A personalized intelligent meal distribution method and system based on user conditions

Abstract

The present invention discloses a personalized intelligent meal distribution method and system based on user conditions, which relates to the fields of personalized nutrition science and intelligent food technology. The method includes collecting comprehensive user health data and multi-dimensional real-time user context data, and preprocessing the two types of data; constructing a user nutrition demand prediction model through a long short-term memory network model combined with a multi-layer perceptron model; analyzing the preprocessed multi-dimensional real-time user context data to obtain an adjustment coefficient affecting nutrition demand, and adjusting the user's personalized nutrition demand through a linear programming algorithm; based on the adjusted user's personalized nutrition demand and combined with the user's taste preferences, analyzing the matching degree of each food ingredient combination, and performing customized adjustment through a particle swarm optimization algorithm to generate a user's personalized recipe. By optimizing the food ingredient combination through the particle swarm optimization algorithm, personalized intelligent meal distribution is successfully achieved, significantly improving the scientificity, accuracy, and user satisfaction of personalized meal distribution.

Description

Technical Field

[0001] The present invention relates to the fields of personalized nutrition and intelligent food technology, and in particular to a personalized intelligent meal - matching method and system based on user conditions. Background Art

[0002] With the continuous improvement of people's health awareness, personalized nutrition meal - matching has gradually become an important research direction in the fields of modern nutrition and food technology. In recent years, personalized nutrition has been gradually developed, and related technologies have gradually shifted from static calculations based on traditional nutrition demand models to dynamic optimization using advanced technologies such as big data, artificial intelligence, and machine learning. In particular, the emergence of the Long Short - Term Memory (LSTM) network model has provided a new method for processing and analyzing user health data and context data. The LSTM model can accurately predict the changes in users' nutritional needs through deep learning of time - series data, thus providing more accurate guidance for personalized nutrition meal - matching. At the same time, the introduction of multi - dimensional real - time user context data can better reflect the fluctuations in users' nutritional needs in different situations. With the application of big data technology, it is possible to collect and process users' health data and behavior data in real time, providing strong support for personalized intelligent meal - matching.

[0003] However, there are still some deficiencies in the implementation of personalized meal - matching in the existing technology. First of all, most of the existing technologies rely on single health data or static nutrition demand prediction models, and fail to fully consider the fluctuations in users' nutritional needs in different situations, which makes it difficult to achieve the best personalized meal - matching effect. Secondly, although some technologies try to combine users' taste preferences for personalized recommendations, they lack in - depth analysis of food ingredient combinations and matching degrees, resulting in poor balance between nutrition and taste in the generated recipes and being difficult to meet users' personalized needs. Therefore, there are still certain limitations in the existing technology in aspects such as combining health data with context data, accurately predicting nutritional needs, and dynamically adjusting personalized recipes. Summary of the Invention

[0004] In view of the above - mentioned existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a personalized intelligent meal - matching method based on user conditions, which solves the problems of insufficient consideration of user context data and lack of food ingredient combination analysis.

[0006] To solve the above - mentioned technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a personalized intelligent meal - matching method based on user conditions, which includes collecting comprehensive user health data and multi - dimensional real - time user context data, and pre - processing the two types of data; constructing a user nutritional requirement prediction model based on a long - short - term memory network model, using the pre - processed comprehensive user health data to obtain user nutritional requirement values, and generating personalized nutritional requirements for users; analyzing the pre - processed multi - dimensional real - time user context data to obtain adjustment coefficients affecting nutritional requirements, and adjusting the personalized nutritional requirements of users through a linear programming algorithm; based on the adjusted personalized nutritional requirements of users and combined with user taste preferences, analyzing the matching degree of each ingredient combination, and performing customized adjustment through a particle swarm optimization algorithm to generate personalized recipes for users.

[0008] As a preferred embodiment of the personalized intelligent meal - matching method based on user conditions according to the present invention, wherein: the comprehensive user health data includes user basic information, user physiological health data, and user activity level;

[0009] The multi - dimensional real - time user context data includes real - time activity time data, real - time activity location data, and real - time activity environment data;

[0010] The pre - processing includes data cleaning and standardization processing.

[0011] As a preferred embodiment of the personalized intelligent meal - matching method based on user conditions according to the present invention, wherein: constructing a user nutritional requirement prediction model based on a long - short - term memory network model, the specific steps are as follows,

[0012] Converting the pre - processed historical comprehensive user health data into a time - series format through a sliding window method;

[0013] Using the long - short - term memory network model as the basic model;

[0014] The input layer receives the pre - processed historical comprehensive user health data converted into a time - series format;

[0015] The LSTM layer, through memory units and gate mechanisms, the LSTM model selectively remembers and forgets information, capturing long - term dependencies in the time series;

[0016] The fully - connected layer extracts features and reduces dimensions based on the output of the LSTM layer, outputting preliminary features;

[0017] Combining with a multi - layer perceptron model, the input layer receives the preliminary features output by the LSTM model;

[0018] The concatenation layer concatenates the preliminary features with other relevant factors together;

[0019] The MLP layer inputs the concatenated vectors into a single-layer MLP for further feature combination and transformation;

[0020] The output layer outputs the final result;

[0021] Finally, a user nutritional requirement prediction model is obtained.

