Nutrition matching recommendation method and system based on nutrition recommendation data

Abstract

This invention discloses a method and system for recommending nutritional combinations based on nutritional recommendation data, belonging to the field of intelligent recommendation technology. The invention collects user dietary structure and behavior data over a period of time based on user dietary habits, and summarizes these data. Simultaneously, it fuses the collected data using a fusion processing method. After fusion processing, it analyzes the processed user dietary behavior data using nutritional analysis. Following the analysis, it collects the user's daily calorie consumption in real time and constructs a nutritional meal planning model based on the principle of food and medicine sharing the same origin. Multiple nutritional meal planning schemes are obtained through this model. Finally, based on the processed user dietary behavior data and the multiple nutritional meal planning schemes, a corresponding nutritional meal planning scheme is recommended to the user through a fusion recommendation method, with real-time modifications and adjustments, thus improving the rationality of the meal planning recommendations.

Description

Technical Field

[0001] This invention relates to the field of intelligent recommendation technology, specifically to a method and system for recommending nutritional combinations based on nutritional recommendation data. Background Technology

[0002] China boasts a highly developed catering industry, with a rich variety of cuisines and a vast array of dishes. As living standards improve, people are not only pursuing nutritional value but also becoming more discerning in their tastes. Faced with a wealth of ingredients, consumers often struggle to choose, as nutrition and flavor seem mutually exclusive.

[0003] Existing technologies, such as the invention patent application with publication number CN112820378A, disclose a method and system for recommending nutritional meal plans based on dietary behavior. The method includes: collecting and storing multi-dimensional data on various ingredients / products and daily standard intake data for each nutrient element to establish a dietary knowledge base; recording user personal information and dietary records, and calculating the nutrient element intake in the user's dietary records based on the dietary knowledge base; performing a benchmark calculation on the nutrient element intake in the user's dietary records according to the recommended daily intake of various foods per person in the food pyramid; and providing nutritional meal plans to the user based on the benchmark calculation results.

[0004] As can be seen from the above solutions, most current nutritional meal recommendation methods fail to fully consider individual user behavior, often focusing only on nutrition or taste. The meal recommendation results often cannot fully take into account factors such as user nutrition, taste, meal frequency, and duration, and thus have certain limitations. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for recommending nutritional combinations based on nutritional recommendation data, which solves the problems existing in the background art.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a method for recommending nutritional combinations based on nutritional recommendation data, specifically including the following steps: S1. Set the user's dietary habits collection period, collect the user's dietary structure and user behavior over a period of time based on the user's dietary habits collection period, and summarize the user's dietary structure data and user behavior data. S2. The collected user dietary structure data and user behavior data are fused through a fusion processing method to obtain fused user dietary behavior data. At the same time, the fused user dietary behavior data is processed through a data processing method to obtain processed user dietary behavior data. S3. Analyze the processed user dietary behavior data using nutritional analysis methods to obtain the analyzed user nutritional evaluation indicators. S4. Real-time collection of users’ daily calorie consumption, and construction of a nutritional meal planning model based on the principle of food and medicine homology. The analyzed user nutritional evaluation indicators are then input into the nutritional meal planning model to obtain multiple sets of nutritional meal planning schemes. S5. Based on the processed user dietary behavior data and multiple sets of nutritional meal plans, the system recommends corresponding nutritional meal plans to users through a fusion recommendation method, and collects user feedback for real-time intelligent adjustment.

[0007] Preferably, the step of setting a user dietary habit collection period, collecting the user's dietary structure and user behavior over a period of time based on the user dietary habit collection period, and summarizing the user dietary structure data and user behavior data includes the following steps: S11. Set the dietary structure and user behavior standards for collection; The dietary structure standards include: grains, vegetables and fruits, protein, dairy products, and oils; The user behavior criteria include: dietary preferences, timing of food intake, quantity of food intake, and frequency of food intake; S12. Based on the collected dietary structure and user behavior standards, define the dietary structure and user behavior of the collected users over a period of time, and summarize the user dietary structure data and user behavior data. The user's dietary structure data set includes: The user behavior data set is defined as including: ; in, This represents dietary preference data within a user behavior dataset. This represents the timing data of food intake within the user behavior dataset. This represents dietary intake data within a user behavior dataset. This represents the frequency of food intake data within the user behavior dataset. These represent grains, vegetables and fruits, protein, dairy products, and oils in the user's dietary structure data set, respectively. Define dietary preferences in user behavior data as a subset of user dietary structure data.

