Premade dish intelligent customization method and system based on big data

Abstract

The invention relates to a big data-based prefabricated dish intelligent customization method and system, and belongs to the technical field of catering services, and the method comprises the steps: obtaining all prefabricated dish varieties ordered by a current consumption customer in a preset time period, and carrying out the evaluation of all prefabricated dish varieties ordered in the preset time period, and obtaining a diet health evaluation result; introducing an exhaustion search algorithm to construct a discrete combination pairing model, and performing discrete combination on all the prefabricated dish varieties in the discrete combination pairing model based on the diet health assessment result to obtain a class of intelligent customization plans; according to the health condition report of the current consumption customer and the diet health assessment result, performing redundancy elimination on all the prefabricated dish varieties to obtain a second-class intelligent customization plan; and introducing an FCM algorithm to perform clustering calculation and screening on the first-class intelligent customization plan, the second-class intelligent customization plan and the nutrition balance manual to obtain a final prefabricated dish intelligent customization scheme. The method can intelligently customize the premade dishes, and improves the diet health quality.

Description

Technical Field

[0001] This invention relates to the field of catering service technology, and in particular to a method and system for intelligent customization of pre-made dishes based on big data. Background Technology

[0002] Pre-cooked meals are foods that are processed or prepared in advance, making them convenient to store and use. They are typically semi-finished products that only require heating or seasoning before consumption, or they can be fully cooked and eaten directly. Pre-cooked meals significantly reduce the time and workload of food preparation in the kitchen, offering convenience and speed. Currently, traditional pre-cooked meal customization methods can only provide simple customization and recommendations based on user taste preferences and meal times. This makes it difficult to tailor pre-cooked meals to the user's health condition, nutritional needs, and overall health indicators, failing to intelligently customize healthy, nutritious, and delicious pre-cooked meals from a health perspective. Long-term consumption of unhealthy pre-cooked meal combinations can lead to nutritional deficiencies, health problems, and loss of appetite, significantly increasing the risk of illness and harming user health. Therefore, a smart pre-cooked meal customization method that considers users' healthy eating habits is needed to solve these problems. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a method and system for intelligent customization of pre-made dishes based on big data.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a method for intelligent customization of pre-made dishes based on big data, comprising the following steps:

[0006] The facial features of the current customer are obtained, and the facial features are analyzed in the face recognition model to obtain all the pre-prepared dishes ordered by the current customer within a preset time period. The pre-prepared dishes ordered within the preset time period are evaluated to obtain the dietary health assessment results.

[0007] Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, then an exhaustive search algorithm is introduced to construct a discrete combination pairing model. At the same time, all the pre-made food varieties in the pre-made food store are obtained. In the discrete combination pairing model, all the pre-made food varieties are discretely combined to obtain a type of intelligent customization plan.

[0008] Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, the health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within the preset time period are redundantly eliminated to obtain the redundancy elimination results. Based on the redundancy elimination results, calculations are performed in the discrete combination pairing model to obtain the second type of intelligent customization plan.

[0009] The nutritional balance manual and taste preference information are obtained. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutritional balance manual, and obtain the nutritional balance clustering results at different time points. Based on the nutritional balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme.

[0010] Furthermore, in a preferred embodiment of the present invention, the step of acquiring the facial features of the current customer, analyzing the facial features in a facial recognition model to obtain all pre-prepared dishes ordered by the current customer within a preset time period, and evaluating all pre-prepared dishes ordered within the preset time period to obtain a dietary health assessment result specifically includes the following steps:

[0011] The system acquires historical facial image information of customers and extracts features from the facial image information to obtain several historical facial feature data. The system then uses a spectral clustering algorithm to cluster the several historical facial feature data to obtain a facial feature cluster set.

[0012] An initial recognition model is constructed based on the support vector machine algorithm. The facial feature cluster set is then imported into the initial recognition model for training and verification to obtain the face recognition model.

[0013] The facial features of the current customer are obtained, and the facial features of the current customer are calculated and recognized in the face recognition model to obtain the recognition result. Based on the recognition result, the historical ordering records of the current customer are retrieved.

[0014] Analyze the historical order records to extract all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the nutritional component index values ​​of each pre-prepared dish based on all the pre-prepared dishes; wherein, the nutritional component index values ​​include dietary fiber content, trans fatty acid content, calories and sodium content;

[0015] Based on big data networks, human dietary health benchmark information is obtained, and a dietary health assessment model is constructed based on the human dietary health benchmark information. At the same time, the age group of the current consumer is obtained, and the nutritional component index values ​​of each pre-prepared dish are evaluated in the dietary health assessment model based on the age group of the current consumer to obtain the dietary health assessment result.

[0016] Furthermore, in a preferred embodiment of the present invention, the step of analyzing the dietary health assessment results, if the dietary health assessment results are within a preset dietary health assessment result range, involves introducing an exhaustive search algorithm to construct a discrete combination pairing model, simultaneously acquiring all pre-prepared food varieties from the pre-prepared food store, and discretely combining all pre-prepared food varieties in the discrete combination pairing model to obtain a type of intelligent customization plan, specifically including the following steps:

[0017] If the dietary health assessment result is within the preset dietary health assessment result range, then the N target nutrient indicators required by the human body every day are obtained based on the big data network, and the first human health coefficient of each target nutrient indicator is obtained at the same time.

[0018] The system obtains all the ready-made food varieties from the ready-made food store, as well as information on several raw materials required for each ready-made food variety. Based on the information on these raw materials, the system constructs search tags and imports these search tags into big data for retrieval. This yields several nutritional component indicators corresponding to each raw material information, and also obtains the second human health coefficient for each nutritional component indicator.

[0019] A discrete combination pairing model is constructed by introducing deep learning networks and exhaustive search algorithms. All the pre-prepared dishes are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes.

