Child nutrition recommendation method based on big data

Abstract

The invention discloses a child nutrition recommendation method based on big data, and relates to the technical field of intelligent recommendation, and the method comprises the steps: collecting child basic information and vitamin D level detection data, and carrying out the preprocessing; using correlation analysis and principal component analysis to identify main features affecting the vitamin D demand, and analyzing the current vitamin D nutrition status of the child through a deep learning model; a deep neural network is utilized, historical child basic information and vitamin D level detection data are combined, a vitamin D nutritional demand prediction model of children is constructed, and the future nutritional demand of children for vitamin D is predicted; personalized vitamin D nutrition suggestions are provided for children; through correlation analysis, principal component analysis and construction of the vitamin D demand prediction model, dynamic adjustment can be performed according to specific conditions of different children, scientificity and practicability of suggestions are ensured, the vitamin D nutritional status of the children can be improved, and the health level of the children is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent recommendation, and in particular to a method for recommending children's nutrition based on big data. Background Art

[0002] Big data, or massive data, refers to the amount of data involved that is so large that it cannot be captured, managed, processed, and organized into information that helps companies make more proactive business decisions within a reasonable time through mainstream software tools.

[0003] Children are in a critical period of growth and development, and their bodies have a greater demand for various nutrients. If certain nutrients are lacking, it will affect their physical health and intellectual development. For example, lack of protein will affect the development of muscles and bones, and lack of iron will affect blood synthesis and immune function. Through reasonable nutritional supplements, the needs of children's physical development and health can be met. According to children's growth and development and physical needs, a reasonable diet plan and nutrient intake can be formulated to promote the healthy development of children's body.

[0004] In the field of child nutrition management, by collecting children's basic information and vitamin D level test data, the nutritional status of children can be effectively assessed and personalized nutrition recommendations can be made. However, most of the existing technical means rely on traditional statistical methods, lack effective feature selection and dimensionality reduction techniques, and it is difficult to accurately identify the key features that affect vitamin D demand, resulting in low accuracy of nutritional recommendations. In addition, the existing vitamin D demand prediction models are mostly based on static data and fail to fully utilize the dynamic change trends in historical data, which limits the model's prediction ability and personalized service level. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a child nutrition recommendation method based on big data to solve the problems of inaccurate vitamin D demand prediction and insufficient personalized nutrition advice in the prior art.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a method for recommending nutrition for children based on big data, which includes collecting basic information of children and vitamin D level detection data, and performing preprocessing;

[0009] Based on the pre-processed basic information of children and vitamin D level test data, correlation analysis and principal component analysis were used to identify the main features that affect vitamin D demand, and the current vitamin D nutritional status of children was analyzed through a deep learning model.

[0010] According to the current vitamin D nutritional status of children, a deep neural network is used to combine historical basic information of children and vitamin D level test data to build a children's vitamin D nutritional demand prediction model to predict the future nutritional demand of children for vitamin D;

[0011] Provide personalized vitamin D nutrition recommendations for children based on the predicted future nutritional requirements of vitamin D for children.

[0012] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, wherein:

[0013] The basic information of the child includes age, gender, height and weight;

[0014] The preprocessing includes data cleaning and missing value processing.

[0015] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, the method includes the following steps: finding out the characteristics closely related to the vitamin D requirement through correlation analysis,

[0016] For each feature, its mean was calculated, and the Pearson correlation coefficient was used to calculate the correlation between each feature and vitamin D level;

[0017] All features are organized into a matrix, where each row represents a sample and each column represents a feature. The matrix is ​​standardized to construct a correlation matrix.

[0018] The correlation coefficient with vitamin D level was extracted from the correlation matrix. For each feature, its correlation coefficient with vitamin D level was calculated. All correlation coefficients were sorted according to their absolute values, and the feature with the largest absolute value of correlation coefficient was selected as the feature most correlated with vitamin D level.

