A white duck soup health formula recommendation system based on nutrient component analysis

Abstract

This application belongs to the field of big data processing technology. It discloses a white duck soup healthy recipe recommendation system based on nutritional component analysis. The system includes: collecting basic user information and white duck soup ingredient attribute data, and cleaning the data to obtain a white duck soup recommendation dataset; performing scenario profile recognition and constructing scenario-based user sub-profiles, outputting a scenario-based profile parameter set; performing nutritional recipe constraint mapping, outputting an individualized nutritional constraint set; performing nutrient balance analysis, outputting a dynamic nutritional budget parameter set; constructing a white duck soup candidate recipe set and performing recipe role adjustment, then performing nutritional component matching analysis, outputting a candidate recipe scoring table; and outputting the target white duck soup healthy recipe recommendation result. This improves the scenario adaptability and nutritional accuracy of the white duck soup healthy recipe recommendation.

Description

Technical Field

[0001] This application relates to the field of big data processing technology, and more specifically, to a white duck soup health recipe recommendation system based on nutritional component analysis. Background Technology

[0002] With the increasing demand for modern health management, recommending healthy recipes for white duck soup based on nutritional analysis has become a necessary requirement in the catering and health fields. However, traditional recipe recommendation systems still face many technical bottlenecks.

[0003] In reality, most traditional recipe recommendation systems remain at the level of general recipe matching or static nutritional label screening. They lack a processing mechanism to build scenario-based user sub-profiles for different types of white duck soup consumption scenarios. They cannot differentiate the recommendation needs of the same user in different scenarios based on conditions such as dining time, ambient temperature, seasonal information, and user personal information. This leads to the recommendation results easily using the same logic and failing to reflect the specificity of the scenario. In addition, traditional recipe recommendation systems usually only perform coarse-grained control of white duck soup candidate recipes based on preset group nutritional standards. They fail to map individual nutritional goals to standardized catering constraints in institutional or scenario templates, and also lack consideration of historical nutritional intake. The impact of input records on the remaining amount of currently ingestible elements means that recommendations often only reflect static goals and cannot reflect the user's real-time remaining nutritional space. On the other hand, when generating candidate recipes for white duck soup, traditional recipe recommendation systems usually only combine ingredients as components of the same level, lacking a mechanism for differentiating, adjusting, and scoring the main ingredient, auxiliary ingredients, and seasonings of white duck at different levels. This makes it difficult to retain the recipe structure and flavor of white duck soup while meeting nutritional constraints, easily leading to problems such as large nutritional deviations, flavor imbalances, and recommendations that do not match actual consumption scenarios. Therefore, it is difficult to meet the requirements of personalized dietary management and scenario-based adaptation for healthy white duck soup recipe recommendations.

[0004] In view of this, this application proposes a white duck soup health recipe recommendation system based on nutritional component analysis to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, this application provides the following technical solution: a white duck soup health recipe recommendation system based on nutritional component analysis, comprising: The data acquisition module collects basic user information and white duck soup ingredient attribute data, and performs data cleaning to obtain a white duck soup recommendation dataset, including user profile attribute parameters, ingredient attribute parameters, and consumption scenario data. The profile construction module performs scene profile recognition based on user profile attribute parameters and consumption scene data to determine the corresponding white duck soup consumption scene for the user; it constructs a scene-based user sub-profile based on the scene profile recognition results and outputs the scene profile parameter set. The nutritional constraint design module extracts scene template parameters corresponding to the white duck soup consumption scenario from the consumption scenario data, performs nutritional formula constraint mapping based on the scene template parameters and the scenario-based profile parameter group, and outputs an individualized nutritional constraint set. The nutrition budget calculation module obtains the historical nutrition intake parameters of the corresponding user, performs the balance of ingested nutrition based on the historical nutrition intake parameters and the individualized nutrition constraint set, determines the latest recommended remaining nutrition budget, and outputs the dynamic nutrition budget parameter set. The recipe combination module constructs a set of candidate recipes for white duck soup based on ingredient attribute parameters and dynamic nutrient budget parameter groups, performs recipe role adjustment on the set of candidate recipes for white duck soup, performs nutrient component matching analysis based on the adjusted candidate recipes for white duck soup, and outputs a candidate recipe scoring table. The recommendation result generation module sorts all candidate white duck soup recipes based on the candidate recipe scoring table, outputs the target white duck soup healthy recipe recommendation result, and sends it to the preset data terminal for storage and updating; the modules are connected to each other via wired and / or wireless means.

[0006] Furthermore, the methods for performing scene profiling recognition include: Extract age, physical health parameters, and recent medical records corresponding to each user from the user profile attribute parameters; identify the dining time, ambient temperature, and seasonal information corresponding to the user from the food scenario data; Calculate the overlap ratio between the age and physical health parameters of each user and the condition value range corresponding to the preset candidate scenario, and use the overlap ratio as the user matching feature value of the corresponding candidate scenario; The text of the user’s recent medical records, meal time, solar term information and ambient temperature are compared with the preset scene tags corresponding to the preset candidate scenes. The number of preset scene tags with the same semantics is counted and used as the semantic matching feature value of the corresponding candidate scene. The scene matching score between the corresponding user and each candidate scene is calculated based on the user matching feature value and the semantic matching feature value; the candidate scene corresponding to the maximum scene matching score is taken as the white duck soup consumption scene of the corresponding user.

[0007] Furthermore, the methods for generating scenario-based profile parameter sets include: The user's white duck soup consumption scenario is taken as the target candidate scenario for that user; a preset scenario parameter table is set, and the health element coefficients and scenario time intervals corresponding to the target candidate scenario are extracted from the preset scenario parameter table; Extract the age and physical health parameters of the corresponding user and compare them with the corresponding preset classification intervals, and output the classification parameters of the corresponding dimension for the user; Based on the hierarchical parameters and the corresponding health element coefficients of the target candidate scenarios, the target upper limit of scene health elements is matched. Extract the user's dietary restrictions and taste preferences data from the user profile attribute parameters, associate and match the scene time interval with the taste preference data, and construct a scene taste matching semantic vector; Compare the dietary restrictions data with the preset list of prohibited ingredients, label the names of the prohibited ingredients, and construct a semantic vector of the prohibited constraint ingredients. By integrating the target upper limit of health elements, taste matching semantic vectors, and disabled constraint food semantic vectors corresponding to users and target candidate scenarios, a set of scenario-based profile parameters is constructed.