[0020] The total human health coefficient of the M pre-prepared dish discrete combinations is calculated based on the second human health coefficient of each nutritional component index, and the actual human health coefficient of the M pre-prepared dish discrete combinations is obtained.

[0021] Determine whether the actual human health coefficient of each of the pre-prepared dish discrete combinations is greater than the first human health coefficient. If it is greater, extract the pre-prepared dish discrete combinations corresponding to whether the actual human health coefficient is greater than the first human health coefficient and package them neatly to obtain a type of intelligent customization plan.

[0022] Furthermore, in a preferred embodiment of the present invention, the step of introducing a deep learning network and an exhaustive search algorithm to construct a discrete combination pairing model, and importing all the pre-prepared dish varieties into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes, specifically includes the following steps:

[0023] A pairing model is built based on a deep learning network. All pre-prepared dishes are imported into the traversal pairing model so that each pre-prepared dish can be trained with one or more of the remaining pre-prepared dishes through traversal pairing. The trained pairing model is then obtained.

[0024] An exhaustive search algorithm is introduced to perform combined calculations on all the pre-prepared dish varieties. Based on all the pre-prepared dish varieties, variable constraints in the solution space are determined. Several preferred solutions in the solution space are generated based on the variable constraints. The nested loop method is used to traverse the several preferred solutions in the solution space and check whether each preferred solution satisfies the variable constraints.

[0025] If the preferred solution satisfies the variable constraints, the pre-prepared vegetable variety information corresponding to the preferred solution that satisfies the variable constraints is extracted and evaluated to obtain the evaluation result. The evaluation result is then embedded into the trained pairing model to obtain the discrete combination pairing model.

[0026] All the prepared dish varieties are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of prepared dishes.

[0027] Furthermore, in a preferred embodiment of the present invention, if the dietary health assessment result is not within a preset dietary health assessment result range, a health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within a preset time period are redundantly eliminated to obtain a redundancy elimination result. Based on the redundancy elimination result, a discrete combination pairing model is used to calculate a second-class intelligent customization plan, specifically including the following steps:

[0028] If the dietary health assessment result is not within the preset dietary health assessment result range, then obtain the current consumer's health status report, and extract one or more abnormal health indicators based on the health status report; wherein, the health indicators include weight, heart rate, blood sugar, blood pressure and blood lipids;

[0029] Obtain the raw material information, seasoning information, and production process information of all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the seasoning ratio during the production process of each pre-prepared dish based on the production process information;

[0030] Factor analysis is introduced to calculate the correlation between the raw material information, seasoning information, and seasoning ratio and each abnormal health indicator item, to obtain multiple factor correlation degrees. Raw material information, seasoning information, or seasoning ratio corresponding to factor correlation degrees greater than preset factor correlation degrees are extracted and integrated to obtain an abnormal redundancy removal set.

[0031] Obtain the raw material information, seasoning information, and seasoning ratio of all pre-prepared dishes in the pre-prepared food store. Import the raw material information, seasoning information, and seasoning ratio of all pre-prepared dishes in the pre-prepared food store into the abnormal redundancy removal set for matching and redundancy removal to obtain the redundancy removal result.

[0032] The remaining pre-prepared dish varieties from the redundancy elimination results are imported into the discrete combination pairing model for calculation to obtain several customized pre-prepared dish combinations. These customized pre-prepared dish combinations are then packaged and organized to obtain a second-class intelligent customization plan.

[0033] Furthermore, in a preferred embodiment of the present invention, the step of obtaining the nutrition balance manual and taste preference information, introducing the FCM algorithm to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points, and filtering the first type of intelligent customization plan and the second type of intelligent customization plan based on the nutrition balance clustering results and taste preference information to obtain the final pre-made intelligent customization scheme, specifically includes the following steps:

[0034] Based on a big data network, a nutritional balance manual is obtained, and the manual is analyzed to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening.

[0035] The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. K clusters are generated based on the standard nutritionally balanced dishes at different time points, and the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan is calculated.

[0036] A preset membership threshold is used to update the cluster center position of each cluster based on the membership degree. During the update iteration process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iteration is stopped immediately, and all cluster centers after the update iteration is stopped are output to obtain the nutrient-balanced clustering results at different time points.

[0037] Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class and second-class intelligent customization plans are screened to obtain the screened first-class and second-class intelligent customization plans.

[0038] The system acquires and analyzes all pre-made dishes ordered by the current customer within a preset time period to obtain the customer's taste preference information. Based on the taste preference information, the system performs a second screening and integration of the first and second types of smart customization plans to obtain the final smart customization solution for pre-made dishes.

[0039] A second aspect of the present invention provides a big data-based intelligent pre-prepared meal customization system. The system includes a memory and a processor. The memory stores a big data-based intelligent pre-prepared meal customization method program. When the processor executes the big data-based intelligent pre-prepared meal customization method program, it performs the following steps:

[0040] The facial features of the current customer are obtained, and the facial features are analyzed in the face recognition model to obtain all the pre-prepared dishes ordered by the current customer within a preset time period. The pre-prepared dishes ordered within the preset time period are evaluated to obtain the dietary health assessment results.

[0041] Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, then an exhaustive search algorithm is introduced to construct a discrete combination pairing model. At the same time, all the pre-made food varieties in the pre-made food store are obtained. In the discrete combination pairing model, all the pre-made food varieties are discretely combined to obtain a type of intelligent customization plan.

[0042] Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, the health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within the preset time period are redundantly eliminated to obtain the redundancy elimination results. Based on the redundancy elimination results, calculations are performed in the discrete combination pairing model to obtain the second type of intelligent customization plan.

[0043] The nutritional balance manual and taste preference information are obtained. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutritional balance manual, and obtain the nutritional balance clustering results at different time points. Based on the nutritional balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme.