[0008] Preferably, the step of fusing the collected user dietary structure data and user behavior data through a fusion processing method to obtain fused user dietary behavior data, and then processing the fused user dietary behavior data through a data processing method to obtain processed user dietary behavior data, includes the following steps: S21. Process the collected user dietary structure data and user behavior data through data fusion to obtain fused user dietary behavior data; Remove the dimensions from user behavior data through data transformation methods; Convert the timing, quantity, and frequency of food intake in user behavior data into numerical data. A set of standard user behavior data must include: dietary preferences, timing of food intake, amount of food intake, and frequency of food intake; The collected user dietary structure data is quantified based on standard user behavior data to obtain fused user dietary behavior data. By collecting dietary preference data from user behavior data, we can determine the foods that users like and classify them based on user dietary structure data. After classification, the user's preferred diet is quantified by the amount and frequency of food intake in the user behavior data. After quantification, the user's dietary behavior data is obtained by data fusion. The integrated user dietary behavior data includes: dietary intake type, corresponding dietary intake amount, dietary intake time, dietary intake frequency, and dietary preferences; S22. The fused user dietary behavior data is processed by data fitting to obtain processed user dietary behavior data.

[0009] Preferably, the process of processing the fused user dietary behavior data through data fitting to obtain processed user dietary behavior data includes the following steps: The timing of food intake was summarized, and the distribution of food intake time was determined through data fitting. The data fitting formula is shown below: ; in, This indicates the time of food intake in the merged user dietary behavior data. The probability density function of the distribution. For scale parameters, The shape parameter representing the time distribution of dietary intake after data fitting; The frequency of various types of food intake of users is summarized and merged from the user's dietary behavior data, and the probability distribution of the frequency of various types of food intake is calculated by normal fitting. The normal fit formula is shown below: ; in, This indicates the first point regarding the integrated user dietary behavior data. The normal fit function of the frequency of various food intakes of users within a time period. It is the natural logarithm. Represents variance. This represents the mean of a normal distribution; The processed user dietary behavior data is obtained by summarizing dietary intake types, corresponding dietary intake amounts, probability distributions of frequency of various dietary intakes, dietary intake time distributions, and dietary preferences.

[0010] Preferably, the step of analyzing the processed user dietary behavior data through nutritional analysis to obtain the analyzed user nutritional evaluation indicators includes the following steps: S31. Analyze the dietary preferences in the processed user dietary behavior data using the enumeration method to determine the user dietary preference index; The dietary preferences in the processed user dietary behavior data are listed using the enumeration method to determine each user's dietary preference; After determining each user's dietary preference, we collected evaluation indicators from multiple groups of nutrition experts on each user's dietary preference. By summarizing the evaluation indicators of multiple groups of nutrition experts on each user's dietary preference, a user dietary preference index is obtained. S32. Based on the determined user dietary preference indicators, determine the weights of the user dietary preference indicators using the entropy weight method. S33. Determine user nutrition evaluation indicators based on the weights of user dietary preference indicators and the probability distribution of dietary intake types, corresponding dietary intake amounts, frequency of various types of user dietary intake, and dietary intake time distribution in the processed user dietary behavior data. Set the user nutrition evaluation index as the weight of the user's dietary preference index. Dietary intake type Corresponding type of dietary intake Probability distribution of the frequency of various dietary intakes of users Distribution of food intake over time.

[0011] Preferably, determining the weights of user dietary preference indicators using the entropy weight method based on the determined user dietary preference indicators includes the following steps: Calculate the information entropy of the j-th item in the evaluation index based on the determined user dietary preference index; ; in, , Indicates the number of evaluation indicators. This represents the weight value of the evaluation index for the j-th dietary preference in the i-th group of processed user dietary behavior data. This represents the information entropy of the j-th item in the evaluation index; The weight of the user's dietary preference index is calculated based on the information entropy of the j-th item in the evaluation index. ; in, This represents the weight of the user's dietary preference index regarding the j-th type of dietary preference.

[0012] Preferably, the real-time collection of users' daily calorie consumption, the construction of a nutritional meal planning model based on the principle of food and medicine sharing the same origin, and the input of the analyzed user nutritional evaluation indicators into the nutritional meal planning model to obtain multiple sets of nutritional meal plans include the following steps: S41. Based on real-time collection of users' daily calorie consumption, determine the user's daily calorie requirement; S42. Based on the user's daily calorie needs, construct a nutritional meal planning model according to the principle of food and medicine sharing the same origin; The user's daily calorie requirements include: 10%-20% protein intake, 20%-30% fat intake, and 50%-60% carbohydrate intake. Grains, vegetables, and fruits are carbohydrates; some dairy products are proteins, and some dairy products and oils are also carbohydrates. The nutritional meal plan model is set as follows: Protein intake percentage + Fat intake percentage + Carbohydrate intake percentage = 100%; S43. Input the analyzed user nutrition evaluation indicators into the nutrition meal planning model to obtain multiple sets of nutrition meal planning schemes.

[0013] Preferably, determining the user's daily calorie needs based on real-time collection of the user's daily calorie consumption includes the following steps: Calculate the user's BMI index; BMI = Actual weight (kg) - [Height (m)] 2; Calculate the minimum required calories based on the user's BMI index; Minimum calorie requirement = ideal body weight * * ; in, * It is a unit representing the standard daily calorie requirement per kilogram of body weight; The user's daily calorie requirement is determined based on the user's daily calorie consumption and minimum calorie requirement; The user's daily calorie requirement = minimum calorie requirement + calorie deficit from previous periods; The previous calorie deficit = previous user calorie consumption - previous user dietary calorie intake; Past user calorie intake = Past user dietary behavior data Heat conversion coefficient.