[0022] As a preferred embodiment of the personalized intelligent meal distribution method based on user conditions according to the present invention, wherein: the use of a health energy calculator in combination with augmented reality (AR) technology, and the application of the Harris-Benedict algorithm to obtain the user's metabolic rate, combined with the preprocessed comprehensive user health data, through the user nutritional requirement prediction model, to obtain the user's nutritional requirement value and generate the user's personalized nutritional requirements. The specific steps are as follows.

[0023] Use a health energy calculator in combination with augmented reality (AR) technology to simulate the user performing lightweight activities in a virtual environment, capture individual differences by observing the user's physical reactions in different situations, and create a unique health profile for each user;

[0024] Through the activity logs provided by the user, and using a sports bracelet to collect and analyze the user's activity level, the average activity intensity of the user is obtained;

[0025] Based on the user's average activity intensity, combined with the user's health profile, adjust according to the proportion of the time occupied by the user in different activity types and the corresponding metabolic equivalent value ratio to obtain a personalized activity factor;

[0026] Based on the user's health profile, through a health energy calculator and applying the Harris-Benedict algorithm, obtain the basal metabolic rate, and combine with the personalized activity factor to obtain the user's metabolic rate;

[0027] Based on the user's metabolic rate and the preprocessed comprehensive user health data, through the user nutritional requirement prediction model, obtain the user's nutritional requirement value.

[0028] As a preferred embodiment of the personalized intelligent meal distribution method based on user conditions according to the present invention, wherein: the generation of the user's personalized nutritional requirements. The specific steps are as follows.

[0029] Based on historical user health data, analyze through a clustering analysis algorithm and set a health threshold ;

[0030] Compare the user's nutritional requirement value with the health threshold to evaluate the user's health status;

[0031] When it indicates that the user's health status is poor and multiple nutrients need to be ingested;

[0032] When When it indicates that the user's health condition is good and there is no need to intake nutrients;

[0033] Based on the user's health condition and the nutrients the user needs, a genetic algorithm is combined with nutritional theory to generate the user's personalized nutritional requirements.

[0034] As a preferred solution of the personalized intelligent meal - matching method based on the user's condition described in the present invention, among them: analyzing the pre - processed multi - dimensional real - time user context data to obtain the adjustment coefficients affecting nutritional requirements, and adjusting the user's personalized nutritional requirements through a linear programming algorithm. The specific steps are as follows:

[0035] Analyze the real - time activity time data through time - series analysis method, and combine with kinematic principles to determine the activity intensity of each time period;

[0036] Use the weight - allocation method to convert different activity intensities into adjustment coefficients of activity intensity affecting nutritional requirements;

[0037] Analyze the real - time activity location data and the user's physiological health data through data - mining method, and map them into adjustment coefficients of activity location affecting nutritional requirements;

[0038] Combine the real - time activity environment data with the influence laws of human body heat consumption and human body water requirement, and calculate the adjustment coefficients of activity environment affecting nutritional requirements through regression analysis method;

[0039] Set the minimum - value and maximum - value constraint conditions according to the user's basic physiological needs and health condition;

[0040] The linear programming algorithm uses the adjustment coefficients of activity intensity affecting nutritional requirements, the adjustment coefficients of activity location affecting nutritional requirements, and the adjustment coefficients of activity environment affecting nutritional requirements, and adjusts the user's personalized nutritional requirements through the minimum - value and maximum - value constraint conditions.

[0041] As a preferred solution of the personalized intelligent meal - matching method based on the user's condition described in the present invention, among them: based on the adjusted user's personalized nutritional requirements and combined with the user's taste preferences, analyze the matching degree of each food ingredient combination, and perform customized adjustment through a particle - swarm optimization algorithm to generate the user's personalized recipe. The specific steps are as follows:

[0042] Obtain the user's taste preferences through the user's diet history records, feedback scores, and taste - preference questionnaires;

[0043] Combine the adjusted user's personalized nutritional requirements with the user's taste preferences, and gradually search for food ingredients and dish combinations that meet the requirements through a particle - swarm optimization algorithm to form a food ingredient meal - matching plan;

[0044] Analyze the matching degree between the nutritional value of ingredient combinations in each ingredient meal plan and the user's taste preferences through the Pearson correlation coefficient;

[0045] Perform a weighted sum of the matching degrees of each ingredient combination to obtain an ingredient combination evaluation score, and screen out the ingredients that meet the user's personalized needs through the ascending sorting method to form a user's personalized recipe.