[0044] Furthermore, in a preferred embodiment of the present invention, the step of obtaining the nutrition balance manual and taste preference information, introducing the FCM algorithm to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points, and filtering the first type of intelligent customization plan and the second type of intelligent customization plan based on the nutrition balance clustering results and taste preference information to obtain the final pre-made intelligent customization scheme, specifically includes the following steps:

[0045] Based on a big data network, a nutritional balance manual is obtained, and the manual is analyzed to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening.

[0046] The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. K clusters are generated based on the standard nutritionally balanced dishes at different time points, and the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan is calculated.

[0047] A preset membership threshold is used to update the cluster center position of each cluster based on the membership degree. During the update iteration process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iteration is stopped immediately, and all cluster centers after the update iteration is stopped are output to obtain the nutrient-balanced clustering results at different time points.

[0048] Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class and second-class intelligent customization plans are screened to obtain the screened first-class and second-class intelligent customization plans.

[0049] The system acquires and analyzes all pre-made dishes ordered by the current customer within a preset time period to obtain the customer's taste preference information. Based on the taste preference information, the system performs a second screening and integration of the first and second types of smart customization plans to obtain the final smart customization solution for pre-made dishes.

[0050] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0051] The system retrieves all pre-prepared food items ordered by current customers within a preset time period and evaluates these items to obtain a dietary health assessment result. Analyzing this result, if it falls within a preset dietary health assessment result range, an exhaustive search algorithm is introduced to construct a discrete combination pairing model. Simultaneously, all pre-prepared food items from the pre-prepared food store are retrieved, and these items are discretely combined within the model to obtain a smart customization plan. Based on the dietary health assessment result, if the result does not fall within a preset range, an exhaustive search algorithm is introduced to construct a discrete combination pairing model. Simultaneously, all pre-prepared food items from the store are retrieved, and discrete combinations are performed on these items within the model to obtain a smart customization plan. Within a given dietary health assessment range, the current customer's health status report is obtained. Based on this report, redundancy is eliminated from all pre-made meal orders placed by the current customer within a preset time period, resulting in a second type of intelligent customization plan. A nutrition balance manual is then obtained, and the FCM algorithm is used to cluster the first type of intelligent customization plan, the second type of intelligent customization plan, and the nutrition balance manual, yielding nutrition balance clustering results at different time points. Based on these clustering results, the first and second types of intelligent customization plans are filtered to obtain the final intelligent pre-made meal customization solution. This invention intelligently provides customers with healthy and nutritious pre-made meal customization solutions, improving their dietary health and ensuring nutrient absorption. Attached Figure Description

[0052] 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 embodiments can be obtained from these drawings without creative effort.

[0053] Figure 1 A flowchart of the first method for intelligent customization of pre-prepared dishes based on big data is shown;

[0054] Figure 2 A flowchart of a second method for intelligent customization of pre-prepared dishes based on big data is shown.

[0055] Figure 3 A flowchart of a third method for intelligent customization of pre-prepared dishes based on big data is shown.

[0056] Figure 4 A system framework diagram of a pre-prepared food intelligent customization system based on big data is shown. Detailed Implementation

[0057] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0058] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0059] The first aspect of this invention provides a method for intelligent customization of pre-made dishes based on big data, such as... Figure 1 As shown, it includes the following steps:

[0060] S102: Obtain the facial features of the current customer, analyze the facial features in the face recognition model, obtain all the pre-prepared dishes ordered by the current customer within a preset time period, and evaluate all the pre-prepared dishes ordered within the preset time period to obtain a dietary health assessment result.

[0061] S104: Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, then introduce an exhaustive search algorithm to construct a discrete combination pairing model. At the same time, obtain all the pre-made food varieties of the pre-made food store. Discretely combine all the pre-made food varieties in the discrete combination pairing model to obtain a type of intelligent customization plan.

[0062] S106: Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, then obtain the current consumer's health status report, and according to the health status report, perform redundancy elimination on all pre-made dishes ordered by the current consumer within the preset time period to obtain redundancy elimination results. Based on the redundancy elimination results, perform calculations in the discrete combination pairing model to obtain a second type of intelligent customization plan.

[0063] S108: Obtain the nutrition balance manual and taste preference information, introduce the FCM algorithm to perform clustering calculation on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutrition balance manual, obtain the nutrition balance clustering results at different time points, and filter the first type of intelligent customization plan and the second type of intelligent customization plan based on the nutrition balance clustering results and taste preference information to obtain the final pre-made intelligent customization scheme.

[0064] Furthermore, in a preferred embodiment of the present invention, the step of acquiring the facial features of the current customer, analyzing the facial features in a facial recognition model to obtain all pre-prepared dishes ordered by the current customer within a preset time period, and evaluating all pre-prepared dishes ordered within the preset time period to obtain a dietary health assessment result specifically includes the following steps:

[0065] The system acquires historical facial image information of customers and extracts features from the facial image information to obtain several historical facial feature data. The system then uses a spectral clustering algorithm to cluster the several historical facial feature data to obtain a facial feature cluster set.

[0066] An initial recognition model is constructed based on the support vector machine algorithm. The facial feature cluster set is then imported into the initial recognition model for training and verification to obtain the face recognition model.

[0067] The facial features of the current customer are obtained, and the facial features of the current customer are calculated and recognized in the face recognition model to obtain the recognition result. Based on the recognition result, the historical ordering records of the current customer are retrieved.

[0068] Analyze the historical order records to extract all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the nutritional component index values ​​of each pre-prepared dish based on all the pre-prepared dishes; wherein, the nutritional component index values ​​include dietary fiber content, trans fatty acid content, calories and sodium content;

[0069] Based on big data networks, human dietary health benchmark information is obtained, and a dietary health assessment model is constructed based on the human dietary health benchmark information. At the same time, the age group of the current consumer is obtained, and the nutritional component index values ​​of each pre-prepared dish are evaluated in the dietary health assessment model based on the age group of the current consumer to obtain the dietary health assessment result.