[0014] Preferably, the step of recommending a corresponding nutritional meal plan to the user based on processed user dietary behavior data and multiple sets of nutritional meal plans through a fusion recommendation method, and collecting user feedback for real-time intelligent adjustment, includes the following steps: S51. Initialize the processed user dietary behavior data and multiple sets of nutritional meal plans. Based on the hybrid genetic algorithm, encode the multiple sets of nutritional meal plans into chromosomes and construct a population set. At the same time, set the population size and the number of iterations. Each chromosome group is assigned a code to represent a set of nutritional meal plans; S52. Randomly select S groups of nutritional meal plans from multiple groups to construct the initial population. ; S53. Calculate the initial population. The similarity between the nutritional meal plan and the dietary preference data in the processed user dietary behavior data is calculated, and the calculated similarity is used as the fitness function of the genetic algorithm to calculate the fitness of each chromosome encoding in the population. The similarity calculation formula is as follows: Set up a set of nutritional meal plans and dietary preference data set ; in, The first of multiple nutritional meal plans Group nutritional meal plan, To process the user's dietary behavior data, the first Group dietary preference data; Calculate the mean and standard deviation of the two sets of data respectively, and determine the initial population based on the calculated mean and standard deviation using the Pearson correlation coefficient. The similarity between the nutritional meal plan in the data and the dietary preference data in the processed user dietary behavior data; The formulas for calculating the fitness of each chromosome are as follows: ; in, Indicates the initial population The fitness encoded by chromosome group 1 in the middle Indicates the initial population The similarity between the nutritional meal plan corresponding to the first group of chromosomes and the dietary preference data in the processed user dietary behavior data; S54. Selecting superior chromosome codes from chromosome codes based on roulette wheel selection; S55. Selected superior chromosome codes are crossovered using a sequential crossover method, resulting in a new population. ; S56. Randomly select a chromosome code in the population and mutate it with a set probability to produce a mutated population. ; S57. Compare the fitness difference between the initial population and the mutated population. ; when <0 indicates that the fitness of the mutated population is higher than that of the original population. Accepting the mutated population means... ≥0 indicates that the fitness of the mutated population is lower than that of the original population, and the mutated population is rejected. S58. Determine whether the maximum number of iterations has been reached based on the number of iterations of the algorithm. If the maximum number of iterations has been reached, output the optimal solution. If the maximum number of iterations has not been reached, continue to execute steps S52-S57. The optimal output solution is set as the recommended nutritional meal plan; S59. Collect user feedback on the recommended nutritional meal plan in real time and make real-time intelligent adjustments. When the user feedback is unsatisfactory, remove the currently recommended nutritional meal plan and recalculate the recommended nutritional meal plan through steps S52-S58.

[0015] The present invention also provides a nutritional combination recommendation system based on nutritional recommendation data, which is used to implement a nutritional combination recommendation method based on nutritional recommendation data. The system includes: a data acquisition module, a data processing module, a data analysis module, a nutritional meal planning module, and a meal planning intelligent recommendation module. The data acquisition module is used to collect the user's dietary structure and user behavior over a period of time; The data processing module is used to fuse the collected dietary structure and user behavior to obtain processed user dietary behavior data. The data analysis module is used to analyze the processed user dietary behavior data to obtain the analyzed user nutrition evaluation index. The nutritional meal planning module is used to construct a nutritional meal planning model based on the principle of food and medicine sharing the same origin, and to generate multiple sets of nutritional meal planning schemes based on the analyzed user nutritional evaluation indicators. The intelligent meal recommendation module is used to recommend corresponding nutritional meal plans to users based on processed user dietary behavior data and multiple sets of nutritional meal plans through a fusion recommendation method, and to make intelligent adjustments in real time.

[0016] The beneficial effects of this invention are as follows: (1) This invention collects the dietary structure and user behavior of users over a period of time based on the user's dietary habits collection cycle, and summarizes the user's dietary structure data and user behavior data. At the same time, it performs fusion processing on the collected user dietary structure data and user behavior data. After the fusion processing is completed, it analyzes the processed user dietary behavior data through nutritional analysis. After the analysis is completed, it collects the user's daily calorie consumption in real time, and constructs a nutritional meal planning model based on the principle of food and medicine homology. It obtains multiple sets of nutritional meal planning schemes through the nutritional meal planning model. Finally, based on the processed user dietary behavior data and multiple sets of nutritional meal planning schemes, it recommends the corresponding nutritional meal planning scheme to the user through fusion recommendation and modifies and adjusts it in real time, thereby improving the rationality of the meal planning recommendation.

[0017] (2) This invention processes the collected user dietary structure data and user behavior data through data fusion, and at the same time, it fits the fused user dietary behavior data through data fitting to determine the distribution of dietary intake time and the probability distribution of the frequency of various types of user dietary intake, thereby improving the accuracy of subsequent nutritional meal recommendations.