[0046] In a second aspect, the present invention provides a personalized intelligent meal planning system based on user conditions, including a data collection module, a model construction module, a model prediction module, a nutritional requirement adjustment module, and a recipe formulation module; the data collection module is used to collect comprehensive user health data and multi-dimensional real-time user context data, and preprocess the two types of data; the model construction module constructs a user nutritional requirement prediction model through a long short-term memory network model combined with a multi-layer perceptron model; the model prediction module is used to use a health energy calculator combined with augmented reality (AR) technology, and apply the Harris-Benedict algorithm to obtain the user's metabolic rate, and combine the preprocessed comprehensive user health data, and through the user nutritional requirement prediction model, obtain the user's nutritional requirement value and generate the user's personalized nutritional requirements; the nutritional requirement adjustment module is used to analyze the preprocessed multi-dimensional real-time user context data to obtain an adjustment coefficient for influencing nutritional requirements, and adjust the user's personalized nutritional requirements through a linear programming algorithm; the recipe formulation module, based on the adjusted user's personalized nutritional requirements and combined with the user's taste preferences, analyzes the matching degree of each ingredient combination, and performs customized adjustment through a particle swarm optimization algorithm to generate the user's personalized recipe.

[0047] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, it implements any step of the personalized intelligent meal planning method based on user conditions as described in the first aspect of the present invention.

[0048] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, it implements any step of the personalized intelligent meal planning method based on user conditions as described in the first aspect of the present invention.

[0049] The beneficial effects of the present invention are as follows: Through the accurate prediction of the user's nutritional requirements by combining the long short-term memory (LSTM) network model with the multi-layer perceptron model, and the optimization of ingredient combinations by the particle swarm optimization algorithm, personalized intelligent meal planning is successfully realized. The beneficial effects are that it can dynamically adjust nutritional requirements based on the user's health status and real-time context data, optimize ingredient combinations in combination with the user's taste preferences, so as to generate recipes that meet both nutritional requirements and personalized tastes, significantly improving the scientificity, accuracy, and user satisfaction of personalized meal planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0051] Figure 1 It is a flowchart of the personalized intelligent meal distribution method based on user status in Embodiment 1.

[0052] Figure 2 It is a schematic diagram of the personalized intelligent meal distribution system based on user status in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the drawings of the specification.

[0054] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0055] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0056] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides a personalized intelligent meal distribution method based on user status, including the following steps:

[0057] S1. The comprehensive user health data includes the user's basic information, the user's physiological health data, and the user's activity level;

[0058] It should be noted that the above content has obtained the consent of the user and is used for legal purposes.

[0059] S1.1. The multi-dimensional real-time user context data includes real-time activity time data, real-time activity location data, and real-time activity environment data.

[0060] S1.2. The preprocessing includes data cleaning and standardization processing.

[0061] It should be noted that the data cleaning and standardization processes for the comprehensive user health data and multi-dimensional real-time user context data are as follows:

[0062] Data cleaning: First, conduct a comprehensive data review to identify and handle missing values. This usually involves analyzing the missing patterns to decide whether to delete the records with missing values or adopt appropriate imputation methods, such as using the mean, median, or prediction models for imputation. Then, correct logical errors and outliers by setting reasonable threshold ranges to detect and correct extreme values that do not conform to physiological common sense or domain knowledge, ensuring data consistency. Especially when the data comes from multiple sources, it is necessary to check whether the date formats, measurement units, etc. are consistent and resolve any detected inconsistencies. Finally, remove completely duplicate or logically redundant records;

[0063] Standardization process: Adopt the Z-score standardization method, that is, calculate the difference between each data point and the overall mean and then divide by the standard deviation to convert the data into the form of a standard normal distribution with a mean of 0 and a standard deviation of 1. This method is particularly suitable for cases where features have different magnitudes.

[0064] S2. Build a user nutrition demand prediction model based on the long short-term memory network model.

[0065] S2.1. Convert the preprocessed historical comprehensive user health data into a time series format through the sliding window method.

[0066] It should be noted that first, determine a fixed-size time window, which will slide step by step on the preprocessed historical comprehensive user health data set according to the set step size. Each time it slides, capture all the data points within the window as a sample; then, organize the data within each time window and convert it into a time series format suitable for input into the long short-term memory (LSTM) model, that is, each sample contains the data points arranged in order from the earliest to the latest time point; in this way, by continuously moving the window, a series of time series samples covering the entire time range can be created.

[0067] S2.2. Use the long short-term memory network model as the base model;

[0068] The input layer receives the preprocessed historical comprehensive user health data converted into a time series format;

[0069] The LSTM layer, through memory units and gate mechanisms, the LSTM model selectively remembers and forgets information to capture long-term dependencies in the time series;

[0070] The fully connected layer extracts features and reduces dimensions based on the output of the LSTM layer and outputs preliminary features;

[0071] Combined with a multi - layer perceptron model, the input layer receives the preliminary features output by the LSTM model;

[0072] The concatenation layer concatenates the preliminary features with other relevant factors;

[0073] The MLP layer inputs the concatenated vector into a single - layer MLP for further feature combination and transformation;

[0074] The output layer outputs the final result;

[0075] Finally, a user nutritional requirement prediction model is obtained.