[0070] It's important to note that the nutritional content of pre-cooked meals varies due to differences in ingredient combinations. Furthermore, the nutritional needs of consumers at different age levels require careful consideration and health management. For example, older consumers require pre-cooked meals that are low in calories, high in dietary fiber, and low in cholesterol to ensure a healthy diet. However, traditional pre-cooked meal customization methods cannot assess the health benefits of previously purchased pre-cooked meals based on age, leading to unreliable recommendations and potentially increased risk of illness. Therefore, a health assessment of previously purchased pre-cooked meals is necessary before intelligent selection and customization. This approach can significantly improve the level of healthy eating and ensure the health of consumers at all ages. This invention aims to improve the nutritional absorption of pre-prepared meals at different stages and reduce the risk of disease. Currently, traditional pre-prepared meal customization methods typically rely on directly reading customers' past order records, making it difficult to accurately determine the customer's age group. Furthermore, this direct method significantly reduces information security and the accuracy of the assessment results, hindering precise and efficient pre-prepared meal customization. Therefore, this invention addresses this issue by acquiring customers' historical facial image information and extracting features to obtain several historical facial feature data. A face recognition model is then constructed based on this data using a support vector machine algorithm. This model identifies the current customer's facial features to retrieve their historical order records, achieving accurate information acquisition, improving security, and accurately determining age information. This invention enables dietary health assessment of pre-prepared meals based on age group, allowing for customized pre-prepared meals for customers of different age groups, thereby improving dietary health and the level of adequate nutrient absorption.

[0071] Furthermore, in a preferred embodiment of the present invention, the step of analyzing the dietary health assessment results, if the dietary health assessment results are within a preset dietary health assessment result range, involves introducing an exhaustive search algorithm to construct a discrete combination pairing model, simultaneously acquiring all pre-prepared food varieties from the pre-prepared food store, and discretely combining all pre-prepared food varieties in the discrete combination pairing model to obtain a type of intelligent customization plan, specifically including the following steps:

[0072] If the dietary health assessment result is within the preset dietary health assessment result range, then the N target nutrient indicators required by the human body every day are obtained based on the big data network, and the first human health coefficient of each target nutrient indicator is obtained at the same time.

[0073] The system obtains all the ready-made food varieties from the ready-made food store, as well as information on several raw materials required for each ready-made food variety. Based on the information on these raw materials, the system constructs search tags and imports these search tags into big data for retrieval. This yields several nutritional component indicators corresponding to each raw material information, and also obtains the second human health coefficient for each nutritional component indicator.

[0074] A discrete combination pairing model is constructed by introducing deep learning networks and exhaustive search algorithms. All the pre-prepared dishes are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes.

[0075] The total human health coefficient of the M pre-prepared dish discrete combinations is calculated based on the second human health coefficient of each nutritional component index, and the actual human health coefficient of the M pre-prepared dish discrete combinations is obtained.

[0076] Determine whether the actual human health coefficient of each of the pre-prepared dish discrete combinations is greater than the first human health coefficient. If it is greater, extract the pre-prepared dish discrete combinations corresponding to whether the actual human health coefficient is greater than the first human health coefficient and package them neatly to obtain a type of intelligent customization plan.

[0077] It should be noted that the dietary health assessment results directly reflect the healthiness of pre-prepared meal options for consumers of different age groups. If the dietary health assessment results fall within the preset range, it indicates that most pre-prepared meal options in the store are healthy and highly nutritious for the target age group, and therefore can be selected from all pre-prepared meal options. However, the nutritional content of the raw materials in different pre-prepared meal options varies, and selection should be based on the daily nutritional requirements of the human body. The first human health coefficient represents the health benefits of the target nutritional requirements for the human body each day. The first index represents the health index of various nutritional components in the raw materials of pre-cooked dishes for the human body. Therefore, by calculating whether the second human health index of all pre-cooked dishes meets the requirements of the first human health index, compliant pre-cooked dishes can be packaged and recommended. To achieve better customization, a discrete combination pairing model is constructed by introducing deep learning networks and exhaustive search algorithms. This model can intelligently calculate the combination of all pre-cooked dishes one by one, achieving diversified pre-cooked dish combination customization based on human health index judgment. This provides consumers with rich, healthy, and nutritious pre-cooked dish customization options, replacing the traditional, less efficient pre-cooked dish customization methods. This invention can provide consumers with diversified, healthy, and nutritious intelligent combination customization of pre-cooked dishes when the dietary health assessment results meet the standards, improving the nutritional value of the combined pre-cooked dishes and increasing the intake of nutrients required by the human body.

[0078] Furthermore, in a preferred embodiment of the present invention, the step of introducing a deep learning network and an exhaustive search algorithm to construct a discrete combination pairing model, and importing all the pre-prepared dish varieties into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes, specifically includes the following steps:

[0079] A pairing model is built based on a deep learning network. All pre-prepared dishes are imported into the traversal pairing model so that each pre-prepared dish can be trained with one or more of the remaining pre-prepared dishes through traversal pairing. The trained pairing model is then obtained.

[0080] An exhaustive search algorithm is introduced to perform combined calculations on all the pre-prepared dish varieties. Based on all the pre-prepared dish varieties, variable constraints in the solution space are determined. Several preferred solutions in the solution space are generated based on the variable constraints. The nested loop method is used to traverse the several preferred solutions in the solution space and check whether each preferred solution satisfies the variable constraints.

[0081] If the preferred solution satisfies the variable constraints, the pre-prepared vegetable variety information corresponding to the preferred solution that satisfies the variable constraints is extracted and evaluated to obtain the evaluation result. The evaluation result is then embedded into the trained pairing model to obtain the discrete combination pairing model.