[0018] (3) This invention analyzes the dietary preferences in the processed user dietary behavior data by listing, determines the weight of the user dietary preference index by entropy weight method, and finally determines the user nutrition evaluation index based on user preferences and nutrition analysis method, thereby improving the accuracy of subsequent nutritional meal recommendations.

[0019] (4) This invention collects the user’s daily calorie consumption in real time, constructs a nutritional meal planning model based on the principle of food and medicine being of the same origin, and inputs the analyzed user nutritional evaluation index into the nutritional meal planning model to obtain multiple nutritional meal planning schemes. Then, the genetic algorithm and similarity matching algorithm are used to combine and recommend multiple nutritional meal planning schemes, thereby improving the real-time performance of nutritional meal planning recommendations. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the nutritional recommendation method for nutritional recommendation data of the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In a specific embodiment of the present invention, Reference Figure 1 As shown, the present invention provides a method for recommending nutritional combinations based on nutritional recommendation data, comprising the following steps: S1. Set the user's dietary habits collection period, collect the user's dietary structure and user behavior over a period of time based on the user's dietary habits collection period, and summarize the user's dietary structure data and user behavior data. S2. The collected user dietary structure data and user behavior data are fused through a fusion processing method to obtain fused user dietary behavior data. At the same time, the fused user dietary behavior data is processed through a data processing method to obtain processed user dietary behavior data. S3. Analyze the processed user dietary behavior data using nutritional analysis methods to obtain the analyzed user nutritional evaluation indicators. S4. Real-time collection of users’ daily calorie consumption, and construction of a nutritional meal planning model based on the principle of food and medicine homology. The analyzed user nutritional evaluation indicators are then input into the nutritional meal planning model to obtain multiple sets of nutritional meal planning schemes. S5. Based on the processed user dietary behavior data and multiple sets of nutritional meal plans, recommend corresponding nutritional meal plans to users through a fusion recommendation method, and collect user feedback for real-time intelligent adjustment; Furthermore, referring to Figure 1 As shown, the process involves setting a user diet habit collection period, collecting user diet structure and behavior data over a specific period, and summarizing the user diet structure data and user behavior data. This includes the following steps: S11. Set the dietary structure and user behavior standards for collection; The dietary structure standards include: grains, vegetables and fruits, protein, dairy products, and oils; The user behavior criteria include: dietary preferences, timing of food intake, quantity of food intake, and frequency of food intake; S12. Based on the collected dietary structure and user behavior standards, define the dietary structure and user behavior of the collected users over a period of time, and summarize the user dietary structure data and user behavior data. The user's dietary structure data set includes: The user behavior data set is defined as including: ; in, This represents dietary preference data within a user behavior dataset. This represents the timing data of food intake within the user behavior dataset. This represents dietary intake data within a user behavior dataset. This represents the frequency of food intake data within the user behavior dataset. These represent grains, vegetables and fruits, protein, dairy products, and oils in the user's dietary structure data set, respectively. Furthermore, dietary preferences in user behavior data are defined as a subset of user dietary structure data; Furthermore, referring to Figure 1 As shown, the collected user dietary structure data and user behavior data are fused using a fusion processing method to obtain fused user dietary behavior data. Simultaneously, the fused user dietary behavior data is processed using data processing methods to obtain processed user dietary behavior data, including the following steps: S21. Process the collected user dietary structure data and user behavior data through data fusion to obtain fused user dietary behavior data; Remove the dimensions from user behavior data through data transformation methods; Convert the timing, quantity, and frequency of food intake in user behavior data into numerical data. Furthermore, a set of standard user behavior data must include: dietary preferences, timing of food intake, quantity of food intake, and frequency of food intake; The collected user dietary structure data is quantified based on standard user behavior data to obtain fused user dietary behavior data. By collecting dietary preference data from user behavior data, we can determine the foods that users like and classify them based on user dietary structure data. After classification, the user's preferred diet is quantified by the amount and frequency of food intake in the user behavior data. After quantification, the user's dietary behavior data is obtained by data fusion. The integrated user dietary behavior data includes: dietary intake type, corresponding dietary intake amount, dietary intake time, dietary intake frequency, and dietary preferences; S22. The fused user dietary behavior data is processed by data fitting to obtain processed user dietary behavior data. The timing of food intake was summarized, and the distribution of food intake time was determined through data fitting. The data fitting formula is shown below: ; in, This indicates the time of food intake in the merged user dietary behavior data. The probability density function of the distribution. For scale parameters, The shape parameter representing the time distribution of dietary intake after data fitting; Furthermore, the frequency of various types of food intake of users in the merged user dietary behavior data is summarized, and the probability distribution of the frequency of various types of food intake of users is calculated by normal fitting method; The normal fit formula is shown below: ; in, This indicates the first point regarding the integrated user dietary behavior data. The normal fit function of the frequency of various food intakes of users within a time period. It is the natural logarithm. Represents variance. This represents the mean of a normal distribution; Furthermore, by summarizing the types of food intake, the amount of each type of food intake, the probability distribution of the frequency of each type of food intake, the time distribution of food intake, and dietary preferences, we obtain the processed user dietary behavior data. Furthermore, referring to Figure 1 As shown, the process of analyzing processed user dietary behavior data using nutritional analysis to obtain user nutritional evaluation indicators includes the following steps: S31. Analyze the dietary preferences in the processed user dietary behavior data using the enumeration method to determine the user dietary preference index; The dietary preferences in the processed user dietary behavior data are listed using the enumeration method to determine each user's dietary preference; Furthermore, after determining each user's dietary preference, we collected evaluation indicators from multiple groups of nutrition experts on each user's dietary preference. Furthermore, by summarizing the evaluation indicators of multiple groups of nutrition experts on each of the user's dietary preferences, a user dietary preference index was obtained. S32. Based on the determined user dietary preference indicators, determine the weights of the user dietary preference indicators using the entropy weight method. Calculate the information entropy of the j-th item in the evaluation index based on the determined user dietary preference index; ; in, , Indicates the number of evaluation indicators. This represents the weight value of the evaluation index for the j-th dietary preference in the i-th group of processed user dietary behavior data. This represents the information entropy of the j-th item in the evaluation index; Furthermore, the weight of the user's dietary preference index is calculated based on the information entropy of the j-th item in the evaluation index; ; in, This represents the weight of the user's dietary preference index