[0076] It should be noted that based on the long short - term memory (LSTM) model and combined with the multi - layer perceptron (MLP) model, a user nutritional requirement prediction model is constructed. This method effectively captures the time - series features and long - term dependencies in the user's comprehensive health data through the LSTM layer, especially providing more accurate modeling capabilities for health parameters that change over time. Subsequently, the MLP is used to further combine and transform the preliminary features output by the LSTM, and after concatenating with other relevant factors, optimization processing is carried out, realizing the in - depth mining and integration of complex and multi - dimensional health data. Compared with the existing technology, this method not only enhances the understanding and prediction accuracy of individual differences and dynamic living habits, but also improves the comprehensiveness and precision of personalized nutritional requirement prediction by introducing additional relevant factors, providing more scientific and reasonable diet suggestions for users.

[0077] S3. Use a health energy calculator combined with augmented reality (AR) technology, and apply the Harris - Benedict algorithm to obtain the user's metabolic rate. Combine it with the pre - processed user's comprehensive health data, and through the user nutritional requirement prediction model, obtain the user's nutritional requirement value and generate the user's personalized nutritional requirement.

[0078] S3.1. Use a health energy calculator and combine it with augmented reality (AR) technology to simulate the user performing lightweight activities in a virtual environment, capture individual differences by observing the user's physical reactions in different scenarios, and create a unique health profile for each user.

[0079] It should be noted that by means of augmented reality (AR) technology, personalized virtual activity scenarios are created, allowing users to perform lightweight activities in these scenarios; the system real - time monitors and records changes in physiological indicators such as the user's heart rate and blood pressure, and analyzes their energy consumption and body responses in specific activities; finally, based on the collected data, a health profile containing detailed physiological information and activity responses is constructed for each user.

[0080] S3.2. Analyze the user's activity level collected by the activity log provided by the user and the sports bracelet to obtain the user's average activity intensity.

[0081] It should be noted that, first, the daily activity information of the user is recorded, such as the activity type, duration, and intensity; then, the sports bracelet monitors and transmits the user's physiological and motion data in real time, such as the number of steps, heart rate, and calories consumed; then, these data are integrated, and comprehensive statistical analysis and machine learning algorithms are applied to calculate the average activity intensity of different activity types, reflecting the overall activity level of the user over a period of time; finally, based on these analysis results, the average activity intensity of the user is determined.

[0082] S3.3. Based on the average activity intensity of the user, combined with the user's health record, adjust according to the proportion of time the user spends on different activity types and the proportion of their corresponding metabolic equivalent values to obtain a personalized activity factor.

[0083] It should be noted that, first, analyze the proportion of time the user spends on different activity types and perform weighted calculations based on the metabolic equivalent (METs) values of each activity type; then, adjust the average activity intensity of the user according to the time proportion of these activity types and their corresponding METs values, so as to obtain a personalized activity factor that reflects the user's lifestyle and health status.

[0084] S3.4. Based on the user's health record, obtain the basal metabolic rate through a health energy calculator and using the Harris-Benedict algorithm, and combine the personalized activity factor to obtain the user's metabolic rate.

[0085] It should be noted that, first, use the Harris-Benedict algorithm through a health energy calculator to calculate the basal metabolic rate (BMR) according to the user's age, gender, weight, and height, and then combine the personalized activity factor calculated according to the proportion of time the user spends on different activity types and their corresponding metabolic equivalent values to adjust the basal metabolic rate to reflect the user's actual daily activity level, so as to obtain the user's metabolic rate that accurately reflects the personal lifestyle and health status.

[0086] S3.5. Based on the user's metabolic rate and the preprocessed comprehensive user health data, obtain the user's nutritional requirement value through the user nutritional requirement prediction model, and the expression is:

[0087] ;

[0088] where is the user's nutritional requirement value at time point ; is the weight parameter of the user's nutritional requirement value; is the function symbol; means processing with model; is the user's metabolic rate; It is a fully connected layer operation, is for the time point of the preprocessed comprehensive user health data processing result, is the time point of the preprocessed comprehensive user health data, is the weight parameter for adjusting the user's metabolic rate, is the adjustment coefficient for controlling the influence intensity of the periodic component in the denominator on the output result, represents the influence of the user's metabolic rate on the periodic factor, is the angular frequency parameter, is the phase offset.