[0019] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, principal component analysis is used to reduce the data dimension and retain important features, including the following steps:

[0020] First, the original feature matrix is ​​standardized and the covariance matrix of the standardized feature matrix is ​​calculated;

[0021] Perform eigendecomposition on the covariance matrix and solve the eigenvalues ​​and corresponding eigenvectors;

[0022] Sort the obtained eigenvalues ​​in descending order and arrange the corresponding eigenvectors accordingly;

[0023] Select the eigenvectors corresponding to the first k largest eigenvalues ​​and construct a new feature matrix;

[0024] The standardized feature matrix is ​​transformed using the selected eigenvector matrix to obtain the reduced-dimensional feature matrix.

[0025] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, the method includes: analyzing the current vitamin D nutritional status of children by a deep learning model, including the following steps:

[0026] Choose a fully connected neural network as the base model, set the size of the input layer to match the dimension of the feature vector after PCA dimensionality reduction, build a network architecture with multiple hidden layers, the output layer corresponds to the number of categories in the classification task, and use the Softmax activation function to output the probability of each category;

[0027] Based on historical children's basic information and vitamin D level test data, the model is trained to output the probability of each category. The argmax function is used to obtain the index of the category with the highest probability. The index is mapped back to the actual category name, and the children's current vitamin D nutritional status is divided into good and poor.

[0028] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, a model for predicting children's vitamin D nutritional needs is constructed based on the current nutritional status of children using a deep neural network, combined with historical basic information of children and vitamin D level detection data, to predict the nutritional needs of children for vitamin D in the future, including the following steps:

[0029] When children's current vitamin D nutritional status is good, maintain the original nutritional package;

[0030] When children's current vitamin D nutritional status is poor, a deep neural network is used to combine historical children's basic information and vitamin D level test results to build a child's vitamin D nutritional demand prediction model to predict the child's future nutritional demand for vitamin D, expressed as,

[0031]

[0032] in, is the predicted nutritional requirement of vitamin D for children in the future, α is the age correlation coefficient, t is the age of the child, β is the age standardization factor, γ is the exponential factor of age, δ is the weight-to-height ratio correlation coefficient, w is the child's weight, h is the child's height, ζ is the vitamin D concentration correlation coefficient, D is the child's current vitamin D concentration, η is the vitamin D concentration standardization factor, θ is the exponential factor of vitamin D concentration, φ is the gender correlation coefficient, g is the child's gender, ξ is the gender-centered constant, and σ is the standard deviation of gender differences.

[0033] As a preferred embodiment of the method for recommending children's nutrition based on big data of the present invention, providing personalized children's vitamin D nutrition recommendations based on the predicted future nutritional requirements of children for vitamin D includes the following steps:

[0034] According to medical guidelines and nutritional research, the range of vitamin D demand is defined as [0,,A1] for low demand, (A1,,A2] for medium demand, and (A2,+∞) for high demand.

[0035] When the predicted vitamin D requirement is in the low range, maintain current dietary habits and monitor vitamin D levels regularly to ensure that vitamin D levels are within the healthy range;

[0036] When the predicted vitamin D requirement is in the medium range, it is recommended to increase the intake of foods rich in vitamin D. In addition, increase the time spent outdoors to promote the synthesis of vitamin D in the skin.

[0037] When the predicted vitamin D requirement is in the high demand range, in addition to increasing food sources, supplement intake under the guidance of a doctor is also required. It is recommended to consult a professional doctor to develop a personalized vitamin D supplement plan;

[0038] Dynamically adjust personalized vitamin D nutrition recommendations for children based on the specific situation of the child.

[0039] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the child nutrition recommendation method based on big data as described in the first aspect of the present invention is implemented.

[0040] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the child nutrition recommendation method based on big data as described in the first aspect of the present invention is implemented.