regarding the j-th dietary preference; S33. Determine user nutrition evaluation indicators based on the weights of user dietary preference indicators and the probability distribution of dietary intake types, corresponding dietary intake amounts, frequency of various types of user dietary intake, and dietary intake time distribution in the processed user dietary behavior data. Set the user nutrition evaluation index as the weight of the user's dietary preference index. Dietary intake type Corresponding type of dietary intake Probability distribution of the frequency of various dietary intakes of users Distribution of food intake over time; Furthermore, referring to Figure 1 As shown, the system collects users' daily calorie consumption data in real time. Simultaneously, a nutritional meal planning model is constructed based on the principle of food and medicine sharing the same origin. The analyzed user nutritional evaluation indicators are then input into the nutritional meal planning model to obtain multiple sets of nutritional meal plans, including the following steps: S41. Based on real-time collection of users' daily calorie consumption, determine the user's daily calorie requirement; Calculate the user's BMI index; BMI = Actual weight (kg) - [Height (m)] 2; Calculate the minimum required calories based on the user's BMI index; Minimum calorie requirement = ideal body weight * * ; in, * It is a unit representing the standard daily calorie requirement per kilogram of body weight; The user's daily calorie requirement is determined based on the user's daily calorie consumption and minimum calorie requirement; The user's daily calorie requirement = minimum calorie requirement + calorie deficit from previous periods; The previous calorie deficit = previous user calorie consumption - previous user dietary calorie intake; Past user calorie intake = Past user dietary behavior data Heat conversion coefficient; S42. Based on the user's daily calorie needs, construct a nutritional meal planning model according to the principle of food and medicine sharing the same origin; The user's daily calorie requirements include: 10%-20% protein intake, 20%-30% fat intake, and 50%-60% carbohydrate intake. Grains, vegetables, and fruits are carbohydrates; some dairy products are proteins, and some dairy products and oils are also carbohydrates. The nutritional meal plan model is set as follows: Protein intake percentage + Fat intake percentage + Carbohydrate intake percentage = 100%; S43. Input the analyzed user nutrition evaluation indicators into the nutrition meal planning model to obtain multiple sets of nutrition meal planning schemes; The analyzed user nutrition evaluation indicators and the user's daily calorie requirement are input into the nutrition meal planning model, and multiple nutrition meal planning schemes are generated through random permutation and enumeration. For example, the generated multiple sets of nutritional meal plans include: Option 1: Grains + High-Quality Protein + Fresh Vegetables; Option 2: Mixed grain porridge + bean products and cold side dishes; Option 3: Grains + Eggs + Nuts + Fruit; Option 4: Oatmeal + Milk + Fruit + Nuts; Option 5: Noodles / rice noodles + lean meat + vegetables + mushrooms; Example combination: 100g whole wheat noodles + 50g lean beef slices + 50g broccoli + 2 shiitake mushrooms + 1 egg; Furthermore, referring to Figure 1 As shown, based on processed user dietary behavior data and multiple sets of nutritional meal plans, a corresponding nutritional meal plan is recommended to the user through a fusion recommendation method, and user feedback is collected for real-time intelligent adjustment, including the following steps: S51. Initialize the processed user dietary behavior data and multiple sets of nutritional meal plans. Based on the hybrid genetic algorithm, encode the multiple sets of nutritional meal plans into chromosomes and construct a population set. At the same time, set the population size and the number of iterations. Each chromosome group is assigned a code to represent a set of nutritional meal plans; S52. Randomly select S groups of nutritional meal plans from multiple groups to construct the initial population. ; S53. Calculate the initial population. The similarity between the nutritional meal plan and the dietary preference data in the processed user dietary behavior data is calculated, and the calculated similarity is used as the fitness function of the genetic algorithm to calculate the fitness of each chromosome encoding in the population. The similarity calculation formula is as follows: Set up a set of nutritional meal plans and dietary preference data set ; in, The first of multiple nutritional meal plans Group nutritional meal plan, To process the user's dietary behavior data, the first Group dietary preference data; Furthermore, the mean and standard deviation of the two sets of data were calculated separately, and the initial population was determined based on the calculated mean and standard deviation using the Pearson correlation coefficient. The similarity between the nutritional meal plan in the data and the dietary preference data in the processed user dietary behavior data; Pearson correlation coefficient: ; in, For data The Pearson correlation coefficient, A collection of nutritional meal plans standard deviation For dietary preference data set Standard deviation; Furthermore, the Pearson correlation coefficient was set as the initial population value. The similarity between the nutritional meal plan in the data and the dietary preference data in the processed user dietary behavior data; The formulas for calculating the fitness of each chromosome are as follows: ; in, Indicates the initial population The fitness encoded by chromosome group 1 in the middle Indicates the initial population The similarity between the nutritional meal plan corresponding to the first group of chromosomes and the dietary preference data in the processed user dietary behavior data; S54. Selecting superior chromosome codes from chromosome codes based on roulette wheel selection; S55. Selected superior chromosome codes are crossovered using a sequential crossover method, resulting in a new population. ; S56. Randomly select a chromosome code in the population and mutate it with a set probability to produce a mutated population. ; S57. Compare the fitness difference between the initial population and the mutated population. ; when <0 indicates that the fitness of the mutated population is higher than that of the original population. Accepting the mutated population means... ≥0 indicates that the fitness of the mutated population is lower than that of the original population, and the mutated population is rejected. S58. Determine whether the maximum number of iterations has been reached based on the number of iterations of the algorithm. If the maximum number of iterations has been reached, output the optimal solution. If the maximum number of iterations has not been reached, continue to execute steps S52-S57. The optimal output solution is set as the recommended nutritional meal plan; S59. Collect user feedback on the recommended nutritional meal plan in real time and make real-time intelligent adjustments. When the user feedback is unsatisfactory, remove the currently recommended nutritional meal plan and recalculate the recommended nutritional meal plan through steps S52-S58. In one specific embodiment, the nutritional combination recommendation system based on nutritional recommendation data is used to implement a nutritional combination recommendation method based on nutritional recommendation data. The system includes: a data acquisition module, a data processing module, a data analysis module, a nutritional meal planning module, and a meal planning intelligent recommendation module. The data acquisition module is used to collect the user's dietary structure and user behavior over a period of time; The data processing module is used to fuse the collected dietary structure and user behavior to obtain processed user dietary behavior data. The data analysis module is used to analyze the processed user dietary behavior data to obtain the analyzed user nutrition evaluation index. The nutritional meal planning module is used to construct a nutritional meal planning model based on the principle of food and medicine sharing the same origin, and to generate multiple sets of nutritional meal planning schemes based on the analyzed user nutritional evaluation indicators. The intelligent meal recommendation module is used to recommend corresponding nutritional meal plans to users based on processed user dietary behavior data and multiple sets of nutritional meal plans through a fusion recommendation method, and to make intelligent adjustments in real time.