[0089] It should be noted that the above expression processes time series data by combining the long short-term memory network (LSTM) and the multi-layer perceptron (MLP) model, and incorporates the influence of the user's metabolic rate ( ) and its personalized activity factor. At the same time, it considers the regulatory effect of periodic factors on nutritional requirements. Compared with the existing technology, this method can not only capture the long-term dependencies in the comprehensive user health data, but also realizes a more refined and personalized assessment of nutritional requirements by introducing the metabolic rate and the periodic component adjustment coefficient ( ), emphasizes the importance of dynamic changes and individual differences, and provides a more accurate and customized user nutritional requirement value.

[0090] S4. Generate the user's personalized nutritional requirements.

[0091] S4.1. Based on the historical user health data, analyze it through the clustering analysis algorithm and set the health threshold .

[0092] It should be noted that first, collect and organize the historical user health data of the user, then apply the clustering analysis algorithm (such as K-means or hierarchical clustering) to group the historical user health data to identify user groups with similar health characteristics; then, calculate the statistical summary (such as mean, standard deviation) of the core health indicators for each cluster, and set the corresponding health threshold for each cluster according to these statistical results .

[0093] S4.2. Compare the user's nutritional requirement value with the health threshold to evaluate the user's health status;

[0094] When , it indicates that the user's health status is poor and multiple nutrients need to be ingested;

[0095] When When it indicates that the user's health condition is good and there is no need to intake nutrients.

[0096] It should be noted that when it indicates that the user's current nutritional requirement value exceeds the set health threshold, which usually indicates that there may be problems with the user's health condition or the user is in a sub-healthy state. In this case, it is necessary to increase the intake of various nutrients specifically to improve the health condition. At this time, foods rich in specific vitamins, minerals, and other essential nutrients will be recommended to help the user restore to an ideal health level;

[0097] When it means that the user's nutritional requirement value is below the health threshold, reflecting that the user's current health condition is relatively ideal and there is no need to increase the intake of nutrients additionally. This means that the user's body can obtain sufficient nutritional support under the existing diet structure to maintain normal physiological functions and good health conditions. Therefore, it is recommended to maintain the existing balanced diet and there is no need to specifically adjust the diet plan or increase additional nutritional supplements.

[0098] S4.3. Generate the user's personalized nutritional requirements based on the user's health condition and the nutrients the user needs through the genetic algorithm combined with nutritional theory.

[0099] It should be explained that first, according to the user's health data and the evaluation results of nutritional requirements, the optimization objectives and constraints are defined, such as the intake range of essential nutrients, calorie restrictions, etc.; then, the genetic algorithm is used to perform encoding, selection, crossover, and mutation operations on various nutritional combinations to simulate the natural selection process to search for the optimal solution space; during this process, the nutritional theory is combined to ensure that the generated nutritional requirement plan conforms to the dietary guidelines and the principle of nutritional balance; finally, a set of personalized nutritional requirements that meet the user's specific health needs and optimize the nutrient ratio are output.

[0100] S5. Analyze the preprocessed multi-dimensional real-time user context data to obtain the adjustment coefficients affecting nutritional requirements, and adjust the user's personalized nutritional requirements through the linear programming algorithm.

[0101] S5.1. Analyze the real-time activity time data through the time series analysis method, and combine the kinematics principle to determine the activity intensity of each time period.

[0102] It should be explained that the time series analysis technology (such as ARIMA model or exponential smoothing method) is applied to identify the activity patterns and trends; then, the kinematics principle, such as parameters like speed and acceleration, is used to quantify the intensity levels of different activity types; then, these physical activity indicators are mapped to the predefined activity intensity levels to determine the specific activity intensity of the user in each time period.

[0103] S5.2. Use a weight distribution method to convert different activity intensities into adjustment coefficients that affect nutritional requirements according to the activity intensity.

[0104] It should be noted that first, determine the basic activity intensity level according to the physical characteristics and metabolic equivalents (METs) of the activity type; then, set corresponding weights for each basic activity intensity level, and these weights reflect the relative influence degree of this activity intensity on nutritional requirements; then, based on the user's actual activity data, calculate an adjustment coefficient that affects nutritional requirements through the method of weighted average.

[0105] S5.3. Analyze the real-time activity location data and user's physiological health data through data mining methods, and map them into adjustment coefficients that affect nutritional requirements according to the activity location.

[0106] It should be noted that first, integrate the user's real-time activity location data with the corresponding physiological health indicators (such as heart rate, blood pressure, etc.), and use data mining techniques (such as cluster analysis or regression models) to identify the association patterns between different locations and specific physiological responses; then, evaluate the actual impacts of the user's activities at different locations on human energy consumption and nutrient absorption under these association patterns; then, quantify these impacts according to the analysis results and convert them into specific adjustment coefficients.

[0107] S5.4. Combine the real-time activity environment data with the influence laws of human heat consumption and human water needs, and calculate the adjustment coefficient that affects nutritional requirements according to the activity environment through regression analysis.

