Diet recommendation method, device, equipment and computer readable storage medium

Abstract

The application provides a meal recommendation method, device and equipment and a computer readable storage medium. The method comprises the following steps: obtaining current user information of a target user, wherein the current user information comprises current dietary information, current target protein information in the body and current metabolic dynamic information; determining first recommended meal information according to the current target protein information in the body based on a meal recommendation prediction network of a target meal recommendation model, and determining second recommended meal information according to the current dietary information and the current metabolic dynamic information; determining target meal recommendation information according to the first recommended meal information and the second recommended meal information based on a meal fusion network of the target meal recommendation model; and pushing the target meal recommendation information to a target terminal of the target user. The application can realize nutrition meal recommendation in the medical and health field, and improve the recommendation accuracy of the nutrition meal.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more particularly to a dietary recommendation method, apparatus, device, and computer-readable storage medium. Background Technology

[0002] The popularity of intelligent food recommendation systems is growing. These systems offer recommendations based on user dietary habits and user health information, such as those based on user dietary needs. However, many of these systems rely on single data points like BMI or nutritional standards (food pyramids) for recommendations. This makes it difficult to tailor recommendations to the user's actual health condition. This can lead to poor absorption of recommended foods, resulting in lower-than-expected benefits. Furthermore, some systems focus excessively on short-term indicators like blood sugar and lipids, neglecting other internal health indicators, thus compromising the effectiveness of the recommendations. Summary of the Invention

[0003] This application provides a dietary recommendation method, apparatus, device, and computer-readable storage medium, aiming to improve the accuracy of nutritional dietary recommendations and thus enhance the efficacy of the nutritional diets provided to users.

[0004] In a first aspect, this application provides a dietary recommendation method, which includes the following steps: Obtain the target user's current user information, which includes current dietary information, current target protein information in the body, and current metabolic dynamic information; The dietary recommendation prediction network based on the target dietary recommendation model determines the first recommended dietary information based on the current target protein information in the body, and determines the second recommended dietary information based on the current dietary information and the current metabolic dynamic information. Based on the target diet recommendation model, the diet fusion network determines the target diet recommendation information according to the first diet information to be recommended and the second diet information to be recommended. The target dietary recommendation information is pushed to the target user's target terminal. Secondly, this application also provides a dietary recommendation device, the dietary recommendation device comprising: The information acquisition module is used to acquire the current user information of the target user, including current dietary information, current target protein information in the body, and current metabolic dynamic information. The information prediction module is used to determine the first recommended dietary information based on the target dietary recommendation model's dietary recommendation prediction network, according to the target protein information in the body, and to determine the second recommended dietary information based on the current dietary information and the current metabolic dynamic information. The information fusion module is used to determine the target diet recommendation information based on the diet fusion network of the target diet recommendation model and the first diet information to be recommended and the second diet information to be recommended. The information recommendation module pushes the target dietary recommendation information to the target user's target terminal.

[0005] Thirdly, this application also provides a computer device, the computer device including a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the dietary recommendation method as described above.

[0006] Fourthly, this application also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the dietary recommendation method described above.

[0007] This application provides a dietary recommendation method, apparatus, device, and computer-readable storage medium. This application determines first recommended dietary information based on the user's current target protein information, thus focusing on the protein content in the user's body. It also determines second recommended dietary information based on current dietary information and current metabolic dynamics information, thus focusing on the user's dietary habits and other bodily indicators. Furthermore, by using the first and second recommended dietary information to determine the target dietary recommendation, it avoids situations where interactions between simultaneously consumed foods lead to less than expected effects, thereby improving the accuracy of dietary recommendations and enhancing the efficacy of the recommended target diet for the user. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic flowchart illustrating a dietary recommendation method provided in an embodiment of this application; Figure 2 This is a schematic diagram illustrating an application scenario of a dietary recommendation method provided in an embodiment of this application; Figure 3A schematic block diagram of a dietary recommendation device provided in an embodiment of this application; Figure 4 This is a schematic block diagram of the structure of a computer device according to an embodiment of this application.

[0010] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0011] 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, 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.

[0012] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0013] This application provides a dietary recommendation method, apparatus, computer equipment, and computer-readable storage medium. The dietary recommendation method can be applied to terminal devices, such as mobile phones, tablets, laptops, desktop computers, personal digital assistants, and wearable devices. It can also be applied to servers, such as standalone servers, server clusters, or cloud servers. The dietary recommendation method provided in this application aims to recommend anti-aging nutritional diets to users, thereby enabling them to manage their own health. The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0014] Please refer to Figure 1 and Figure 2 , Figure 1 This is a schematic flowchart of a dietary recommendation method provided for an embodiment of this application. Figure 2 This is a schematic diagram illustrating an application scenario of a dietary recommendation method provided in an embodiment of this application.