[0024] It should be noted that, The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A method for recommending nutritional combinations based on nutritional recommendation data, characterized in that, Includes the following steps: S1. Set the user's dietary habits collection period, collect the user's dietary structure and user behavior over a period of time based on the user's dietary habits collection period, and summarize the user's dietary structure data and user behavior data. S2. The collected user dietary structure data and user behavior data are fused through a fusion processing method to obtain fused user dietary behavior data. At the same time, the fused user dietary behavior data is processed through a data processing method to obtain processed user dietary behavior data. S3. Analyze the processed user dietary behavior data using nutritional analysis methods to obtain the analyzed user nutritional evaluation indicators. S4. Real-time collection of users’ daily calorie consumption, and construction of a nutritional meal planning model based on the principle of food and medicine homology. The analyzed user nutritional evaluation indicators are then input into the nutritional meal planning model to obtain multiple sets of nutritional meal planning schemes. S5. Based on the processed user dietary behavior data and multiple sets of nutritional meal plans, the system recommends corresponding nutritional meal plans to users through a fusion recommendation method, and collects user feedback for real-time intelligent adjustment.

2. The nutritional combination recommendation method based on nutritional recommendation data according to claim 1, characterized in that, The process of setting a user dietary habit collection period, collecting user dietary structure and user behavior data over a period of time based on the user dietary habit collection period, and summarizing the user dietary structure data and user behavior data includes the following steps: S11. Set the dietary structure and user behavior standards for collection; The dietary structure standards include: grains, vegetables and fruits, protein, dairy products, and oils; The user behavior criteria include: dietary preferences, timing of food intake, quantity of food intake, and frequency of food intake; S12. Based on the collected dietary structure and user behavior standards, define the dietary structure and user behavior of the collected users over a period of time, and summarize the user dietary structure data and user behavior data. The user's dietary structure data set includes: The user behavior data set is defined as including: ; in, This represents dietary preference data within a user behavior dataset. This represents the timing data of food intake within the user behavior dataset. This represents dietary intake data within a user behavior dataset. This represents the frequency of food intake data within the user behavior dataset. These represent grains, vegetables and fruits, protein, dairy products, and oils in the user's dietary structure data set, respectively. Define dietary preferences in user behavior data as a subset of user dietary structure data.