[0041] The beneficial effects of the present invention are as follows: through correlation analysis and principal component analysis, features closely related to vitamin D demand are effectively screened out, the interference of redundant information on the prediction results is eliminated, and the generalization ability of the model is improved. The deep learning model is used to classify the current vitamin D nutritional status of children. Compared with traditional methods, it can better capture nonlinear relationships and enhance the prediction ability of the model. The vitamin D demand prediction model proposed in the present invention not only takes into account the basic information of children and vitamin D level test data, but also introduces factors such as age, gender, and weight-to-height ratio, which improves the accuracy and reliability of the prediction results. The personalized nutritional recommendations provided based on the prediction results can be dynamically adjusted according to the specific circumstances of different children, ensuring the scientificity and practicality of the recommendations, and helping to improve children's vitamin D nutritional status and improve their health level. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0043] Figure 1 This is a flow chart of the child nutrition recommendation method based on big data in Example 1. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0045] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0046] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0047] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and provides a method for recommending children's nutrition based on big data, comprising the following steps:

[0048] S1. Collect children's basic information and vitamin D level test data, and perform preprocessing.

[0049] Children’s basic information includes age, gender, height and weight;

[0050] Preprocessing includes data cleaning and missing value processing.

[0051] It should be noted that through the preprocessing steps, data standardization and cleaning were achieved to ensure the reliability and accuracy of subsequent analysis. First, the basic information of children (including age, gender, height and weight) and vitamin D level test data were collected, and then the data were cleaned and missing values ​​were processed. This step effectively eliminated invalid or erroneous data and improved data quality. Through these high-quality data, the vitamin D nutritional status of children can be more accurately assessed, providing a solid foundation for subsequent feature selection and model building. Finally, this step ensures the integrity and consistency of the data and provides reliable data support for subsequent steps.

[0052] S2. Based on the pre-processed basic information of children and vitamin D level test data, correlation analysis and principal component analysis were used to identify the main features affecting vitamin D demand, and the current vitamin D nutritional status of children was analyzed through a deep learning model.

[0053] Correlation analysis was used to identify features that were closely related to vitamin D requirements, including the following steps:

[0054] For each feature, its mean value was calculated, and the Pearson correlation coefficient was used to calculate the correlation between each feature and vitamin D level, expressed as,

[0055]

[0056] Among them, r xy is the Pearson correlation coefficient of features x and y, x i and i are the two eigenvalues ​​of the i-th sample, and is the average value of features x and y, and n is the number of samples;

[0057] All features are organized into a matrix, where each row represents a sample and each column represents a feature. The matrix is ​​standardized so that the mean of each feature is 0 and the standard deviation is 1. A correlation matrix is ​​constructed to show the correlation between the features.

[0058] Extract the correlation coefficient with vitamin D level (assuming it is feature y) from the correlation matrix. For each feature, calculate its correlation coefficient with vitamin D level. Sort all correlation coefficients by absolute value, and select the feature with the largest absolute value of correlation coefficient as the feature most correlated with vitamin D level.

[0059] Using principal component analysis to reduce data dimensions and retain important features includes the following steps:

[0060] First, the original feature matrix is ​​standardized so that the mean of each feature is 0 and the standard deviation is 1. The covariance matrix of the standardized feature matrix is ​​calculated and expressed as:

[0061]

[0062] Among them, C is the covariance matrix, Z is the standardized feature matrix. Through standardization, the mean of each feature is 0 and the standard deviation is 1, thereby eliminating the dimension effect and ensuring the comparability between different features. n is the number of samples. When calculating the covariance matrix, n-1 is used as the denominator for unbiased estimation. The calculated covariance matrix is ​​closer to the true covariance matrix of the population, Z T is the transpose of the matrix Z. After the transposition, matrix multiplication can be performed to obtain the covariance matrix;

[0063] Perform eigendecomposition on the covariance matrix and solve the eigenvalues ​​and corresponding eigenvectors, expressed as,

[0064] Cv=λv;

[0065] Among them, C is the covariance matrix, which reflects the linear relationship between each feature. The covariance matrix is ​​a symmetric matrix. Its diagonal elements represent the variance of each feature itself, and the non-diagonal elements represent the covariance between the features. λ is the eigenvalue, which represents the degree of data variation along the direction of the eigenvector. The larger the eigenvalue, the greater the degree of data variation along the direction of the eigenvector, that is, the higher the importance in this direction. v is the eigenvector, which represents the projection of the data in the direction of the principal component. The direction of the eigenvector indicates the direction of the maximum variation of the data, and the size of the eigenvalue represents the degree of variation along this direction.