[0108] It should be noted that first, collect the user's real-time activity environment data (such as temperature, humidity, etc.) and the corresponding human heat consumption and water demand data; then, use the regression analysis method to explore the relationships between environmental factors and heat consumption and water demand; quantify the specific impacts on human nutritional requirements under different environmental conditions according to the results, for example, increasing the demand for water and electrolytes in a high-temperature environment; finally, convert these impacts into specific adjustment coefficients;

[0109] It should also be noted that the influence laws of human heat consumption are mainly reflected in the basal metabolic rate (BMR), daily activity level, and the thermic effect of food. The basal metabolic rate refers to the energy required by the body to maintain basic physiological functions at rest; the daily activity level includes physical activities and exercises, which significantly increase the total heat consumption; the thermic effect of food refers to the energy consumed during the processes of digesting, absorbing, and metabolizing food. These data are usually obtained through indirect calorimetry, direct measurement, or estimation using specific formulas;

[0110] The influencing rules of human body's water requirement are based on various factors, including environmental temperature, humidity, individual activity intensity, and individual health status, etc. In high-temperature and high-humidity environments, the human body will increase heat dissipation through sweating, resulting in increased water loss; high-intensity exercise will also accelerate the loss of body fluids. In addition, certain health conditions (such as fever, vomiting, or diarrhea) will also increase the water requirement. The data sources of these rules are extensive and can be obtained through research on water balance under laboratory conditions, epidemiological investigations, and individualized health monitoring devices.

[0111] S5.5. Set minimum and maximum constraint conditions according to the user's basic physiological needs and health status.

[0112] It should be noted that first, evaluate the user's basic information such as age, gender, weight, height, activity level, etc., and combine their health status (such as whether there are chronic diseases, allergy history, or special nutritional needs) to determine the basic physiological needs such as the recommended intakes of daily required calories, macronutrients, and micronutrients; then, based on medical guidelines and nutritional standards, set safe and effective minimum and maximum ranges for these nutritional indicators.

[0113] S5.6. The linear programming algorithm utilizes the adjustment coefficients of activity intensity affecting nutritional requirements, the adjustment coefficients of activity location affecting nutritional requirements, and the adjustment coefficients of activity environment affecting nutritional requirements, and adjusts the user's personalized nutritional requirements through the minimum and maximum constraint conditions.

[0114] It should be noted that first, determine each adjustment coefficient, which quantifies the influence of different factors (such as activity intensity, location, environment) on nutritional requirements; then, set the minimum and maximum constraint conditions for nutrient intake based on the user's physiological needs and health status; then construct a linear programming model, input the above adjustment coefficients as variables, and set the objective function to optimize nutritional distribution while ensuring that all set constraint conditions are met; by solving the linear programming model, calculate the most suitable nutritional intake under the given conditions, so as to make precise adjustments to the user's personalized nutritional requirements.

[0115] S6. Based on the adjusted user's personalized nutritional requirements and combined with the user's taste preferences, analyze the matching degree of each food ingredient combination, and conduct customized adjustments through the particle swarm optimization algorithm to generate the user's personalized recipe.

[0116] S6.1. Obtain the user's taste preferences through the user's diet history records, feedback scores, and taste preference questionnaires.

[0117] S6.2. Combine the adjusted user's personalized nutritional requirements with the user's taste preferences, and gradually search for food ingredients and dish combinations that meet the requirements through the particle swarm optimization algorithm to form a food ingredient meal plan.

[0118] It should be noted that first, the personalized nutritional needs and taste preferences of the user are clarified, such as the need for specific nutrients or the preference for certain foods; then a target function is defined, which comprehensively considers the nutritional matching degree and taste satisfaction, and the particle swarm optimization algorithm is used to search in the space of possible ingredient and dish combinations; each "particle" in the algorithm represents a potential meal plan, and they update their positions according to the individual optimal solution and the social optimal solution, continuously evaluating and adjusting each plan during the process to ensure that it meets both nutritional needs and taste preferences, and finally outputs the ingredient meal plan.

[0119] S6.3. Analyze the matching degree between the nutritional value of the ingredient combinations in each ingredient meal plan and the user's taste preferences through the Pearson correlation coefficient.

[0120] It should be noted that first, for each meal plan, determine all the ingredients included therein and their corresponding nutritional value indicators (such as protein content, vitamin level, etc.) and indicators reflecting the user's taste preferences (such as the degree of preference for a certain ingredient or the preference for cooking methods); then calculate the Pearson correlation coefficient between these nutritional value indicators and taste preference indicators to quantify the strength and direction of the linear relationship between the two; evaluate the matching degree of the ingredient combination in meeting the nutritional needs and the user's taste preferences according to the obtained Pearson correlation coefficient.

[0121] S6.4. Perform a weighted sum of the matching degrees of each ingredient combination to obtain the evaluation score of the ingredient combination, and screen out the ingredients that meet the user's personalized needs through the ascending sorting method to form the user's personalized recipe.