[0008] Furthermore, the scenario template parameters include standardized nutritional baseline data and standardized catering constraint data.

[0009] Furthermore, methods for mapping nutritional formula constraints include: Extract the target upper limit of the scene health elements in the scene profile parameter group, identify the target upper limit of each nutritional element, and calculate the difference with the corresponding nutritional dimension parameter in the standard nutritional benchmark data to obtain the nutritional target deviation of each nutritional dimension parameter. The deviation of the nutritional target is compared with the preset deviation risk threshold. If the deviation of the nutritional target exceeds the preset deviation risk threshold, the corresponding nutritional dimension parameter in the standardized nutritional benchmark data is replaced with the target upper limit of each nutritional element in the scenario-based profile parameter group to obtain a preliminary constraint set. If the deviation of the nutritional target is not greater than the preset deviation risk threshold, the mean of the target upper limit and the corresponding nutritional dimension parameter for each nutritional element is calculated and updated to the preliminary constraint set. Extract the preset recipe cost limits and recipe material weight ranges from the standardized catering constraint data, and perform condition screening on the preliminary constraint set to output an individualized nutritional constraint set.

[0010] Furthermore, methods for achieving a balanced intake of nutrients include: Extract a portion of the user's historical nutrient intake parameters within a preset analysis period, and sum these parameters according to the type of nutrient element to obtain the cumulative intake of each type of nutrient element. Identify all nutrient intake timestamps in the historical nutrient intake parameters corresponding to each preset analysis period, and calculate the time difference between each nutrient intake timestamp and the current time. Match the nutrient type and time difference with the preset nutrient decay table, and output the metabolic decay ratio of each type of nutrient. Calculate the product of the cumulative intake and the metabolic attenuation ratio of the corresponding type of nutrient element to obtain the effective element retention of that type of nutrient element. The difference between the target upper limit of the corresponding type of nutrient element in the individualized nutrient constraint set and the effective element retention is calculated to obtain the remaining amount of ingestible elements for the corresponding analysis period. Obtain the preset allocation ratio corresponding to the white duck soup consumption scenario, calculate the product of the remaining amount of ingestible elements and the preset allocation ratio, and integrate the calculation results into an array to obtain a dynamic nutrition budget parameter group.

[0011] Furthermore, the methods for constructing a candidate recipe set for white duck soup include: Identify the recipe role tags belonging to each candidate ingredient in the ingredient attribute parameters, including main ingredient tags, auxiliary ingredient tags, and seasoning tags; at the same time, extract the corresponding nutritional component parameter group for each candidate ingredient and compare the nutritional component parameter group with the dynamic nutritional budget parameter group. If any nutrient parameter of a corresponding type is higher than the budget limit of that type of nutrient element, the corresponding candidate ingredients will be removed. At the same time, the candidate ingredients corresponding to the semantic vector of the prohibited constraint ingredients will be removed, and the candidate ingredients corresponding to the semantic vector of the scene taste matching will be retained. The preliminary set of filtered ingredients will be output. Ingredients labeled with "main ingredient" are used as primary ingredients, ingredients labeled with "auxiliary ingredient" are used as complementary ingredients, and ingredients labeled with "seasoning" are used as basic seasonings. These ingredients are then arranged and combined according to the proportions required for making white duck soup to obtain a set of candidate recipes for white duck soup.

[0012] Furthermore, the methods for implementing recipe role adjustments include: The nutritional parameters of each type of main ingredient, complementary ingredient and basic seasoning in each candidate recipe of white duck soup are summed to obtain the total amount of each type of nutritional component in the candidate recipe of white duck soup. The total amount of each type of nutrient in the corresponding white duck soup candidate recipe is compared with the preset reasonable nutrient range. If the total amount of a corresponding type of nutrient exceeds the preset reasonable nutrient range, then the nutrient element of that type is regarded as a deviating nutrient element. Calculate the proportion of deviating nutrients in the main ingredients, complementary ingredients, and basic seasonings, and mark the corresponding ingredients with the highest proportion as the role adjustment ingredients. By repeatedly adjusting the actual amount of ingredients used in the corresponding white duck soup candidate recipe, until the total amount of each type of nutrient in the white duck soup candidate recipe is within the preset reasonable component range, the adjusted white duck soup candidate recipe is obtained.

[0013] Furthermore, methods for conducting nutrient matching analysis include: Extract the total amount of each type of adjusted nutrient in each candidate recipe of adjusted white duck soup, calculate the difference between it and the budget upper limit of the nutrient element of that type, and take the absolute value to obtain the analysis deviation of the nutrient element of the corresponding type. Obtain the nutritional evaluation weight of each type of nutrient element, and calculate the matching loss of that type of nutrient element by multiplying the analysis bias of the corresponding type of nutrient element by the nutritional evaluation weight. The total loss of all types of nutrients is calculated as the nutrient loss score; the formula score of the adjusted white duck soup candidate formula is obtained by subtracting the nutrient loss score from the full score of the preset formula evaluation. All adjusted white duck soup candidate recipes in the white duck soup candidate recipe set are incorporated into a preset table structure for storage, resulting in a candidate recipe scoring table.

[0014] Furthermore, the methods for generating the recommended healthy recipe for target white duck soup include: Candidate white duck soup recipes with scores within the preset recommended recipe score range are selected, sorted in descending order of recipe score, and integrated into the target white duck soup healthy recipe recommendation results.