[0066] Sort the obtained eigenvalues ​​in descending order and arrange the corresponding eigenvectors accordingly;

[0067] Select the eigenvectors corresponding to the first k largest eigenvalues ​​and construct a new feature matrix, expressed as,

[0068] Y=ZV k ;

[0069] Among them, Y is the new feature matrix, which is obtained by combining the standardized feature matrix Z with the k most important eigenvector matrices V k The new matrix obtained by matrix multiplication, this new matrix Y contains the data after dimensionality reduction, retaining only the most important feature information, V k is the matrix composed of the eigenvectors corresponding to the first k largest eigenvalues, V k Contains the k most important eigenvectors, which correspond to the largest eigenvalues ​​and represent the most important directions in the data. By selecting these eigenvectors, the most important information in the data can be retained and dimensionality reduction can be performed;

[0070] The standardized feature matrix is ​​transformed using the selected eigenvector matrix to obtain the reduced-dimensional feature matrix.

[0071] The deep learning model is used to analyze the current vitamin D nutritional status of children, including the following steps:

[0072] Select a fully connected neural network as the base model, set the size of the input layer to match the dimension of the feature vector after PCA dimensionality reduction, build a network architecture with multiple hidden layers for feature extraction and classification, the output layer corresponds to the number of categories in the classification task, and use the Softmax activation function to output the probability of each category;

[0073] Based on historical children's basic information and vitamin D level test data, the model is trained to output the probability of each category. The argmax function is used to obtain the index of the category with the highest probability. The index is mapped back to the actual category name, and the current vitamin D nutritional status of children is divided into good and poor.

[0074] When the index is 0, it means that the child’s current vitamin D nutritional status is good;

[0075] When the index is 1, it means that the child’s current vitamin D nutritional status is poor.

[0076] It should be noted that by applying the steps of correlation analysis and principal component analysis, the identification of the main features that affect vitamin D demand was achieved, and the current vitamin D nutritional status of children was analyzed through a deep learning model. First, by calculating the Pearson correlation coefficient, the features closely related to vitamin D demand were found. This step helped us identify which features had the most significant impact on vitamin D demand. Then, principal component analysis (PCA) was used to reduce the data dimension and retain important features, thereby reducing the time cost of model training and avoiding the overfitting problem. Finally, a deep learning model was constructed to analyze the current vitamin D nutritional status of children, and a fully connected neural network was used for classification to divide the current vitamin D nutritional status of children into good and poor. This step not only improved the accuracy of the prediction, but also provided valuable feature input for the subsequent vitamin D demand prediction model. Finally, this step achieved the effective identification of factors affecting vitamin D demand and improved the accuracy of vitamin D nutritional status analysis through a deep learning model.

[0077] S3. Based on the children’s current vitamin D nutritional status, a deep neural network is used to combine historical children’s basic information and vitamin D level test data to build a children’s vitamin D nutritional demand prediction model to predict children’s future nutritional demand for vitamin D.

[0078] When children's current vitamin D nutritional status is good, maintain the original nutritional package;

[0079] When children's current vitamin D nutritional status is poor, a deep neural network is used to combine historical children's basic information and vitamin D level test results to build a child's vitamin D nutritional demand prediction model to predict the child's future nutritional demand for vitamin D, expressed as,

[0080]

