task of performing the life habit behavior, the robot is equipped with a cloud camera according to the A jieduan The corresponding life habit behavior plan execution time period information collects the real-time action image information of the user performing the life habit behavior online, and generates the user's life habit behavior execution real-time image data O.

6. The AI-based user health intelligent management method according to claim 5, characterized in that: The S5 comprises the following steps: S51, importing the O into the dialog box of the Internet search platform to perform graphic translation processing on the real-time action object feature information of the user's living habit behavior, and generating the user's living habit real-time action analysis data P; S52, the P and the Performing lifestyle behavior keyword matching, generating B based on the lifestyle behavior keyword matching result, and executing the specific operation steps of generating the user lifestyle behavior real-time performance status judgment data B are as follows: S521, initialization, updating the maximum number of iterations T and updating the execution status to identify the location of the pelican population; S522, exploration phase, the pelican determines the location of the prey, that is, in the Search the search space to find the location of the prey with the life habit behavior keyword that matches the P, and model the performance state recognition pelican approaching prey strategy so that the algorithm can The search space is scanned, and the prey with the life habit behavior keyword matching the P is searched in the algorithm. The search space is randomly generated; S523, development phase, implementation status identification Pelican adopts water flight strategy in the Hunt out the prey with the life habit behavior keyword matching the P in the search space, and model the hunting behavior process of the pelican in the performance state recognition so that the algorithm Search the search space to find the life habit behavior keyword prey that best matches the P; S524, when the algorithm meets the maximum number of iterations, according to the P and the Performing lifestyle behavior keyword matching results to generate user lifestyle behavior real-time performance status judgment data B; When P and If no matching of the keywords of the lifestyle behavior is successful, then the output of B is that the above-mentioned B is not performed normally; When P and If the keyword matching of lifestyle behavior is successful, the output of B is normal execution. At this time, the steps S4 and S5 are repeated until B is not normally executed to supervise the user to complete the After the corresponding current stage life habit behavior plan content is implemented, steps S2, S3, S4, and S5 are repeated until B is not implemented normally, or the user completes all the life habit behavior plan contents in A' in the implementation stage normally and then directly executes step S7.

7. The AI-based user health intelligent management method according to claim 6, characterized in that: The S6 comprises the following steps: S61. When B is not performed normally, establish a data set of penalty plans for different types of life habits and behaviors not performed normally, C = (c1, ..., c n ,…,c γ ), n=1,2,3,…,γ; where c n represents the data of different types of non-compliance penalty plans for lifestyle behaviors corresponding to the nth lifestyle behavior type, and γ represents the maximum number of lifestyle behavior types; S62, using a bidirectional search algorithm to compare the P with the c in the C n Perform character matching of life habits and search for the c corresponding to the P n , and generate data on user lifestyle behavior that fails to comply with the penalty plan through data identification At this point, continue to supervise the user to complete the After the corresponding current stage of implementation of the life habit behavior plan content, repeat steps S2, S3, S4, and S5.

8. The AI-based user health intelligent management method according to claim 7, characterized in that: The S7 comprises the following steps: S71, the A, the A', the user's life habit behavior execution stage judgment result is the A in the execution stage jieduan , said B and said Combining data to construct user life habit behavior fulfillment result data K; S72. Transmit the K to the user health management platform via the mobile communication network to perform feedback on the user's lifestyle behavior performance results.

9. An AI-based user health intelligent management system, used to implement the AI-based user health intelligent management method according to any one of claims 1 to 8, characterized in that: The system includes a user life habit behavior stage prompt module, a user life habit behavior fulfillment management module, and a user life habit behavior fulfillment feedback management module.

Images