[0015] The technical effects and advantages of the white duck soup health recipe recommendation system based on nutritional component analysis proposed in this application are as follows: By collecting user basic information, white duck soup ingredient attribute data, and consumption scenario data, and performing data cleaning, a white duck soup recommendation dataset was obtained, and a white duck soup healthy recipe recommendation system was constructed. Compared with existing experience, by identifying user age, physical health parameters, meal time, ambient temperature, and solar term information, the system eliminates the problem of using the same recommendation logic for the same user in different white duck soup consumption scenarios. By constraining and mapping scenario-based user sub-profiles with scenario template parameters, the system avoids direct conflicts between group-level catering standards and individual health needs. By introducing historical nutrient intake time decay, the system avoids the problem of static accumulation leading to distortion of the remaining nutrient budget. By performing recipe role screening, adjustment, and scoring ranking on white duck soup candidate recipes, the system avoids indiscriminate deletion of ingredients during ingredient adjustments that could lead to flavor imbalance, thus improving the scenario adaptability and nutritional accuracy of the white duck soup healthy recipe recommendations. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a white duck soup health recipe recommendation system based on nutritional component analysis, as proposed in this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Example 1 Please see Figure 1 As shown in this embodiment, a white duck soup health recipe recommendation system based on nutritional component analysis includes: The data acquisition module collects basic user information and white duck soup ingredient attribute data, and performs data cleaning to obtain a white duck soup recommendation dataset, including user profile attribute parameters, ingredient attribute parameters, and consumption scenario data. The profile construction module performs scene profile recognition based on user profile attribute parameters and consumption scene data to determine the corresponding white duck soup consumption scene for the user; it constructs a scene-based user sub-profile based on the scene profile recognition results and outputs the scene profile parameter set. The nutritional constraint design module extracts scene template parameters corresponding to the white duck soup consumption scenario from the consumption scenario data, performs nutritional formula constraint mapping based on the scene template parameters and the scenario-based profile parameter group, and outputs an individualized nutritional constraint set. The nutrition budget calculation module obtains the historical nutrition intake parameters of the corresponding user, performs the balance of ingested nutrition based on the historical nutrition intake parameters and the individualized nutrition constraint set, determines the latest recommended remaining nutrition budget, and outputs the dynamic nutrition budget parameter set. The recipe combination module constructs a set of candidate recipes for white duck soup based on ingredient attribute parameters and dynamic nutrient budget parameter groups, performs recipe role adjustment on the set of candidate recipes for white duck soup, performs nutrient component matching analysis based on the adjusted candidate recipes for white duck soup, and outputs a candidate recipe scoring table. The recommendation result generation module sorts all candidate white duck soup recipes based on the candidate recipe scoring table, outputs the target white duck soup healthy recipe recommendation result, and sends it to the preset data terminal for storage and updating; the modules are connected to each other via wired and / or wireless means.

[0019] In this embodiment, a big data acquisition terminal is used to obtain relevant data of users who need to recommend healthy white duck soup recipes, as well as specific ingredient attribute data related to white duck soup. Missing value imputation and filtering are then used to obtain a higher-quality white duck soup recommendation dataset. User profile attribute parameters include data reflecting the user's profile information, such as age, gender, height, weight, blood pressure, uric acid, blood sugar, dietary restrictions, taste preferences, and recent medical records. Ingredient attribute parameters include data reflecting the inherent properties of the ingredients, such as ingredient name, category, role tag, unit weight, unit nutritional components, and cost. Consumption scenario data refers to data reflecting the dining scenario, such as dining time, seasonal weather information, ambient temperature, ambient humidity, and dining location.

[0020] The methods for performing scene profiling recognition include: Extract the age, physical health parameters, and recent medical records corresponding to each user from the user profile attribute parameters.

[0021] The physical health parameters include parameters that reflect physical health status, such as blood pressure, uric acid, and blood sugar; recent medical records refer to text information that records whether the corresponding user has recently been ill and sought medical treatment.

[0022] Identify the dining time, ambient temperature, and seasonal information corresponding to the user in the food consumption scenario data.

[0023] Meal times include, for example, breakfast, lunch, dinner, or a specific time period; seasonal information includes the season and weather conditions.

[0024] Calculate the overlap ratio between the age and physical health parameters of each user and the condition value range corresponding to the preset candidate scenario, and use the overlap ratio as the user matching feature value of the corresponding candidate scenario.

[0025] In this embodiment, the preset candidate scenarios refer to various scenarios for consuming white duck soup, such as postoperative recovery scenarios, elderly dining scenarios, seasonal tonic scenarios, or low-fat control scenarios. These scenarios are set based on relevant data on consuming white duck soup obtained from a big data acquisition terminal.

[0026] By matching each user's age and physical health parameters with the conditional value ranges related to age and corresponding physical health parameters in each preset candidate scenario, overlapping conditional value ranges that can be matched are identified, and the proportion of the overlapping part to all corresponding conditional value ranges is calculated as the user matching feature value between the user and the corresponding candidate scenario.

[0027] The system compares the user's recent medical records, meal times, seasonal information, and ambient temperature with the preset scene tags corresponding to the preset candidate scenes, and counts the number of preset scene tags with the same semantics, which is used as the semantic matching feature value of the corresponding candidate scene.

[0028] Semantic similarity refers to the fact that the preset scene tags can represent the scene reflected by the user's recent medical records, meal time, solar term information, and ambient temperature. For example, if the user's recent medical records include post-operative recovery records, the meal time is dinner time, and the solar term information is winter, then the preset scene tags related to post-operative recovery, dinner, and winter nourishment can all be counted as semantically similar. The preset scene tags refer to the representative tags corresponding to each preset candidate scene used to describe the situation of that scene.

[0029] The scene matching score between the corresponding user and each candidate scene is calculated based on the user matching feature value and the semantic matching feature value.

[0030] The formula for calculating the scene matching score between the corresponding user and each candidate scene is as follows: ;in This represents the scenario matching score between a user and each candidate scenario; This indicates the user's matching feature value with the user in the corresponding candidate scenario; This represents the semantic matching feature value between the user and the corresponding candidate scenario; and These represent the weights, and their sum is 1. The specific values ​​are set based on the information contribution ratio of user matching feature values ​​and semantic matching feature values ​​in historical scene matching score calculation experience. It should be noted that the user matching feature values ​​and semantic matching feature values ​​in the above calculation formula have been normalized and there is no difference in dimensions.

[0031] The candidate scene corresponding to the maximum scene matching score is taken as the white duck soup consumption scene for the corresponding user. The candidate scene with the highest scene matching score is selected to ensure that the candidate scene is the most suitable for the corresponding user, thus obtaining the white duck soup consumption scene for the user.

[0032] The methods for generating contextualized profile parameter sets include: The user's white duck soup consumption scenario is taken as the target candidate scenario for that user; a preset scenario parameter table is set, and the health element coefficients and scenario time intervals corresponding to the target candidate scenario are extracted from the preset scenario parameter table.

[0033] In this embodiment, scenario parameters for an ideal health scenario are set based on relevant medical theories, forming a preset scenario parameter table. By matching the target candidate scenario with the preset scenario parameter table, the corresponding health element coefficients and scenario time intervals are identified. The health element coefficients include parameters such as protein, fat, purine, and calories in the corresponding target candidate scenario. The scenario time interval is used to represent the appropriate dining time range for the target candidate scenario.