3. The nutritional combination recommendation method based on nutritional recommendation data according to claim 1, characterized in that, The process of fusing the collected user dietary structure data and user behavior data through a fusion processing method to obtain fused user dietary behavior data, and then processing the fused user dietary behavior data to obtain processed user dietary behavior data, includes the following steps: S21. Process the collected user dietary structure data and user behavior data through data fusion to obtain fused user dietary behavior data; Remove the dimensions from user behavior data through data transformation methods; Convert the timing, quantity, and frequency of food intake in user behavior data into numerical data. A set of standard user behavior data must include: dietary preferences, timing of food intake, amount of food intake, and frequency of food intake; The collected user dietary structure data is quantified based on standard user behavior data to obtain fused user dietary behavior data. By collecting dietary preference data from user behavior data, we can determine the foods that users like and classify them based on user dietary structure data. After classification, the user's preferred diet is quantified by the amount and frequency of food intake in the user behavior data. After quantification, the user's dietary behavior data is obtained by data fusion. The integrated user dietary behavior data includes: dietary intake type, corresponding dietary intake amount, dietary intake time, dietary intake frequency, and dietary preferences; S22. The fused user dietary behavior data is processed by data fitting to obtain processed user dietary behavior data.

4. The nutritional combination recommendation method based on nutritional recommendation data according to claim 3, characterized in that, The process of processing the fused user dietary behavior data through data fitting to obtain processed user dietary behavior data includes the following steps: The timing of food intake was summarized, and the distribution of food intake time was determined through data fitting. The data fitting formula is shown below: ； in, This indicates the time of food intake in the merged user dietary behavior data. The probability density function of the distribution. For scale parameters, The shape parameter representing the time distribution of dietary intake after data fitting; The frequency of various types of food intake of users is summarized and merged from the user's dietary behavior data, and the probability distribution of the frequency of various types of food intake is calculated by normal fitting. The normal fit formula is shown below: ； in, This indicates the first point regarding the integrated user dietary behavior data. The normal fit function of the frequency of various food intakes of users within a time period. It is the natural logarithm. Represents variance. This represents the mean of a normal distribution; The processed user dietary behavior data is obtained by summarizing dietary intake types, corresponding dietary intake amounts, probability distributions of frequency of various dietary intakes, dietary intake time distributions, and dietary preferences.

5. The nutritional combination recommendation method based on nutritional recommendation data according to claim 1, characterized in that, The process of analyzing processed user dietary behavior data using nutritional analysis to obtain analyzed user nutritional evaluation indicators includes the following steps: S31. Analyze the dietary preferences in the processed user dietary behavior data using the enumeration method to determine the user dietary preference index; The dietary preferences in the processed user dietary behavior data are listed using the enumeration method to determine each user's dietary preference; After determining each user's dietary preference, we collected evaluation indicators from multiple groups of nutrition experts on each user's dietary preference. By summarizing the evaluation indicators of multiple groups of nutrition experts on each user's dietary preference, a user dietary preference index is obtained. S32. Based on the determined user dietary preference indicators, determine the weights of the user dietary preference indicators using the entropy weight method. S33. Determine user nutrition evaluation indicators based on the weights of user dietary preference indicators and the probability distribution of dietary intake types, corresponding dietary intake amounts, frequency of various types of user dietary intake, and dietary intake time distribution in the processed user dietary behavior data. Set the user nutrition evaluation index as the weight of the user's dietary preference index. Dietary intake type Corresponding type of dietary intake Probability distribution of the frequency of various dietary intakes of users Distribution of food intake over time.

6. The nutritional combination recommendation method based on nutritional recommendation data according to claim 5, characterized in that, The process of determining the weights of user dietary preference indicators using the entropy weight method based on established user dietary preference indicators includes the following steps: Calculate the information entropy of the j-th item in the evaluation index based on the determined user dietary preference index; ； in, , Indicates the number of evaluation indicators. This represents the weight value of the evaluation index for the j-th dietary preference in the i-th group of processed user dietary behavior data. This represents the information entropy of the j-th item in the evaluation index; The weight of the user's dietary preference index is calculated based on the information entropy of the j-th item in the evaluation index. ； in, This represents the weight of the user's dietary preference index regarding the j-th type of dietary preference.