[0122] It should be noted that first, calculate the scores for each ingredient combination according to the matching degree between its nutritional value and the user's taste preferences, and assign weights to these scores according to the importance of the nutrients or the user's specific preferences; then perform a weighted sum of the scores of all ingredient combinations in each meal plan to obtain the overall evaluation score of the plan; then sort all meal plans in ascending order according to the evaluation scores from high to low; finally, select the top-ranked plans, that is, the ingredient combinations that meet both the user's nutritional needs and taste preferences, to form the final personalized recipe.

[0123] This embodiment also provides a personalized intelligent meal planning system based on the user's condition, including: a data collection module, a model construction module, a model prediction module, a nutritional requirement adjustment module, and a recipe formulation module;

[0124] The data collection module is used to collect the user's comprehensive health data and multi-dimensional real-time user context data, and preprocess the two types of data;

[0125] A model construction module constructs a user nutrition requirement prediction model by combining a long short-term memory network model with a multi-layer perceptron model;

[0126] A model prediction module is used to obtain the user's metabolic rate by using a healthy energy calculator in combination with augmented reality (AR) technology and applying the Harris-Benedict algorithm, and in combination with the preprocessed comprehensive user health data, obtain the user's nutrition requirement value through the user nutrition requirement prediction model, and generate the user's personalized nutrition requirement;

[0127] A nutrition requirement adjustment module is used to analyze the preprocessed multi-dimensional real-time user context data to obtain an adjustment coefficient that affects the nutrition requirement, and adjust the user's personalized nutrition requirement through a linear programming algorithm;

[0128] A recipe formulation module analyzes the matching degree of each ingredient combination based on the adjusted user's personalized nutrition requirement and in combination with the user's taste preference, and performs customized adjustment through a particle swarm optimization algorithm to generate the user's personalized recipe.

[0129] This embodiment also provides a computer device applicable to the case of a personalized intelligent meal distribution method based on the user's situation, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the personalized intelligent meal distribution method based on the user's situation proposed in the above embodiment.

[0130] This computer device can be a terminal. This computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the computer device housing, or an external keyboard, touchpad, or mouse, etc.

[0131] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the personalized intelligent meal distribution method based on user conditions proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0132] In summary, the present invention: accurately predicts the nutritional needs of users through the combination of the long short-term memory network (LSTM) model and the multi-layer perceptron model, and optimizes the food ingredients combination through the particle swarm optimization algorithm, successfully realizing personalized intelligent meal distribution. The beneficial effects are that it can dynamically adjust the nutritional needs based on the user's health status and real-time context data, and optimize the food ingredients combination in combination with the user's taste preferences, so as to generate a recipe that not only meets the nutritional needs but also satisfies the personalized taste, significantly improving the scientificity, accuracy and user satisfaction of personalized meal distribution.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A personalized intelligent meal distribution method based on user conditions, characterized in that: including, collecting comprehensive user health data and multi-dimensional real-time user context data, and preprocessing the two types of data; constructing a user nutritional requirement prediction model through a long short-term memory network model combined with a multi-layer perceptron model; using a healthy energy calculator combined with enhanced AR technology, and applying the Harris-Benedict algorithm to obtain the user's metabolic rate. Combining with the preprocessed comprehensive user health data, through the user nutritional requirement prediction model, obtaining the user's nutritional requirement value, and generating the user's personalized nutritional requirement. The specific steps are as follows. Using a healthy energy calculator and combined with enhanced AR technology, simulating the user to perform lightweight activities in a virtual environment, capturing individual differences by observing the user's physical reactions in different situations, and creating a unique health profile for each user; Analyzing the user's activity level collected by a sports bracelet through the activity log provided by the user to obtain the user's average activity intensity; Based on the user's average activity intensity, combined with the user's health profile, adjusting according to the proportion of time occupied by the user in different activity types and the corresponding proportion of metabolic equivalent values to obtain a personalized activity factor; Based on the user's health profile, using a healthy energy calculator and applying the Harris-Benedict algorithm to obtain the basal metabolic rate, and obtaining the user's metabolic rate by combining the personalized activity factor; Based on the user's metabolic rate and the preprocessed comprehensive user health data, obtaining the user's nutritional requirement value through the user nutritional requirement prediction model; The specific steps for generating the user's personalized nutritional requirement are as follows. Based on historical user health data, analyze through the clustering analysis algorithm and set a health threshold ; Compare the user's nutritional requirement value with the health threshold to evaluate the user's health status; When it indicates that the user's health condition is poor and multiple nutrients need to be ingested; When it indicates that the user's health condition is good and there is no need to intake nutrients; Based on the user's health status and the nutrients required by the user, generating the user's personalized nutritional requirement through a genetic algorithm combined with nutritional theory; Analyzing the preprocessed multi-dimensional real-time user context data to obtain an adjustment coefficient affecting nutritional requirements, and adjusting the user's personalized nutritional requirement through a linear programming algorithm. The specific steps are as follows. Analyzing the real-time activity time data through a time series analysis method, and combining with kinematic principles to determine the activity intensity of each time period; Using a weight allocation method to convert different activity intensities into adjustment coefficients for the impact of activity intensity on nutritional requirements; Analyzing the real-time activity location data and the user's physiological health data through a data mining method, and mapping them into adjustment coefficients for the impact of activity location on nutritional requirements; Combining the real-time activity environment data with the influence laws of human heat consumption and human water requirements, and calculating the adjustment coefficient for the impact of activity environment on nutritional requirements through a regression analysis method; Setting minimum and maximum constraint conditions according to the user's basic physiological needs and health status; The linear programming algorithm uses the adjustment coefficient for the impact of activity intensity on nutritional requirements, the adjustment coefficient for the impact of activity location on nutritional requirements, and the adjustment coefficient for the impact of activity environment on nutritional requirements, and adjusts the user's personalized nutritional requirement through the minimum and maximum constraint conditions; Based on the adjusted user's personalized nutritional requirement and combined with the user's taste preferences, analyzing the matching degree of each ingredient combination, and performing customized adjustment through a particle swarm optimization algorithm to generate the user's personalized recipe.