[0034] Extract the age and physical health parameters of the corresponding user and compare them with the corresponding preset classification intervals, and output the classification parameters of the corresponding dimension of the user.

[0035] The corresponding preset grading interval refers to the set grading interval corresponding to each dimension of the age and physical health parameters. Based on relevant medical theoretical knowledge, it outputs the grading parameters of all dimensions of the user's age and physical health parameters. For example, blood pressure is matched with the corresponding preset grading interval to output blood pressure grading parameters, and age is matched with the corresponding preset grading interval to output grading parameters for age groups such as youth, middle-aged, or elderly.

[0036] Based on the hierarchical parameters and the corresponding health element coefficients of the target candidate scenarios, the target upper limit of scene health elements is matched.

[0037] The classification parameters determine the level of the corresponding element or age dimension, and the target upper limit value corresponding to each dimension and the corresponding level is found in the preset scenario parameter table. For example, when the classification parameter corresponds to the elderly classification and the health element coefficient corresponds to low salt, a lower sodium target upper limit value can be matched. When the classification parameter corresponds to the postoperative recovery classification and the health element coefficient corresponds to high protein content, a higher protein target upper limit value can be matched.

[0038] Extract the user's dietary restrictions and taste preferences from the user profile attribute parameters, associate and match the scene time interval with the taste preference data, and construct a scene taste matching semantic vector.

[0039] In this embodiment, dietary restriction data refers to descriptive information about foods that should be avoided, such as aversion to cilantro, allergy to a certain type of food, or dietary restriction due to medical treatment; taste preference data refers to descriptive information reflecting the user's preferred taste, such as light, mildly spicy, or medicinal cuisine preferences, or the requirement to eat certain foods due to medical advice; by associating the time period of a scenario with the taste preference data, the taste type belonging to that time period is determined, such as associating dinner time with light taste.

[0040] The dietary restrictions data are compared with the preset list of prohibited ingredients, the names of the prohibited ingredients are marked, and a semantic vector of the prohibited ingredients is constructed.

[0041] The preset prohibited food list is a list of foods that cannot be eaten based on the corresponding user's medical advice or allergies. By comparing the dietary restrictions data with the preset prohibited food list, the union result is output, and the names of all prohibited foods that the corresponding user cannot eat are marked. These names are then combined into a vector form to form a constraint food semantic vector.

[0042] By integrating the target upper limit of health elements, taste matching semantic vectors, and disabled constraint food semantic vectors corresponding to users and target candidate scenarios, a set of scenario-based profile parameters is constructed.

[0043] This involves storing the target upper limit of scene health elements, taste matching semantic vector, and disabled constraint food semantic vector for each user and the corresponding target candidate scene in a unified vector form, and then concatenating them into a scene-based profile parameter group that matches the corresponding user with the target candidate scene.

[0044] The scenario template parameters include standardized nutritional baseline data and standardized catering constraint data.

[0045] In this embodiment, the standardized nutritional baseline data refers to the set of baseline values ​​for each type of nutrient element in a pre-set white duck soup consumption scenario in institutions such as nursing homes or rehabilitation centers, which is used to represent the basic levels of various types of nutrients allowed in this scenario; the standardized catering constraint data refers to the restriction data used to limit the white duck soup recipe, such as the standard weight of a single serving of white duck soup, recipe cost restrictions, serving size restrictions, and common or prohibited ingredients.

[0046] Methods for performing nutritional formula constraint mapping include: Extract the target upper limit of the scene health elements in the scene profile parameter group, identify the target upper limit of each nutritional element, and calculate the difference between the target upper limit and the corresponding nutritional dimension parameter in the standard nutritional benchmark data to obtain the nutritional target deviation of each nutritional dimension parameter.

[0047] The method involves extracting the target upper limit corresponding to each nutritional element from the target upper limit of the scene health elements, and calculating the difference between this target upper limit and the benchmark value of the same type of nutritional element in the standardized nutritional benchmark data to obtain the nutritional target deviation of the corresponding nutritional dimension parameter. This deviation is used to indicate the difference between the target value of each type of nutritional element in the scene-based profile parameter group and the benchmark value of the corresponding type of nutritional element. The larger the deviation, the more obvious the deviation between the individual's nutritional needs and the standard.

[0048] The deviation of the nutritional target is compared with the preset deviation risk threshold. If the deviation exceeds the preset deviation risk threshold, the corresponding nutritional dimension parameter in the standardized nutritional benchmark data is replaced with the target upper limit of each nutritional element in the scenario-based profile parameter group to obtain a preliminary constraint set.

[0049] The system sets a preset deviation risk threshold based on the ideal nutritional intake situation. If the deviation of the nutritional target is higher than the preset deviation risk threshold, it indicates that the individual's nutritional needs are more obvious and may conflict with the nutritional intake requirements of certain scenarios. Therefore, the parameters of the corresponding nutritional dimensions in the standardized nutritional benchmark data are replaced with the target upper limit of each nutritional element in the scenario-based profile parameter group corresponding to the specific white duck soup consumption scenario, so as to obtain the preliminary constraint set.

[0050] For example, if a user has a clear need for low sodium, but the standard sodium limit set by the institution is high, if the difference exceeds the preset deviation risk threshold, the low sodium target limit corresponding to the specific scenario will be directly used as the initial constraint value, thereby avoiding the subsequent combination of white duck soup recipes from causing the sodium content to continue to be too high.

[0051] If the deviation of the nutritional target is not greater than the preset deviation risk threshold, the mean of the target upper limit and the corresponding nutritional dimension parameter for each nutritional element is calculated and updated to the preliminary constraint set.

[0052] When the deviation from the nutritional target is not greater than the preset deviation risk threshold, it indicates that the user's individual needs are not significantly different from the standard. In order to ensure that the constraints are not too strict or too relaxed, a relatively compromise constraint value is formed by averaging the target upper limit of each type of nutritional element with the parameters of the corresponding nutritional dimension in the standardized nutritional benchmark data. The calculation results are then updated to the preliminary constraint set, replacing the parameters of the nutritional elements corresponding to the case where the deviation from the nutritional target is not greater than the preset deviation risk threshold.