[0082] All the prepared dish varieties are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of prepared dishes.

[0083] It should be noted that the construction of the discrete combination pairing model first requires training a model capable of performing traversal combination pairings one by one. This traversal pairing model is constructed using a deep learning network. This model serves as the initial training model, providing a training framework for data training. Information on several raw materials is imported into the traversal pairing model, allowing each raw material to be paired with one or more of the remaining raw materials for training. This results in a fully trained traversal pairing model, which forms the basic framework for intelligent raw material matching, improving the computational performance of the exhaustive search algorithm and significantly increasing the speed of pre-prepared food customization. After the traversal pairing model is trained, an exhaustive search algorithm is introduced. This algorithm can calculate the quality of combinations and even find the optimal solution. The exhaustive search algorithm is used to further calculate and evaluate the optimal solution for all pre-prepared food combinations within the trained traversal pairing model, ultimately yielding multiple discrete combinations of pre-prepared foods. This enables precise and rapid diversification of pre-prepared food combinations, achieving the goal of customizing combinations based on customer needs and selected pre-prepared food varieties.

[0084] Furthermore, in a preferred embodiment of the present invention, based on the dietary health assessment results, if the dietary health assessment results are not within a preset dietary health assessment result range, then a health status report of the current consumer is obtained. Based on the health status report, all pre-prepared dishes ordered by the current consumer within a preset time period are redundantly eliminated to obtain redundancy elimination results. Based on the redundancy elimination results, a discrete combination pairing model is used to calculate a second-class intelligent customization plan, such as... Figure 2 As shown, the specific steps include:

[0085] S202: If the dietary health assessment result is not within the preset dietary health assessment result range, then obtain the current consumer's health status report, and extract one or more abnormal health indicators based on the health status report; wherein, the health indicators include weight, heart rate, blood sugar, blood pressure and blood lipids;

[0086] S204: Obtain the raw material information, seasoning information, and production process information of all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the seasoning ratio in the production process of each pre-prepared dish based on the production process information;

[0087] S206: Introduce factor analysis to calculate the correlation between the raw material information, seasoning information, and seasoning ratio and each abnormal health indicator item, obtain multiple factor correlation degrees, extract the raw material information, seasoning information, or seasoning ratio corresponding to the factor correlation degree greater than the preset factor correlation degree, and integrate them to obtain an abnormal redundancy removal set.

[0088] S208: Obtain the raw material information, seasoning information and seasoning ratio of all pre-prepared dishes in the pre-prepared food store, import the raw material information, seasoning information and seasoning ratio of all pre-prepared dishes in the pre-prepared food store into the abnormal redundancy removal set for matching and redundancy removal, and obtain the redundancy removal result.

[0089] S210: Import the remaining pre-prepared dish varieties from the redundancy elimination results into the discrete combination pairing model for calculation to obtain several customized pre-prepared dish combinations. Then, package and organize these customized pre-prepared dish combinations to obtain a second-class intelligent customization plan.

[0090] It should be noted that the production process information includes steaming, frying, heat level, seasoning ratio, etc. If the dietary health assessment result is not within the preset dietary health assessment result range, it means that the pre-prepared food in the pre-prepared food store is unhealthy and may harm the health of the consumer's age group, at least in terms of the pre-prepared food varieties purchased by the consumer. Long-term consumption has led to abnormal health conditions in the consumer. Therefore, it is necessary to analyze the purchased pre-prepared food varieties based on the abnormal health indicators shown in the consumer's health status report, analyze the raw material information, seasoning information, and seasoning ratio that caused the consumer's health abnormalities, and then screen all pre-prepared food varieties in the pre-prepared food store based on the analyzed raw material information, seasoning information, and seasoning ratio. Pre-prepared food varieties containing the same raw material information, seasoning information, and seasoning ratio are redundantly removed. The remaining varieties after the redundancy removal are then imported into a discrete combination pairing model for calculation and packaging, and output a second type of intelligent customization plan. This invention can select and match pre-made dishes according to the health status of consumers, thereby improving the nutritional value and health of customized pre-made dishes, avoiding the adverse effects of traditional pre-made dish customization on the health of consumers, and ensuring high safety and reliability.

[0091] Furthermore, in a preferred embodiment of the present invention, the acquisition of the nutrition balance manual and taste preference information involves introducing an FCM algorithm to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points. Based on the nutrition balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme, such as... Figure 3 As shown, the specific steps include:

[0092] S302: Obtain a nutritional balance manual based on a big data network, and analyze the nutritional balance manual to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening;

[0093] S304: Introduce the FCM algorithm to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. Generate K clusters based on the standard nutritionally balanced dishes at different time points, and calculate the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan.

[0094] S306: A membership threshold is preset. The cluster center position of each cluster is updated iteratively based on the membership degree. During the update iterative process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iterative is stopped immediately, and all cluster centers after the update iterative is stopped are output to obtain the nutrient-balanced clustering results at different time points.

[0095] S308: Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class intelligent customization plan and the second-class intelligent customization plan are screened to obtain the screened first-class intelligent customization plan and the screened second-class intelligent customization plan.

[0096] S310: Obtain all pre-made dishes ordered by the current customer within a preset time period, analyze the current customer's taste preference information, and perform secondary screening and integration on the first-class and second-class intelligent customization plans based on the taste preference information to obtain the final pre-made dish intelligent customization solution.