7. The nutritional combination recommendation method based on nutritional recommendation data according to claim 1, characterized in that, The process of collecting users' daily calorie consumption in real time, constructing a nutritional meal planning model based on the principle of food and medicine sharing the same origin, and inputting the analyzed user nutritional evaluation indicators into the nutritional meal planning model to obtain multiple sets of nutritional meal plans includes the following steps: S41. Based on real-time collection of users' daily calorie consumption, determine the user's daily calorie requirement; S42. Based on the user's daily calorie needs, construct a nutritional meal planning model according to the principle of food and medicine sharing the same origin; The user's daily calorie requirements include: 10%-20% protein intake, 20%-30% fat intake, and 50%-60% carbohydrate intake. Grains, vegetables, and fruits are carbohydrates; some dairy products are proteins, and some dairy products and oils are also carbohydrates. The nutritional meal plan model is set as follows: Protein intake percentage + Fat intake percentage + Carbohydrate intake percentage = 100%; S43. Input the analyzed user nutrition evaluation indicators into the nutrition meal planning model to obtain multiple sets of nutrition meal planning schemes.

8. The nutritional combination recommendation method based on nutritional recommendation data according to claim 7, characterized in that, The process of determining a user's daily calorie needs based on real-time data collection includes the following steps: Calculate the user's BMI index; BMI = Actual weight (kg) - [Height (m)] 2; Calculate the minimum required calories based on the user's BMI index; Minimum calorie requirement = ideal body weight * * ; in, * It is a unit representing the standard daily calorie requirement per kilogram of body weight; The user's daily calorie requirement is determined based on the user's daily calorie consumption and minimum calorie requirement; The user's daily calorie requirement = minimum calorie requirement + calorie deficit from previous periods; The previous calorie deficit = previous user calorie consumption - previous user dietary calorie intake; Past user calorie intake = Past user dietary behavior data Heat conversion coefficient.

9. A method for recommending nutritional combinations based on nutritional recommendation data according to claim 1, characterized in that, The process of recommending corresponding nutritional meal plans to users based on processed user dietary behavior data and multiple sets of nutritional meal plans, and collecting user feedback for real-time intelligent adjustments, includes the following steps: S51. Initialize the processed user dietary behavior data and multiple sets of nutritional meal plans. Based on the hybrid genetic algorithm, encode the multiple sets of nutritional meal plans into chromosomes and construct a population set. At the same time, set the population size and the number of iterations. Each chromosome group is assigned a code to represent a set of nutritional meal plans; S52. Randomly select S groups of nutritional meal plans from multiple groups to construct the initial population. ; S53. Calculate the initial population. The similarity between the nutritional meal plan and the dietary preference data in the processed user dietary behavior data is calculated, and the calculated similarity is used as the fitness function of the genetic algorithm to calculate the fitness of each chromosome encoding in the population. The similarity calculation formula is as follows: Set up a set of nutritional meal plans and dietary preference data set ; in, The first of multiple nutritional meal plans Group nutritional meal plan, To process the user's dietary behavior data, the first Group dietary preference data; Calculate the mean and standard deviation of the two sets of data respectively, and determine the initial population based on the calculated mean and standard deviation using the Pearson correlation coefficient. The similarity between the nutritional meal plan in the data and the dietary preference data in the processed user dietary behavior data; The formulas for calculating the fitness of each chromosome are as follows: ； in, Indicates the initial population The fitness encoded by chromosome group 1 in the middle Indicates the initial population The similarity between the nutritional meal plan corresponding to the first group of chromosomes and the dietary preference data in the processed user dietary behavior data; S54. Selecting superior chromosome codes from chromosome codes based on roulette wheel selection; S55. Selected superior chromosome codes are crossovered using a sequential crossover method, resulting in a new population. ; S56. Randomly select a chromosome code in the population and mutate it with a set probability to produce a mutated population. ; S57. Compare the fitness difference between the initial population and the mutated population. ; when <0 indicates that the fitness of the mutated population is higher than that of the original population. Accepting the mutated population means... ≥0 indicates that the fitness of the mutated population is lower than that of the original population, and the mutated population is rejected. S58. Determine whether the maximum number of iterations has been reached based on the number of iterations of the algorithm. If the maximum number of iterations has been reached, output the optimal solution. If the maximum number of iterations has not been reached, continue to execute steps S52-S57. The optimal output solution is set as the recommended nutritional meal plan; S59. Collect user feedback on the recommended nutritional meal plan in real time and make real-time intelligent adjustments. When the user feedback is unsatisfactory, remove the currently recommended nutritional meal plan and recalculate the recommended nutritional meal plan through steps S52-S58.

10. A system for implementing the nutritional combination recommendation method based on nutritional recommendation data as described in claim 1, characterized in that, include: The system includes a data acquisition module, a data processing module, a data analysis module, a nutrition and meal planning module, and an intelligent meal recommendation module. The data acquisition module is used to collect the user's dietary structure and user behavior over a period of time; The data processing module is used to fuse the collected dietary structure and user behavior to obtain processed user dietary behavior data. The data analysis module is used to analyze the processed user dietary behavior data to obtain the analyzed user nutrition evaluation index. The nutritional meal planning module is used to construct a nutritional meal planning model based on the principle of food and medicine sharing the same origin, and to generate multiple sets of nutritional meal planning schemes based on the analyzed user nutritional evaluation indicators. The intelligent meal recommendation module is used to recommend corresponding nutritional meal plans to users based on processed user dietary behavior data and multiple sets of nutritional meal plans through a fusion recommendation method, and to make intelligent adjustments in real time.

Images