[0053] Extract the preset recipe cost limits and recipe material weight ranges from the standardized catering constraint data, and perform condition screening on the preliminary constraint set to output an individualized nutritional constraint set.

[0054] In this embodiment, due to limited food resources, there are restrictions on formula cost and formula material weight range. The formula cost restriction is used to limit the economic efficiency of the white duck soup formula in actual serving, and to prevent the use of too many high-cost ingredients from leading to food resource shortages and budget shortfalls. The formula material weight range refers to the weight range of various ingredients in a single serving of white duck soup, such as the specific weight of duck meat and the weight of other auxiliary ingredients, to prevent the use of a certain type of ingredient from causing the weight of the whole soup to not meet the serving standard.

[0055] By using preset formula cost limits and formula material weight ranges as constraint screening conditions, food-related parameters that exceed the formula cost limits are eliminated, and food-related parameters that exceed the formula material weight range are recalculated or only the amount of food that meets the formula material weight range is directly obtained, ensuring that the food can simultaneously meet the cost and weight constraints, thus obtaining an individualized nutritional constraint set.

[0056] Methods for achieving a balanced intake of nutrients include: Extract a portion of the user's historical nutrient intake parameters within a preset analysis period, and then sum these parameters according to the type of nutrient element to obtain the cumulative intake of each type of nutrient element.

[0057] In this embodiment, the preset analysis period refers to, for example, the breakfast period or a specific time period after a meal that needs to be analyzed. Some historical nutrient intake parameters are the intake of various nutrients after a meal during the preset analysis period of the historical record, including, for example, protein, fat, sodium, and purines.

[0058] For each preset analysis period, the historical nutrient intake parameters are divided according to the type of nutrient element, and the cumulative sum of the intake of each type of nutrient element is calculated. The cumulative sum means that from any analysis period to the current real time, the intake of nutrient elements is accumulated to obtain the cumulative intake of each type of nutrient element. For example, if the analysis period is breakfast, only the intake within that period is extracted. If the next analysis period is lunch, the intake of lunch and breakfast periods needs to be accumulated.

[0059] Identify all nutrient intake timestamps in the historical nutrient intake parameters corresponding to each preset analysis period, and calculate the time difference between each nutrient intake timestamp and the current time.

[0060] The process involves identifying the specific intake timestamps of certain historical nutrient intake parameters for each analysis period, and calculating the difference between each specific intake timestamp and the current real-time time to obtain the time difference. This time difference refers to the length of time that has elapsed since each historical nutrient intake record was recorded. For example, if the specific intake timestamp for a certain intake record is noon, and the current real-time time is 7 pm, then the time difference can be expressed as seven hours.

[0061] Match the nutrient type and time difference with a preset nutrient decay table to output the metabolic decay ratio of each type of nutrient.

[0062] The preset nutrient decay table is a data table based on biological and medical theories that reflects the decay of the intake of each nutrient element over time due to metabolism. Each nutrient element type is used as an index to query the metabolic decay ratio of the corresponding nutrient element type and the time difference from the preset nutrient decay table. This metabolic decay ratio is used to indicate the degree to which a certain element can still retain its effective influence after a certain period of time.

[0063] The effective element retention of a certain type of nutrient is obtained by multiplying the cumulative amount ingested by the metabolic attenuation ratio of the corresponding nutrient element.

[0064] The formula for calculating the effective element retention is as follows: ;in Indicates the amount of available element retention for a certain type of nutrient; This indicates the cumulative amount of this type of nutrient that has been ingested; This indicates the metabolic decay rate of the corresponding type of nutrient element; the effective element retention amount refers to the actual retention value of the cumulative amount ingested over a historical period that still has an effect on the user's nutritional status at the current moment after being metabolically decayed over time.

[0065] The difference between the target upper limit of the corresponding type of nutrient element in the individualized nutrient constraint set and the effective element retention amount is calculated to obtain the remaining amount of ingestible elements for the corresponding analysis period.

[0066] The method involves calculating the difference between the target upper limit for each type of nutrient and the effective retention of that nutrient. This determines the remaining amount of that nutrient that a user can still consume within the corresponding analysis period after deducting historical effective retention. For example, if a user's target upper limit for sodium in the current period is low, but the historical effective retention is already close to that upper limit, then the remaining amount of the nutrient is small, which serves as a reminder to control the sodium content in subsequent white duck soup recipes.

[0067] Obtain the preset allocation ratio corresponding to the white duck soup consumption scenario, calculate the product of the remaining amount of ingestible elements and the preset allocation ratio, and integrate the calculation results into an array to obtain a dynamic nutrition budget parameter group.

[0068] The preset allocation ratio refers to the proportion of the remaining elements that a user can ingest under ideal health conditions in the corresponding white duck soup consumption scenario, which can be allocated to the white duck soup consumed at this time. Therefore, by calculating the product of the remaining ingestible elements and the preset allocation ratio, the upper limit value of each type of nutrient element in the white duck soup at the current moment can be obtained. The upper limit values ​​of all types of elements are integrated into the dynamic nutrient budget parameter group recommended for this white duck soup recipe.

[0069] Methods for constructing a candidate recipe set for white duck soup include: Identify the recipe role tags belonging to each candidate ingredient in the ingredient attribute parameters, including main ingredient tags, auxiliary ingredient tags, and seasoning tags.

[0070] In this embodiment, the main ingredient label is used to mark ingredients such as duck meat, duck carcass, or duck leg that form the main body of white duck soup; the auxiliary ingredient label is used to mark ingredients such as yam, winter melon, and lotus root that are used to supplement the soup; and the seasoning label is used to mark ingredients such as ginger, scallions, salt, and cooking wine that are used to adjust the flavor.

[0071] At the same time, the nutritional component parameter set for each candidate ingredient is extracted and compared with the dynamic nutritional budget parameter set.

[0072] The nutrient composition parameter group is used to represent the specific content of all nutrients per unit mass of each ingredient; comparing the nutrient composition parameter group with the dynamic nutrient budget parameter group means comparing each corresponding nutrient composition parameter of each candidate ingredient with the corresponding nutrient budget upper limit in the dynamic nutrient budget parameter group for the current white duck soup consumption scenario.