[0097] It's important to note that a person's dietary health is usually inseparable from regular mealtimes. For example, breakfast needs to ensure a certain intake of protein and vitamins to guarantee the body's energy supply and nutritional health throughout the day. Therefore, it's necessary to rationally allocate the nutritional balance of the pre-prepared dishes in both the Type I and Type II smart customization plans according to meal times to ensure a balanced intake of nutrients throughout the day. The nutritional balance manual records standard nutritionally balanced dishes for different time points. Then, the FCM algorithm is used to calculate the membership degree of each pre-prepared dish in the Type I and Type II smart customization plans with the standard nutritionally balanced dishes at different time points. This allows for the customization of pre-prepared dishes that meet the corresponding nutritional balance at different time points for consumers based on the Type I and Type II smart customization plans. This ensures that consumers obtain more nutrients under the standard nutritional balance benchmark, improving their dietary health and immune function, reducing the chance of disease, and enhancing their health and safety. At the same time, pre-prepared dishes can be further screened and customized based on consumers' taste preferences to increase their enjoyment and desire to eat them.

[0098] Furthermore, the aforementioned method for intelligent customization of pre-prepared dishes based on big data also includes the following steps:

[0099] The historical calorie variation values ​​of consumers under different customized combinations of pre-prepared dishes are obtained. A prediction model is constructed based on the convolutional neural network algorithm. The historical calorie variation values ​​under different customized combinations of pre-prepared dishes are imported into the prediction model for training to obtain the trained prediction model.

[0100] Obtain the current customized combination of pre-cooked dishes, import the current customized combination of pre-cooked dishes into the trained prediction model, and obtain the predicted value of calorie change;

[0101] Obtain the actual calorie change value of the consumer. If the actual calorie change value is greater than the predicted calorie change value, obtain the relevant weight loss plan in the big data network based on the actual calorie change value, and at the same time obtain the execution difficulty of each weight loss plan.

[0102] The Euclidean distance algorithm is introduced to calculate the Euclidean distance between the execution difficulty of each weight loss plan and the age group of the current consumer. Based on the Euclidean distance, the matching degree between the execution difficulty of each weight loss plan and the age group of the current consumer is determined, and several matching degrees are obtained. The weight loss plan corresponding to the matching degree greater than the preset matching degree is extracted to obtain the weight loss recommendation plan.

[0103] Obtain the success rate of each fat loss plan in the recommended fat loss program, construct a sorting table, import the success rates of each fat loss plan in the recommended fat loss program into the sorting table and sort them from largest to smallest to generate a fat loss success rate sorting table, and extract the fat loss plan corresponding to the highest fat loss success rate from the fat loss success rate sorting table and output it.

[0104] It should be noted that pre-cooked meals contain a certain amount of calories. Long-term consumption of pre-cooked meals can lead to increased body fat, obesity, mobility issues, and various obesity-related diseases, negatively impacting the physical and mental health of consumers. Therefore, consumers should be encouraged to engage in post-meal fat-burning exercises to burn body fat and reduce the likelihood of obesity-related diseases. Various fat-burning programs exist, each with varying levels of difficulty. High-difficulty programs are clearly unsuitable for older users, as they can easily cause injury. Therefore, the Euclidean distance algorithm is used to calculate the compatibility between the difficulty of each fat-burning program and the current age group of the consumer. Based on this compatibility, a suitable fat-burning program is selected for the current consumer, improving the safety of fat loss. This invention can intelligently recommend fat-burning programs to consumers after consuming pre-cooked meals, reducing fat accumulation from high-calorie diets and obesity-related diseases, improving the physical and mental health of consumers and ensuring the safety of fat loss. It is highly reliable.

[0105] A second aspect of the present invention provides a big data-based intelligent pre-prepared meal customization system. This system includes a memory 41 and a processor 42. The memory 41 stores a big data-based intelligent pre-prepared meal customization method program. When the processor 42 executes the big data-based intelligent pre-prepared meal customization method program, such as... Figure 4As shown, the following steps are performed:

[0106] The facial features of the current customer are obtained, and the facial features are analyzed in the face recognition model to obtain all the pre-prepared dishes ordered by the current customer within a preset time period. The pre-prepared dishes ordered within the preset time period are evaluated to obtain the dietary health assessment results.

[0107] Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, obtain information on several kinds of raw materials required to make pre-prepared dishes. Discretely combine the information on several kinds of raw materials in a discrete combination pairing model to obtain a type of intelligent customization plan.

[0108] Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, the health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within the preset time period are redundantly eliminated to obtain the redundancy elimination results. Based on the redundancy elimination results, calculations are performed in the discrete combination pairing model to obtain the second type of intelligent customization plan.

[0109] The nutritional balance manual and taste preference information are obtained. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutritional balance manual, and obtain the nutritional balance clustering results at different time points. Based on the nutritional balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme.

[0110] Furthermore, in a preferred embodiment of the present invention, the step of obtaining the nutrition balance manual and taste preference information, introducing the FCM algorithm to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points, and filtering the first type of intelligent customization plan and the second type of intelligent customization plan based on the nutrition balance clustering results and taste preference information to obtain the final pre-made intelligent customization scheme, specifically includes the following steps:

[0111] Based on a big data network, a nutritional balance manual is obtained, and the manual is analyzed to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening.

[0112] The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. K clusters are generated based on the standard nutritionally balanced dishes at different time points, and the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan is calculated.

[0113] A preset membership threshold is used to update the cluster center position of each cluster based on the membership degree. During the update iteration process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iteration is stopped immediately, and all cluster centers after the update iteration is stopped are output to obtain the nutrient-balanced clustering results at different time points.

[0114] Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class and second-class intelligent customization plans are screened to obtain the screened first-class and second-class intelligent customization plans.

[0115] The system acquires and analyzes all pre-made dishes ordered by the current customer within a preset time period to obtain the customer's taste preference information. Based on the taste preference information, the system performs a second screening and integration of the first and second types of smart customization plans to obtain the final smart customization solution for pre-made dishes.