[0081] in, is the predicted nutritional requirement of vitamin D for children in the future, indicating the demand of children for vitamin D in the future predicted based on the current physiological state of children and other relevant information. α is the age correlation coefficient, which controls the influence of age t on the demand for vitamin D. Through regression analysis or other statistical methods, the value of α is fitted based on a large amount of historical data on children's age and vitamin D demand. For example, a multivariate linear regression model can be used, with age as one of the independent variables and vitamin D demand as the dependent variable, to obtain an estimated value of ɑ. t is the age of the child, and β is the age standardization factor, which is used to standardize age so that the contribution of children of different age groups in the model is more balanced. According to the distribution of children's ages, a suitable value is selected as the standardization benchmark. The mean or median of children's age is usually selected as the value of β. γ is the exponential factor of age, which determines the nonlinear degree of the effect of age on vitamin D demand. By analyzing the nonlinear relationship between children's age and vitamin D demand, nonlinear regression or other curve fitting methods can be used to determine the value of γ. δ is the correlation coefficient of weight-to-height ratio, which controls the influence of the ratio of weight w to height h on vitamin D demand. By analyzing the correlation between weight and height on vitamin D demand, the value of δ is determined using statistical methods (such as correlation coefficient analysis). w is the child's weight, h is the child's height, ζ is the vitamin D concentration correlation coefficient, which controls the influence of current vitamin D concentration D on future demand. By analyzing the relationship between vitamin D concentration and demand, the value of ζ is determined using statistical methods such as regression analysis. D is the child's current vitamin D concentration. η is the vitamin D concentration normalization factor, which is used to standardize vitamin D concentration so that children with different concentration levels contribute more to the model. Add balance, according to the distribution of children's vitamin D concentration, select a suitable value as the standardization benchmark, usually choose the average concentration or median concentration as the value of η, θ is the exponential factor of vitamin D concentration, which determines the nonlinear degree of the effect of vitamin D concentration on demand. By analyzing the relationship between vitamin D concentration and demand, the nonlinear regression method is used to determine the optimal index value, φ is the gender correlation coefficient, which controls the degree of influence of gender g on vitamin D demand. By analyzing the correlation between gender and vitamin D demand, the value of φ is determined using statistical methods (such as analysis of variance), g is the child's gender, ξ is the gender centering constant, which is used to standardize gender so that gender differences are more obvious in the model. The value of males is usually 0 and the value of females is 1, that is, ξ=0.5, σ is the standard deviation of gender differences, which describes the degree of fluctuation of gender differences. By analyzing the data distribution of gender differences, the standard deviation is calculated. Usually, the standard deviation of gender differences in historical data can be used as the value of σ.

[0082] It should be noted that the formula for predicting children's vitamin D nutritional needs is designed based on a comprehensive consideration of children's age, weight-to-height ratio, current vitamin D concentration, and gender characteristics. Future vitamin D nutritional needs are quantitatively predicted through nonlinear combinations to reflect the complex impact of these factors on demand. Taking into account the effect of age on vitamin D requirements, the requirement for vitamin D may increase with age. The ratio of weight to height was taken into account, and the nonlinear relationship between weight and height was emphasized through a logarithmic function to reflect the impact of children's body shape on their vitamin D needs. The effect of current vitamin D concentrations on demand was considered; lower vitamin D concentrations may mean higher demand,

[0083] It should be noted that by constructing a vitamin D nutritional demand prediction model, an accurate prediction of children's future vitamin D nutritional demand is achieved. When a child's current vitamin D nutritional status is poor, a prediction model is constructed using a deep neural network combined with historical basic information of children and vitamin D level test results. The model comprehensively considers the child's age, weight-to-height ratio, current vitamin D concentration and gender characteristics, and quantitatively predicts future vitamin D nutritional requirements through nonlinear combinations. This step quantifies the impact of various factors on vitamin D requirements through complex mathematical formulas, thereby improving the accuracy of the prediction. Ultimately, this step achieves an accurate prediction of vitamin D requirements, providing a scientific basis for subsequent personalized nutritional recommendations.

[0084] S4. Provide personalized vitamin D nutrition recommendations for children based on their predicted future nutritional requirements for vitamin D.

[0085] According to medical guidelines and nutritional research, the range of vitamin D demand is defined as [0, A1] for low demand, (A1, A2] for medium demand, and (A2, +∞) for high demand.

[0086] Specifically, first of all, a large amount of data on children's vitamin D requirements needs to be collected and analyzed. These data can be obtained from clinical trials, epidemiological surveys and long-term follow-up studies. The purpose of the analysis is to understand the distribution of children's vitamin D requirements in different age groups, different genders and different physiological states; refer to authoritative medical guidelines at home and abroad, such as the vitamin D nutrition guidelines issued by the World Health Organization (WHO), to understand the recommended vitamin D intake standards. These guidelines usually give the recommended intake range for people of different ages, combined with the latest nutritional research results, especially the research on children's vitamin D requirements, to understand the latest scientific discoveries. These studies usually explore the vitamin D requirements under different conditions, such as different seasons, geographical regions and individual differences. Based on the above data and guidelines, the approximate range of vitamin D requirements can be preliminarily divided. Generally, the range of vitamin D requirements can be divided into three main ranges according to the level of demand, including low demand range, medium demand range and high demand range.