[0073] If any nutrient parameter of a corresponding type exceeds the budget limit for that type of nutrient element, the corresponding candidate ingredients will be removed. At the same time, the candidate ingredients corresponding to the semantic vector of the prohibited constraint ingredients will be removed, and the candidate ingredients corresponding to the semantic vector of the scene taste matching will be retained. The preliminary set of filtered ingredients will be output.

[0074] In this embodiment, if the nutrient composition parameter of any type of nutrient element is higher than the upper limit of the nutrient element budget for that type of nutrient element, in order to avoid the content of a single type of nutrient element exceeding the limit when combining formulas in the future, the alternative ingredients corresponding to the excess elements are removed from all alternative ingredients.

[0075] After removing candidate ingredients that exceed the limits, in order to match the contextualized user sub-profile of the corresponding user, the candidate ingredients corresponding to the semantic vector of the prohibited constraint ingredients are removed from the remaining candidate ingredients. After removing the candidate ingredients corresponding to the prohibited constraint ingredients and the semantic vector of the prohibited constraint ingredients, the candidate ingredients that match the semantic vector of the corresponding user's contextual taste are retained first, thus obtaining a preliminary set of selected ingredients.

[0076] Ingredients labeled with "main ingredient" are used as primary ingredients, ingredients labeled with "auxiliary ingredient" are used as complementary ingredients, and ingredients labeled with "seasoning" are used as basic seasonings. These ingredients are then arranged and combined according to the proportions required for making white duck soup to obtain a set of candidate recipes for white duck soup.

[0077] The main ingredients constitute the core of white duck soup, while complementary ingredients are used to adjust the nutritional structure of the white duck soup. Basic seasonings are used to adjust the taste and flavor of the white duck soup. The white duck soup preparation ratio refers to the reasonable ratio of ingredients for white duck soup based on culinary knowledge. By using the white duck soup preparation ratio as a basis, all ingredients are arranged and combined to obtain several candidate recipes that can meet the basic structure of white duck soup. All candidate recipes are integrated to obtain a set of candidate recipes for white duck soup.

[0078] The methods for adjusting the role of the recipe include: The nutritional parameters of each type of main ingredient, complementary ingredient, and basic seasoning in each candidate recipe for white duck soup are summed to obtain the total amount of each type of nutritional component in the candidate recipe for white duck soup.

[0079] The formula for calculating the summation is as follows: ;in This represents the total amount of nutrients of any type in a candidate recipe for white duck soup; This indicates the quantity of the main ingredients in the recipe. Since the ingredients are weighed and purchased by unit weight, each portion of ingredients can be divided into several portions. This indicates the first [item] in the formula. Nutritional parameters of the corresponding types of nutrients in the main ingredients; This indicates the quantity of ingredients in the recipe; This indicates the first [item] in the formula. Nutritional parameters of the corresponding types of nutrients in the ingredients; This indicates the quantity of basic seasonings in the recipe; This indicates the first [item] in the formula. Nutritional parameters of the corresponding type of nutrients in a basic seasoning.

[0080] The total amount of nutrients is used to reflect the total content of each type of nutrient element in the corresponding white duck soup candidate recipe, and to reflect the overall nutrient distribution of the white duck soup corresponding to the white duck soup candidate recipe.

[0081] The total amount of each type of nutrient in the corresponding white duck soup candidate recipe is compared with the preset reasonable nutrient range. If the total amount of a corresponding type of nutrient exceeds the preset reasonable nutrient range, then the nutrient element of that type is regarded as a deviating nutrient element.

[0082] The preset reasonable component range refers to the nutritional control range for white duck soup set based on nutritional and health requirements. It is used to indicate the acceptable range of each nutrient element for the candidate white duck soup formula. If the content of one or more types of nutrients is higher than the preset reasonable component range, it means that the corresponding nutrient element has deviated from the health requirements, and the corresponding type of nutrient element is regarded as the deviated nutrient element.

[0083] Calculate the proportion of deviated nutrients in the main ingredients, complementary ingredients, and basic seasonings, and mark the corresponding ingredients with the highest proportion as the role adjustment ingredients.

[0084] The content ratio refers to the proportion of the source of the deviated nutrient element in different food roles. The food that contributes the most to the deviated nutrient element among the main food, the matching food and the basic seasoning is selected and used as the role adjustment food.

[0085] By repeatedly adjusting the actual amount of ingredients used in the corresponding white duck soup candidate recipe, until the total amount of each type of nutrient in the white duck soup candidate recipe is within the preset reasonable component range, the adjusted white duck soup candidate recipe is obtained.

[0086] The actual amount used refers to the specific weight, volume, or number of slices of the corresponding ingredient in the candidate recipe. Repeated adjustment refers to making multiple corrections to the actual amount of the ingredient used according to a preset adjustment ratio based on the distribution of nutrients in the corresponding ingredients. For example, reducing the amount of duck skin, reducing the amount of salt, or reducing the amount of a certain type of high-fat auxiliary material. After each correction, the total amount of each type of nutrient in the candidate recipe is recalculated until the total amount of all nutrients falls within the preset reasonable range. Through repeated adjustment, a white duck soup recipe that meets the nutritional control requirements is obtained.

[0087] Methods for conducting nutritional component matching analysis include: Extract the total amount of adjusted nutrients for each type in each candidate recipe for adjusted white duck soup, calculate the difference between this amount and the upper limit of the budget for that type of nutrient element, and take the absolute value to obtain the analytical deviation of the corresponding type of nutrient element.

[0088] The adjusted total nutrient content refers to the final total amount of each type of nutrient element obtained after adjustment of the corresponding white duck soup candidate recipe. The deviation of the candidate recipe from the budgeted amount of each type of nutrient element is obtained by calculating the difference between the adjusted total nutrient content and the budgeted upper limit of the corresponding type of nutrient element and taking the absolute value. This value is used as the analysis deviation. For example, if the fat content that a candidate recipe can provide is close to the budgeted upper limit of the element, the analysis deviation is small.

[0089] Obtain the nutritional evaluation weight for each type of nutrient element, and calculate the matching loss of that type of nutrient element by multiplying the analysis bias of the corresponding nutrient element by its nutritional evaluation weight.

[0090] The nutritional evaluation weight refers to the weight set based on nutritional and health requirements for evaluating each type of nutrient element related to white duck soup. This value is used to represent the difference in importance of different types of nutrient elements in the healthy white duck soup formula. By calculating the product of the nutritional evaluation weight and the analysis deviation, different types of nutrient elements are weighted according to their importance, so that the deviation of more important nutrient elements has a greater impact in subsequent scoring, and the matching loss of the corresponding type of nutrient element is obtained.