[0015] like Figure 1 As shown, the dietary recommendation method includes steps S101 to S104.

[0016] Step S101: Obtain the current user information of the target user, which includes current diet information, current target protein information in the body, and current metabolic dynamic information.

[0017] For example, the current user information corresponding to the target user for whom dietary information recommendations need to be made is obtained, so as to make dietary recommendations to the target user based on the current user information. In a specific implementation, the current user information is obtained by acquiring the input information of the target user on the target terminal.

[0018] For example, current user information is used to indicate the user's information corresponding to a first preset time period; current user information includes current dietary information, current target protein information in the body, and current metabolic dynamic information; for example, dietary information includes the user's personal dietary habits and / or regional dietary information, target protein information in the body is used to indicate one or more protein information of the user, and metabolic dynamic information is used to indicate the user's metabolic state. For example, current target protein information in the body can be predicted based on the user's historically input personal genotype information, such as APOE ε4 carrier, and historical target protein information in the body, or based on the user's recent physical examination information; while the user's metabolic dynamic information can be determined based on the user's specific physiological state information, such as being in a chronic inflammation phase, exercise recovery phase, or based on whether the user stayed up late or engaged in high-intensity exercise.

[0019] By obtaining the target user's current user information and recommending dietary information based on that information, the recommended dietary information can be more aligned with the user's recent lifestyle and physical condition, thereby enhancing the effectiveness of the recommended dietary information on the user's body.

[0020] Step S102: Based on the dietary recommendation prediction network of the target dietary recommendation model, determine the first recommended dietary information according to the current target protein information in the body, and determine the second recommended dietary information according to the current dietary information and the current metabolic dynamic information.

[0021] For example, the current user information is input into the target diet recommendation model to recommend dietary information; specifically, the target diet recommendation model includes a diet recommendation prediction network, in which the first recommended dietary information is predicted by the current target protein information in the body, and the second recommended dietary information is determined by the current diet information and the current metabolic dynamic information.

[0022] In a specific implementation, the dietary recommendation prediction network is built on the LSTM network architecture. The dietary recommendation prediction network can convert the input current target protein information in the body into a first vector, and determine the matching diet through Euclidean distance or other methods, thereby generating the first dietary information to be recommended; similarly, it can convert the current diet information and the current metabolic dynamic information into corresponding vectors to determine the second dietary information to be recommended.

[0023] It should be noted that the first and second recommended dietary information should include at least a variety of foods or dishes, as well as the proportions of each food or dish.

[0024] In some embodiments, the dietary recommendation prediction network based on the target dietary recommendation model determines the first recommended dietary information based on the current in vivo target protein information, including: a first recommendation layer based on the dietary recommendation prediction network, determining telomerase activity information and a first recommended food based on the current in vivo target protein information; a first ratio recommendation layer based on the dietary recommendation prediction network, adjusting the first recommended ratio between nucleic acid-containing foods and polyphenol-containing foods based on the telomerase activity information when the first recommended food is determined to include nucleic acid-containing foods and polyphenol-containing foods; and a first information determination layer based on the dietary recommendation prediction network, determining the first recommended dietary information based on the nucleic acid-containing foods, the polyphenol-containing foods, and the first recommended ratio.

[0025] For example, the dietary recommendation prediction network includes multiple hierarchical structures to process input information based on each layer, thereby determining the first recommended dietary information within the network. Specifically, in the first recommendation layer, telomerase activity information and the first recommended food are determined based on current target protein information in the body. This first recommended food is then passed to the first ratio recommendation layer to determine a first recommended ratio between nucleic acid-containing foods and polyphenol-containing foods. The first recommended food and the first recommended ratio are then used to determine the first recommended dietary information.

[0026] For example, current information about the target protein in the body includes telomerase concentration and the concentration of telomerase synthesis products. Therefore, telomerase activity information can be determined by the concentration of telomerase and the concentration of telomerase synthesis products. Furthermore, based on the concentrations of other proteins included in the current target protein information, such as interleukin-6 (IL-6) concentration, the first recommended food can be determined. In specific implementation, if the first-generation recommended food includes nucleic acid-containing foods such as salmon roe and polyphenol-rich foods, the ratio of these two types of foods is adjusted according to the telomerase activity information to slow down the annual telomerase loss rate, thereby achieving the user's anti-aging nutritional dietary recommendations.