[0116] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for intelligent customization of pre-prepared dishes based on big data, characterized in that, Includes the following steps: The facial features of the current customer are obtained, and the facial features are analyzed in the face recognition model to obtain all the pre-prepared dishes ordered by the current customer within a preset time period. The pre-prepared dishes ordered within the preset time period are evaluated to obtain the dietary health assessment results. Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, then an exhaustive search algorithm is introduced to construct a discrete combination pairing model. At the same time, all the pre-made food varieties in the pre-made food store are obtained. In the discrete combination pairing model, all the pre-made food varieties are discretely combined to obtain a type of intelligent customization plan. Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, the health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within the preset time period are redundantly eliminated to obtain the redundancy elimination results. Based on the redundancy elimination results, calculations are performed in the discrete combination pairing model to obtain the second type of intelligent customization plan. The nutritional balance manual and taste preference information are obtained. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutritional balance manual, and obtain the nutritional balance clustering results at different time points. Based on the nutritional balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme.

2. The method for intelligent customization of pre-prepared dishes based on big data according to claim 1, characterized in that, The process of acquiring the facial features of the current customer, analyzing these features in a facial recognition model to obtain all pre-prepared food items ordered by the current customer within a preset time period, and evaluating all pre-prepared food items ordered within the preset time period to obtain a dietary health assessment result includes the following steps: The system acquires historical facial image information of customers and extracts features from the facial image information to obtain several historical facial feature data. The system then uses a spectral clustering algorithm to cluster the several historical facial feature data to obtain a facial feature cluster set. An initial recognition model is constructed based on the support vector machine algorithm. The facial feature cluster set is then imported into the initial recognition model for training and verification to obtain the face recognition model. The facial features of the current customer are obtained, and the facial features of the current customer are calculated and recognized in the face recognition model to obtain the recognition result. Based on the recognition result, the historical ordering records of the current customer are retrieved. Analyze the historical order records to extract all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the nutritional component index values ​​of each pre-prepared dish based on all the pre-prepared dishes; wherein, the nutritional component index values ​​include dietary fiber content, trans fatty acid content, calories and sodium content; Based on big data networks, human dietary health benchmark information is obtained, and a dietary health assessment model is constructed based on the human dietary health benchmark information. At the same time, the age group of the current consumer is obtained, and the nutritional component index values ​​of each pre-prepared dish are evaluated in the dietary health assessment model based on the age group of the current consumer to obtain the dietary health assessment result.

3. The method for intelligent customization of pre-prepared dishes based on big data according to claim 1, characterized in that, The analysis of the dietary health assessment results, if the results fall within a preset dietary health assessment result range, then an exhaustive search algorithm is introduced to construct a discrete combination pairing model. Simultaneously, all pre-prepared food varieties from the pre-prepared food store are obtained. In the discrete combination pairing model, all pre-prepared food varieties are discretely combined to obtain a type of intelligent customization plan, specifically including the following steps: If the dietary health assessment result is within the preset dietary health assessment result range, then the N target nutrient indicators required by the human body every day are obtained based on the big data network, and the first human health coefficient of each target nutrient indicator is obtained at the same time. The system obtains all the ready-made food varieties from the ready-made food store, as well as information on several raw materials required for each ready-made food variety. Based on the information on these raw materials, the system constructs search tags and imports these search tags into big data for retrieval. This yields several nutritional component indicators corresponding to each raw material information, and also obtains the second human health coefficient for each nutritional component indicator. A discrete combination pairing model is constructed by introducing deep learning networks and exhaustive search algorithms. All the pre-prepared dishes are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes. The total human health coefficient of the M pre-prepared dish discrete combinations is calculated based on the second human health coefficient of each nutritional component index, and the actual human health coefficient of the M pre-prepared dish discrete combinations is obtained. Determine whether the actual human health coefficient of each of the pre-prepared dish discrete combinations is greater than the first human health coefficient. If it is greater, extract the pre-prepared dish discrete combinations corresponding to whether the actual human health coefficient is greater than the first human health coefficient and package them neatly to obtain a type of intelligent customization plan.

4. The method for intelligent customization of pre-prepared dishes based on big data according to claim 3, characterized in that, The process of introducing a deep learning network and an exhaustive search algorithm to construct a discrete combination pairing model, and then importing all the pre-prepared dish varieties into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of pre-prepared dishes, specifically includes the following steps: A pairing model is built based on a deep learning network. All pre-prepared dishes are imported into the traversal pairing model so that each pre-prepared dish can be trained with one or more of the remaining pre-prepared dishes through traversal pairing. The trained pairing model is then obtained. An exhaustive search algorithm is introduced to perform combined calculations on all the pre-prepared dish varieties. Based on all the pre-prepared dish varieties, variable constraints in the solution space are determined. Several preferred solutions in the solution space are generated based on the variable constraints. The nested loop method is used to traverse the several preferred solutions in the solution space and check whether each preferred solution satisfies the variable constraints. If the preferred solution satisfies the variable constraints, the pre-prepared vegetable variety information corresponding to the preferred solution that satisfies the variable constraints is extracted and evaluated to obtain the evaluation result. The evaluation result is then embedded into the trained pairing model to obtain the discrete combination pairing model. All the prepared dish varieties are imported into the discrete combination pairing model for traversal combination calculation to obtain M discrete combinations of prepared dishes.