[0091] The total loss of all types of nutrients is calculated as the nutrient loss score.

[0092] The total matching loss refers to the sum of the matching losses of all types of nutrients in a candidate formula. It is used to represent the overall deviation of the candidate formula from the health target at the nutritional level and serves as the nutritional loss score of the corresponding candidate formula. The smaller the nutritional loss score, the closer the candidate formula is to the target nutritional control requirements and the more suitable it is as a formula recommendation result; conversely, the larger the score, the greater the deviation of the candidate formula from one or more nutrients.

[0093] The formula score of the adjusted white duck soup candidate formula is obtained by subtracting the nutrient loss score from the full score of the preset formula evaluation.

[0094] In this embodiment, a predefined score is used as the maximum score for the recommended formula. This maximum score is greater than the nutrient loss score of any candidate formula. The nutrient loss score and the maximum score are first unified to the same order of magnitude. The formula score of the candidate formula is obtained by subtracting the nutrient loss score of the corresponding candidate formula from the maximum score.

[0095] The candidate recipes for white duck soup with lower nutrient loss scores have higher recipe scores, indicating that the candidate recipes are closer to the nutritional and health requirements of the current white duck soup consumption scenario.

[0096] All adjusted white duck soup candidate recipes in the white duck soup candidate recipe set are incorporated into a preset table structure for storage, resulting in a candidate recipe scoring table. In this embodiment, a table structure is constructed to store all adjusted white duck soup candidate recipes and their corresponding recipe scores, resulting in a candidate recipe evaluation table.

[0097] The methods used to generate the recommended healthy recipe for target white duck soup include: Candidate white duck soup recipes with scores within the preset recommended recipe score range are selected, sorted in descending order of recipe score, and integrated into the target white duck soup healthy recipe recommendation results.

[0098] The process involves setting a preset recommended recipe score range based on nutritional and health requirements, screening candidate white duck soup recipes whose scores fall within this preset range, and arranging them in descending order of score. This prioritizes recommending candidate white duck soup recipes that better match the current white duck soup consumption scenario and the corresponding user. The sorted recipe sequence is then output as the target healthy white duck soup recipe recommendation result.

[0099] This embodiment constructs a white duck soup healthy recipe recommendation system by collecting user basic information, white duck soup ingredient attribute data, and consumption scenario data, and then cleaning the data to obtain a white duck soup recommendation dataset. Compared with existing experience, by identifying user age, physical health parameters, meal time, ambient temperature, and solar term information, it eliminates the problem of using the same recommendation logic for the same user in different white duck soup consumption scenarios. By constraining and mapping scenario-based user sub-profiles with scenario template parameters, it avoids direct conflicts between group-level catering standards and individual health needs. By introducing historical nutrient intake time decay, it avoids the problem of static accumulation leading to distortion of the remaining nutrient budget. By performing recipe role screening, adjustment, and scoring and ranking on white duck soup candidate recipes, it avoids indiscriminate deletion of ingredients during ingredient adjustments that would lead to flavor imbalance, thus improving the scenario adaptability and nutritional accuracy of the white duck soup healthy recipe recommendation.

[0100] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0101] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0102] Although embodiments of this application have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of this application, the scope of which is defined by the claims and their equivalents.

Claims

1. A system for recommending healthy recipes for white duck soup based on nutritional component analysis, characterized in that, include: The data acquisition module collects basic user information and white duck soup ingredient attribute data, and performs data cleaning to obtain a white duck soup recommendation dataset, including user profile attribute parameters, ingredient attribute parameters, and consumption scenario data. The profile construction module performs scene profile recognition based on user profile attribute parameters and consumption scene data to determine the corresponding white duck soup consumption scene for the user. Based on the scene profile recognition results, construct scene-based user sub-profiles and output scene profile parameter sets; The nutritional constraint design module extracts scene template parameters corresponding to the white duck soup consumption scenario from the consumption scenario data, performs nutritional formula constraint mapping based on the scene template parameters and the scenario-based profile parameter group, and outputs an individualized nutritional constraint set. The nutrition budget calculation module obtains the historical nutrition intake parameters of the corresponding user, performs the balance of ingested nutrition based on the historical nutrition intake parameters and the individualized nutrition constraint set, determines the latest recommended remaining nutrition budget, and outputs the dynamic nutrition budget parameter set. The recipe combination module constructs a set of candidate recipes for white duck soup based on ingredient attribute parameters and dynamic nutrient budget parameter groups, performs recipe role adjustment on the set of candidate recipes for white duck soup, performs nutrient component matching analysis based on the adjusted candidate recipes for white duck soup, and outputs a candidate recipe scoring table. The recommendation result generation module sorts all candidate white duck soup recipes based on the candidate recipe scoring table, outputs the target white duck soup healthy recipe recommendation result, and sends it to the preset data terminal for storage and updating; the modules are connected to each other via wired and / or wireless means.

2. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 1, characterized in that, The methods for performing scene profile recognition include: Extract age, physical health parameters, and recent medical records corresponding to each user from the user profile attribute parameters; identify the dining time, ambient temperature, and seasonal information corresponding to the user from the food scenario data; Calculate the overlap ratio between the age and physical health parameters of each user and the condition value range corresponding to the preset candidate scenario, and use the overlap ratio as the user matching feature value of the corresponding candidate scenario; The text of the user’s recent medical records, meal time, solar term information and ambient temperature are compared with the preset scene tags corresponding to the preset candidate scenes. The number of preset scene tags with the same semantics is counted and used as the semantic matching feature value of the corresponding candidate scene. The scene matching score between the corresponding user and each candidate scene is calculated based on the user matching feature value and the semantic matching feature value; the candidate scene corresponding to the maximum scene matching score is taken as the white duck soup consumption scene of the corresponding user.

3. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 2, characterized in that, The generation methods for the contextualized profile parameter group include: The user's white duck soup consumption scenario is taken as the target candidate scenario for that user; a preset scenario parameter table is set, and the health element coefficients and scenario time intervals corresponding to the target candidate scenario are extracted from the preset scenario parameter table; Extract the age and physical health parameters of the corresponding user and compare them with the corresponding preset classification intervals, and output the classification parameters of the corresponding dimension for the user; Based on the hierarchical parameters and the corresponding health element coefficients of the target candidate scenarios, the target upper limit of scene health elements is matched. Extract the user's dietary restrictions and taste preferences data from the user profile attribute parameters, associate and match the scene time interval with the taste preference data, and construct a scene taste matching semantic vector; Compare the dietary restrictions data with the preset list of prohibited ingredients, label the names of the prohibited ingredients, and construct a semantic vector of the prohibited constraint ingredients. By integrating the target upper limit of health elements, taste matching semantic vectors, and disabled constraint food semantic vectors corresponding to users and target candidate scenarios, a set of scenario-based profile parameters is constructed.

4. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 3, characterized in that, The scenario template parameters include standardized nutritional benchmark data and standardized catering constraint data.

5. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 4, characterized in that, The methods for performing nutritional formula constraint mapping include: Extract the target upper limit of the scene health elements in the scene profile parameter group, identify the target upper limit of each nutritional element, and calculate the difference with the corresponding nutritional dimension parameter in the standard nutritional benchmark data to obtain the nutritional target deviation of each nutritional dimension parameter. The deviation of the nutritional target is compared with the preset deviation risk threshold. If the deviation of the nutritional target exceeds the preset deviation risk threshold, the corresponding nutritional dimension parameter in the standardized nutritional benchmark data is replaced with the target upper limit of each nutritional element in the scenario-based profile parameter group to obtain a preliminary constraint set. If the deviation of the nutritional target is not greater than the preset deviation risk threshold, the mean of the target upper limit and the corresponding nutritional dimension parameter for each nutritional element is calculated and updated to the preliminary constraint set. Extract the preset recipe cost limits and recipe material weight ranges from the standardized catering constraint data, and perform condition screening on the preliminary constraint set to output an individualized nutritional constraint set.

6. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 5, characterized in that, The methods for achieving a balanced intake of nutrients include: Extract some historical nutrient intake parameters of the corresponding user within the preset analysis period, and accumulate these historical nutrient intake parameters according to the type of nutrient element to obtain the cumulative intake of each type of nutrient element. Identify all nutrient intake timestamps in the historical nutrient intake parameters corresponding to each preset analysis period, and calculate the time difference between each nutrient intake timestamp and the current time. Match the nutrient type and time difference with the preset nutrient decay table, and output the metabolic decay ratio of each type of nutrient. Calculate the product of the cumulative intake and the metabolic attenuation ratio of the corresponding type of nutrient element to obtain the effective element retention of that type of nutrient element. The difference between the target upper limit of the corresponding type of nutrient element in the individualized nutrient constraint set and the effective element retention is calculated to obtain the remaining amount of ingestible elements for the corresponding analysis period. Obtain the preset allocation ratio corresponding to the white duck soup consumption scenario, calculate the product of the remaining amount of ingestible elements and the preset allocation ratio, and integrate the calculation results into an array to obtain a dynamic nutrition budget parameter group.

7. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 6, characterized in that, The method for constructing the candidate recipe set for white duck soup includes: Identify the recipe role tags belonging to each candidate ingredient in the ingredient attribute parameters, including main ingredient tags, auxiliary ingredient tags, and seasoning tags; at the same time, extract the corresponding nutritional component parameter group for each candidate ingredient and compare the nutritional component parameter group with the dynamic nutritional budget parameter group. If any nutrient parameter of a corresponding type is higher than the budget limit of that type of nutrient element, the corresponding candidate ingredients will be removed. At the same time, the candidate ingredients corresponding to the semantic vector of the prohibited constraint ingredients will be removed, and the candidate ingredients corresponding to the semantic vector of the scene taste matching will be retained. The preliminary set of filtered ingredients will be output. Ingredients labeled with "main ingredient" are used as primary ingredients, ingredients labeled with "auxiliary ingredient" are used as complementary ingredients, and ingredients labeled with "seasoning" are used as basic seasonings. These ingredients are then arranged and combined according to the proportions required for making white duck soup to obtain a set of candidate recipes for white duck soup.

8. The white duck soup health recipe recommendation system based on nutritional component analysis according to claim 7, characterized in that, The methods for adjusting the execution of the recipe role include: The nutritional parameters of each type of main ingredient, complementary ingredient and basic seasoning in each candidate recipe of white duck soup are summed to obtain the total amount of each type of nutritional component in the candidate recipe of white duck soup. The total amount of each type of nutrient in the corresponding white duck soup candidate recipe is compared with the preset reasonable nutrient range. If the total amount of a corresponding type of nutrient exceeds the preset reasonable nutrient range, then the nutrient element of that type is regarded as a deviating nutrient element. Calculate the proportion of deviating nutrients in the main ingredients, complementary ingredients, and basic seasonings, and mark the corresponding ingredients with the highest proportion as the role adjustment ingredients. By repeatedly adjusting the actual amount of ingredients used in the corresponding white duck soup candidate recipe, until the total amount of each type of nutrient in the white duck soup candidate recipe is within the preset reasonable component range, the adjusted white duck soup candidate recipe is obtained.

9. A white duck soup health recipe recommendation system based on nutritional component analysis according to claim 8, characterized in that, The methods for performing nutritional component matching analysis include: Extract the total amount of each type of adjusted nutrient in each candidate recipe of adjusted white duck soup, calculate the difference between it and the budget upper limit of the nutrient element of that type, and take the absolute value to obtain the analysis deviation of the nutrient element of the corresponding type. Obtain the nutritional evaluation weight of each type of nutrient element, and calculate the matching loss of that type of nutrient element by multiplying the analysis bias of the corresponding type of nutrient element by the nutritional evaluation weight. The total loss of all types of nutrients is calculated as the nutrient loss score; the formula score of the adjusted white duck soup candidate formula is obtained by subtracting the nutrient loss score from the full score of the preset formula evaluation. All adjusted white duck soup candidate recipes in the white duck soup candidate recipe set are incorporated into a preset table structure for storage, resulting in a candidate recipe scoring table.

10. A white duck soup health recipe recommendation system based on nutritional component analysis according to claim 9, characterized in that, The method for generating the recommended healthy recipe for the target white duck soup includes: Candidate white duck soup recipes with scores within the preset recommended recipe score range are selected, sorted in descending order of recipe score, and integrated into the target white duck soup healthy recipe recommendation results.

Images