[0027] In some embodiments, the dietary recommendation prediction network based on the target dietary recommendation model determines second recommended dietary information based on the current dietary information and the current metabolic dynamic information, including: a second recommendation layer based on the dietary recommendation prediction network, determining multiple second recommended foods and a second recommended ratio among the second recommended foods based on the current metabolic dynamic information; a second ratio recommendation layer based on the dietary recommendation prediction network, adjusting the flavor of each second recommended food and adjusting the second recommended ratio of each second recommended food based on the current dietary information; and a second information determination layer based on the dietary recommendation prediction network, determining the second recommended dietary information based on the second recommended foods with adjusted flavors and the adjusted second recommended ratios.

[0028] For example, the dietary recommendation prediction network also includes a second recommendation layer, a second ratio recommendation layer, and a second information determination layer. Specifically, in the second recommendation layer, multiple second recommended foods and second recommended ratios among these foods are determined based on current metabolic dynamics information. Then, the second recommended foods and their ratios are input into the second ratio recommendation layer to adjust the ratios. Finally, in the second information determination layer, the second recommended dietary information is determined based on the adjusted ratios and the flavor-adjusted second recommended foods. It should be noted that the dietary recommendation prediction network can be adjusted according to actual data processing needs; therefore, the specific structure of the dietary recommendation prediction network is not limited here.

[0029] For example, current metabolic dynamics information can be used to determine multiple second recommended foods and their recommended proportions. Specifically, this information can be used to determine the user's physical state, such as whether they have just stayed up late or engaged in high-intensity exercise, and then recommend corresponding foods to help them adjust. In practice, if the current metabolic dynamics information determines that the user has just stayed up late or engaged in high-intensity exercise, then onions and spinach can be identified as second recommended foods to provide quercetin. After identifying the second recommended foods, their flavors are adjusted based on the user's current dietary information, including their taste preferences and food preferences. This allows for adjustments to the flavors and proportions of the recommended foods to better suit the user's tastes, thereby increasing the user's adherence to the recommended dietary information.

[0030] Step S103: Based on the dietary fusion network of the target dietary recommendation model, determine the target dietary recommendation information according to the first recommended dietary information and the second recommended dietary information.

[0031] For example, after determining the first and second recommended dietary information in the dietary recommendation prediction network, they are input into the dietary fusion network so that the target dietary recommendation information can be determined in the dietary fusion network based on the first and second recommended dietary information.

[0032] It is important to understand that since the first and second recommended dietary information are predicted based on different current user information, there may be situations where consuming them together is not effective or suitable. Through the dietary fusion network, the first and second recommended dietary information are fused to make the target dietary recommendations more reasonable and better for the user.

[0033] In some embodiments, the dietary fusion network based on the target dietary recommendation model determines target dietary recommendation information based on the first recommended dietary information and the second recommended dietary information, including: a conditional decision layer based on the dietary fusion network determining whether the first recommended dietary information and the second recommended dietary information satisfy a non-mutually exclusive condition; a first information processing layer based on the dietary fusion network concatenating the first recommended dietary information and the second recommended dietary information to obtain the target dietary recommendation information if the non-mutually exclusive condition is satisfied; and a second information processing layer based on the dietary fusion network adjusting the first recommended dietary information and / or adjusting the second recommended dietary information if the non-mutually exclusive condition is not satisfied, and determining the target dietary recommendation information based on the adjusted recommended dietary information.

[0034] For example, it is determined whether the first recommended dietary information and the second recommended dietary information satisfy the non-mutually exclusive condition based on the first food contained in the first recommended dietary information and the second food contained in the second recommended dietary information.

[0035] In a specific implementation, the first recommended dietary information and the second recommended dietary information are determined to meet the non-mutually exclusive condition based on information such as whether the first food and the second food are suitable to be eaten together and whether eating them together will reduce the nutritional value of the food.

[0036] For example, if the condition determination layer determines that the first recommended dietary information and the second recommended dietary information meet the non-mutually exclusive condition, then the first recommended dietary information and the second recommended dietary information are transmitted to the first information processing layer for processing; otherwise, if the non-mutually exclusive condition is not met, then the first recommended dietary information and the second recommended dietary information are transmitted to the second information processing layer for processing.

[0037] For example, if the first recommended dietary information and the second recommended dietary information meet the condition of non-mutual exclusion, the first recommended dietary information and the second recommended dietary information are spliced ​​together to obtain the target dietary recommendation information; it should be understood that the information splicing can be context splicing and / or information insertion, and this application does not limit the specific method of information splicing.

[0038] If the first and second recommended dietary information do not meet the non-mutually exclusive condition, the first and / or second recommended dietary information shall be adjusted to determine the target dietary recommendation information. It should be understood that adjusting the recommended dietary information includes, but is not limited to, replacing, adding, and / or reducing the foods included in the recommended dietary information, and adjusting the recommended proportions, so that the adjusted first and second recommended dietary information can meet the non-mutually exclusive condition. The target dietary recommendation information is then determined based on the adjusted first and second recommended dietary information that meet the non-mutually exclusive condition.