2. The personalized intelligent meal - matching method based on user conditions according to claim 1, characterized in that: The comprehensive user health data includes the user's basic information, the user's physiological health data, and the user's activity level; The multi-dimensional real-time user context data includes real-time activity time data, real-time activity location data, and real-time activity environment data; The preprocessing includes data cleaning and normalization processing.

3. The personalized intelligent meal distribution method based on user conditions according to claim 1, wherein: To construct the user nutritional requirement prediction model, the specific steps are as follows. Convert the preprocessed historical comprehensive user health data into a time series format through the sliding window method; Use the long short-term memory network model as the basic model; The input layer receives the preprocessed historical comprehensive user health data converted into a time series format; The LSTM layer, through memory units and gate mechanisms, the LSTM model selectively remembers and forgets information to capture long-term dependencies in the time series; The fully connected layer extracts features and reduces dimensions based on the output of the LSTM layer, and outputs preliminary features; In combination with the multi-layer perceptron model, the input layer receives the preliminary features output by the LSTM model; The concatenation layer concatenates the preliminary features with other relevant factors; The MLP layer inputs the concatenated vector into a single-layer MLP for further feature combination and transformation; The output layer outputs the final result; Finally, the user nutritional requirement prediction model is obtained.

4. The personalized intelligent meal distribution method based on user conditions according to claim 1, characterized in that: Based on the adjusted user personalized nutritional requirements and combined with the user's taste preferences, analyze the matching degree of each food ingredient combination, and perform customized adjustment through the particle swarm optimization algorithm to generate the user's personalized recipe. The specific steps are as follows. Obtain the user's taste preferences through the user's diet history records, feedback scores, and taste preference questionnaires; Combine the adjusted user personalized nutritional requirements with the user's taste preferences, and gradually search for food ingredients and dish combinations that meet the requirements through the particle swarm optimization algorithm to form a food ingredient meal plan; Analyze the matching degree between the nutritional value of the food ingredient combinations in each food ingredient meal plan and the user's taste preferences through the Pearson correlation coefficient; Perform weighted summation on the matching degrees of each food ingredient combination to obtain the food ingredient combination evaluation score, and screen out the food ingredients that meet the user's personalized needs through the positive sorting method to form the user's personalized recipe.

5. A personalized intelligent meal distribution system based on user conditions, based on the personalized intelligent meal distribution method based on user conditions according to any one of claims 1 to 4, characterized in that: It includes a data collection module, a model construction module, a model prediction module, a nutritional requirement adjustment module, and a recipe formulation module; The data collection module is used to collect the comprehensive user health data and multi-dimensional real-time user context data, and preprocess the two types of data; The model construction module constructs the user nutritional requirement prediction model through the long short-term memory network model combined with the multi-layer perceptron model; The model prediction module is used to utilize the healthy energy calculator combined with the augmented reality technology, and apply the Harris-Benedict algorithm to obtain the user's metabolic rate. Combined with the preprocessed comprehensive user health data, through the user nutritional requirement prediction model, obtain the user's nutritional requirement value, and generate the user's personalized nutritional requirements; The nutritional requirement adjustment module is used to analyze the preprocessed multi-dimensional real-time user context data to obtain the adjustment coefficient affecting the nutritional requirements, and adjust the user's personalized nutritional requirements through the linear programming algorithm; The recipe formulation module, based on the adjusted user personalized nutritional requirements and combined with the user's taste preferences, analyzes the matching degree of each food ingredient combination, and performs customized adjustment through the particle swarm optimization algorithm to generate the user's personalized recipe.

6. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the personalized intelligent meal allocation method based on user conditions according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the personalized intelligent meal allocation method based on user conditions according to any one of claims 1 to 4.

Images