5. The method for intelligent customization of pre-prepared dishes based on big data according to claim 1, characterized in that, Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, a health status report of the current consumer is obtained. Redundancy is eliminated from all pre-made dishes ordered by the current consumer within a preset time period based on the health status report, resulting in a redundancy elimination result. Based on the redundancy elimination result, a discrete combination pairing model is used to calculate a second-class intelligent customization plan, specifically including the following steps: If the dietary health assessment result is not within the preset dietary health assessment result range, then obtain the current consumer's health status report, and extract one or more abnormal health indicators based on the health status report; wherein, the health indicators include weight, heart rate, blood sugar, blood pressure and blood lipids; Obtain the raw material information, seasoning information, and production process information of all pre-prepared dishes ordered by the current customer within a preset time period, and obtain the seasoning ratio during the production process of each pre-prepared dish based on the production process information; Factor analysis is introduced to calculate the correlation between the raw material information, seasoning information, and seasoning ratio and each abnormal health indicator item, to obtain multiple factor correlation degrees. Raw material information, seasoning information, or seasoning ratio corresponding to factor correlation degrees greater than preset factor correlation degrees are extracted and integrated to obtain an abnormal redundancy removal set. Obtain the raw material information, seasoning information, and seasoning ratio of all pre-prepared dishes in the pre-prepared food store. Import the raw material information, seasoning information, and seasoning ratio of all pre-prepared dishes in the pre-prepared food store into the abnormal redundancy removal set for matching and redundancy removal to obtain the redundancy removal result. The remaining pre-prepared dish varieties from the redundancy elimination results are imported into the discrete combination pairing model for calculation to obtain several customized pre-prepared dish combinations. These customized pre-prepared dish combinations are then packaged and organized to obtain a second-class intelligent customization plan.

6. The method for intelligent customization of pre-prepared dishes based on big data according to claim 1, characterized in that, The process of obtaining a nutrition balance manual and taste preference information, and then using the FCM algorithm to cluster the first-type smart customization plan, the second-type smart customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points, and then filtering the first-type and second-type smart customization plans based on the nutrition balance clustering results and taste preference information to obtain the final pre-made meal smart customization solution, specifically includes the following steps: Based on a big data network, a nutritional balance manual is obtained, and the manual is analyzed to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. K clusters are generated based on the standard nutritionally balanced dishes at different time points, and the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan is calculated. A preset membership threshold is used to update the cluster center position of each cluster based on the membership degree. During the update iteration process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iteration is stopped immediately, and all cluster centers after the update iteration is stopped are output to obtain the nutrient-balanced clustering results at different time points. Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class and second-class intelligent customization plans are screened to obtain the screened first-class and second-class intelligent customization plans. The system acquires and analyzes all pre-made dishes ordered by the current customer within a preset time period to obtain the customer's taste preference information. Based on the taste preference information, the system performs a second screening and integration of the first and second types of smart customization plans to obtain the final smart customization solution for pre-made dishes.

7. A pre-prepared meal intelligent customization system based on big data, characterized in that, The big data-based intelligent pre-prepared meal customization system includes a memory and a processor. The memory stores a big data-based intelligent pre-prepared meal customization method program. When the processor executes the big data-based intelligent pre-prepared meal customization method program, it performs the following steps: The facial features of the current customer are obtained, and the facial features are analyzed in the face recognition model to obtain all the pre-prepared dishes ordered by the current customer within a preset time period. The pre-prepared dishes ordered within the preset time period are evaluated to obtain the dietary health assessment results. Analyze the dietary health assessment results. If the dietary health assessment results are within the preset dietary health assessment result range, then an exhaustive search algorithm is introduced to construct a discrete combination pairing model. At the same time, all the pre-made food varieties in the pre-made food store are obtained. In the discrete combination pairing model, all the pre-made food varieties are discretely combined to obtain a type of intelligent customization plan. Based on the dietary health assessment results, if the dietary health assessment results are not within the preset dietary health assessment result range, the health status report of the current consumer is obtained. Based on the health status report, all pre-made dishes ordered by the current consumer within the preset time period are redundantly eliminated to obtain the redundancy elimination results. Based on the redundancy elimination results, calculations are performed in the discrete combination pairing model to obtain the second type of intelligent customization plan. The nutritional balance manual and taste preference information are obtained. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan and the nutritional balance manual, and obtain the nutritional balance clustering results at different time points. Based on the nutritional balance clustering results and taste preference information, the first type of intelligent customization plan and the second type of intelligent customization plan are screened to obtain the final pre-made intelligent customization scheme.

8. The intelligent customized pre-made meal system based on big data according to claim 7, characterized in that, The process of obtaining a nutrition balance manual and taste preference information, and then using the FCM algorithm to cluster the first-type smart customization plan, the second-type smart customization plan, and the nutrition balance manual to obtain nutrition balance clustering results at different time points, and then filtering the first-type and second-type smart customization plans based on the nutrition balance clustering results and taste preference information to obtain the final pre-made meal smart customization solution, specifically includes the following steps: Based on a big data network, a nutritional balance manual is obtained, and the manual is analyzed to extract standard nutritional balance dishes for different time periods; wherein, the different time periods include morning, noon and evening. The FCM algorithm is introduced to perform clustering calculations on the first type of intelligent customization plan, the second type of intelligent customization plan, and standard nutritionally balanced dishes at different time points. K clusters are generated based on the standard nutritionally balanced dishes at different time points, and the membership degree between each cluster for the first type of intelligent customization plan and the second type of intelligent customization plan is calculated. A preset membership threshold is used to update the cluster center position of each cluster based on the membership degree. During the update iteration process, it is determined whether the membership degree is less than the membership threshold. If it is less, the update iteration is stopped immediately, and all cluster centers after the update iteration is stopped are output to obtain the nutrient-balanced clustering results at different time points. Based on the nutritional balance clustering results at different time points, the pre-prepared dish combinations in the first-class and second-class intelligent customization plans are screened to obtain the screened first-class and second-class intelligent customization plans. The system acquires and analyzes all pre-made dishes ordered by the current customer within a preset time period to obtain the customer's taste preference information. Based on the taste preference information, the system performs a second screening and integration of the first and second types of smart customization plans to obtain the final smart customization solution for pre-made dishes.

Images