[0087] When the predicted vitamin D requirement is in the low range, maintain current dietary habits and monitor vitamin D levels regularly to ensure that vitamin D levels are within the healthy range;

[0088] When the predicted vitamin D requirement is in the medium range, it is recommended to increase the intake of foods rich in vitamin D, such as cod liver oil, fatty fish (such as salmon, mackerel), egg yolks, etc. In addition, appropriately increase the time of outdoor activities to promote the synthesis of vitamin D in the skin;

[0089] When the predicted vitamin D requirement is in the high demand range, in addition to increasing food sources, supplement intake under the guidance of a doctor may be required. It is recommended to consult a professional doctor to develop a personalized vitamin D supplement plan;

[0090] Dynamically adjust personalized vitamin D nutrition recommendations for children based on the specific situation of the child.

[0091] It should be noted that by providing personalized vitamin D nutrition advice for children, personalized nutrition advice is provided to children based on the predicted results. According to medical guidelines and nutrition research, different intervals of vitamin D requirements are defined, and targeted advice is provided according to different intervals. When the predicted vitamin D requirements are in different intervals, corresponding diet and lifestyle advice is provided, such as increasing the intake of foods rich in vitamin D, increasing outdoor activity time, etc. This step not only provides personalized nutrition advice for children, but also encourages a healthy lifestyle, which helps to improve children's vitamin D nutritional status. Ultimately, this step achieves personalized nutrition advice based on predicted results, which helps to improve children's overall health.

[0092] This embodiment also provides a computer device, which is suitable for the case of a child nutrition recommendation method based on big data, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the child nutrition recommendation method based on big data proposed in the above embodiment.

[0093] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the 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 the 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, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0094] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing a child nutrition recommendation method based on big data as proposed in the above embodiment is implemented; 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, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0095] In summary, the present invention effectively screens out features closely related to vitamin D demand through correlation analysis and principal component analysis, eliminates the interference of redundant information on the prediction results, and improves the generalization ability of the model. The deep learning model is used to classify the current vitamin D nutritional status of children. Compared with traditional methods, it can better capture nonlinear relationships and enhance the prediction ability of the model. The vitamin D demand prediction model proposed in the present invention not only takes into account the basic information of children and vitamin D level test data, but also introduces factors such as age, gender, and weight-to-height ratio, which improves the accuracy and reliability of the prediction results. The personalized nutritional recommendations provided based on the prediction results can be dynamically adjusted according to the specific circumstances of different children, ensuring the scientificity and practicality of the recommendations, and helping to improve children's vitamin D nutritional status and improve their health level.

Claims

1. A method for recommending children's nutrition based on big data, characterized by: include, Collect children's basic information and vitamin D level test data, and perform preprocessing; Based on the pre-processed basic information of children and vitamin D level test data, correlation analysis and principal component analysis were used to identify the main features that affect vitamin D demand, and the current vitamin D nutritional status of children was analyzed through a deep learning model. According to the current vitamin D nutritional status of children, a deep neural network is used to combine historical basic information of children and vitamin D level test data to build a children's vitamin D nutritional demand prediction model to predict the future nutritional demand of children for vitamin D; Provide personalized vitamin D nutrition recommendations for children based on the predicted future nutritional requirements of vitamin D for children.

2. The method for recommending children's nutrition based on big data according to claim 1, characterized in that: The basic information of the child includes age, gender, height and weight; The preprocessing includes data cleaning and missing value processing.

3. The method for recommending children's nutrition based on big data according to claim 1, characterized in that: Correlation analysis was used to identify features that were closely related to vitamin D requirements, including the following steps: For each feature, its mean was calculated, and the Pearson correlation coefficient was used to calculate the correlation between each feature and vitamin D level; All features are organized into a matrix, where each row represents a sample and each column represents a feature. The matrix is ​​standardized to construct a correlation matrix. The correlation coefficient with vitamin D level was extracted from the correlation matrix. For each feature, its correlation coefficient with vitamin D level was calculated. All correlation coefficients were sorted according to their absolute values, and the feature with the largest absolute value of correlation coefficient was selected as the feature most correlated with vitamin D level.

4. The method for recommending children's nutrition based on big data according to claim 3, characterized in that: Using principal component analysis to reduce data dimensions and retain important features includes the following steps: First, the original feature matrix is ​​standardized and the covariance matrix of the standardized feature matrix is ​​calculated; Perform eigendecomposition on the covariance matrix and solve the eigenvalues ​​and corresponding eigenvectors; Sort the obtained eigenvalues ​​in descending order and arrange the corresponding eigenvectors accordingly; Select the eigenvectors corresponding to the first k largest eigenvalues ​​and construct a new feature matrix; The standardized feature matrix is ​​transformed using the selected eigenvector matrix to obtain the reduced-dimensional feature matrix.

5. The method for recommending children's nutrition based on big data according to claim 4, characterized in that: The deep learning model is used to analyze the current vitamin D nutritional status of children, including the following steps: Choose a fully connected neural network as the base model, set the size of the input layer to match the dimension of the feature vector after PCA dimensionality reduction, build a network architecture with multiple hidden layers, the output layer corresponds to the number of categories in the classification task, and use the Softmax activation function to output the probability of each category; Based on historical children's basic information and vitamin D level test data, the model is trained to output the probability of each category. The argmax function is used to obtain the index of the category with the highest probability. The index is mapped back to the actual category name, and the children's current vitamin D nutritional status is divided into good and poor.

6. The method for recommending children's nutrition based on big data according to claim 5, characterized in that: According to the current vitamin D nutritional status of children, a deep neural network is used to combine historical basic information of children and vitamin D level test data to build a children's vitamin D nutritional demand prediction model to predict the future nutritional demand of children for vitamin D, including the following steps: When children's current vitamin D nutritional status is good, maintain the original nutritional package; When children's current vitamin D nutritional status is poor, a deep neural network is used to combine historical basic information of children and vitamin D level test results to build a child's vitamin D nutritional demand prediction model to predict the child's future nutritional demand for vitamin D, expressed as, in, is the predicted nutritional requirement of vitamin D for children in the future, α is the age correlation coefficient, t is the child’s age, β is the age standardization factor, γ is the age exponential factor, δ is the weight-to-height ratio correlation coefficient, w is the child’s weight, h is the child’s height, ζ is the vitamin D concentration correlation coefficient, D is the child’s current vitamin D concentration, η is the vitamin D concentration standardization factor, θ is the vitamin D concentration exponential factor, φ is the gender correlation coefficient, g is the child’s gender, ξ is the gender-centered constant, and σ is the standard deviation of gender differences.

7. The method for recommending children's nutrition based on big data according to claim 6, characterized in that: Based on the predicted future nutritional requirements of children for vitamin D, provide personalized vitamin D nutritional recommendations for children, including the following steps: According to medical guidelines and nutritional research, the range of vitamin D demand is defined as [0, A1] for low demand, (A1, A2] for medium demand, and (A2, +∞) for high demand. When the predicted vitamin D requirement is in the low range, maintain current dietary habits and monitor vitamin D levels regularly to ensure that vitamin D levels are within the healthy range; When the predicted vitamin D requirement is in the medium range, it is recommended to increase the intake of foods rich in vitamin D. In addition, increase the time spent outdoors to promote the synthesis of vitamin D in the skin. When the predicted vitamin D requirement is in the high demand range, in addition to increasing food sources, supplement intake under the guidance of a doctor is also required. It is recommended to consult a professional doctor to develop a personalized vitamin D supplement plan; Dynamically adjust personalized vitamin D nutrition recommendations for children based on the specific situation of the child.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the child nutrition recommendation method based on big data described in any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the child nutrition recommendation method based on big data described in any one of claims 1 to 7 are implemented.

Images