[0039] Step S104: Push the target diet recommendation information to the target user's target terminal.

[0040] For example, after obtaining the target dietary recommendation information, the target dietary recommendation information is sent to the target user's target terminal so that the target user is aware of the target dietary recommendation information and can then follow the recommended diet for a healthy lifestyle, thereby achieving the user's anti-aging dietary goals. In some implementations, the target terminal includes, but is not limited to, the user's mobile phone, personal computer, wearable device, etc.

[0041] In some embodiments, the method further includes: acquiring target training data, the target training data including historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, and historical dietary recommendation information; updating the model based on preset parameters, determining update parameters according to the target training data; and updating the parameters of the dietary recommendation model according to the update parameters to obtain the target dietary recommendation model.

[0042] For example, target training data is obtained to obtain update parameters for the dietary recommendation model, and the dietary recommendation model is updated by updating the update parameters to improve the recommendation accuracy of the dietary recommendation model.

[0043] In the specific implementation process, historical dietary information, historical target protein information, historical metabolic dynamics information, and historical dietary recommendation information are input into the preset parameter update model to obtain the update parameters output by the preset parameter update model. In this way, the dietary recommendation model can be updated by updating the update parameters to obtain the target dietary recommendation model.

[0044] For example, the update time can be preset, such as once a week, so that the dietary recommendation model is updated every week to improve the recommendation accuracy of the dietary recommendation model.

[0045] In some embodiments, determining the update parameters based on the target training data includes: acquiring learning parameters, discount parameters, and historical reward parameters; determining the current reward parameters based on the historical metabolic dynamics information and the historical reward parameters; and determining the update parameters based on the learning parameters, historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, historical dietary recommendation information, the current reward parameters, and the discount parameters.

[0046] For example, the learning parameters, discount parameters, and historical reward parameters are obtained. The sum of the learning parameters and discount parameters is 1, and their specific values ​​can be set according to the actual model update requirements. The historical reward parameters are the reward parameters in the historical parameter updates.

[0047] In some implementation processes, the current reward parameters are determined by historical metabolic dynamics information and historical reward parameters. After determining the current reward parameters, the updated parameters are determined based on learning parameters, historical dietary information, historical target protein information in the body, historical metabolic dynamics information, historical dietary recommendations, current reward parameters, and discount parameters.

[0048] In some embodiments, determining the current reward parameter based on the historical metabolic dynamics information and the historical reward parameter includes: when the ratio of coenzyme I to reduced coenzyme I included in the metabolic dynamics information is greater than or equal to a preset ratio threshold, multiplying the historical reward parameter by a preset coefficient to obtain the current reward parameter; and when the ratio of coenzyme I to reduced coenzyme I is less than the preset ratio threshold, determining the historical reward parameter as the current reward parameter.

[0049] For example, taking the ratio of coenzyme I (NAD+) to reduced coenzyme I (NADH) as an example of metabolic dynamic information, if the ratio is greater than or equal to a preset ratio threshold, the historical reward parameter is multiplied by a preset coefficient to obtain the current reward parameter; if the ratio is less than the preset ratio threshold, the reward parameter is determined to remain unchanged, that is, the historical reward parameter is determined as the current reward parameter.

[0050] Specifically, the preset ratio threshold is 0.3, and the preset coefficient is 1.5.

[0051] For example, historical dietary information is divided into first historical dietary information and second historical dietary information, with the time corresponding to the first historical dietary information being earlier than the time corresponding to the second historical dietary information; and historical target protein information and historical metabolic dynamic information are processed in the same way as historical dietary information to obtain the corresponding first historical information and second historical information, and the first historical dietary recommendation information and the second historical dietary recommendation information are determined.

[0052] The relevant values ​​are determined based on the first historical dietary information and the corresponding first historical dietary information, first historical target protein information in vivo, and first historical metabolic dynamic information. The maximum value of the update parameter is determined based on the second historical dietary information, second historical target protein information in vivo, and second historical metabolic dynamic information. The process of determining the update parameter can be as follows: Update parameter = (1 - learning parameter) × relevant value + learning parameter × (current reward parameter + discount parameter × maximum update parameter value).

[0053] In some implementations, the learning parameter is set to 0.1 and the discount parameter is set to 0.9.

[0054] It should be noted that after determining the update parameters as described above, the dietary recommendation model is updated using these parameters to improve its accuracy.

[0055] The dietary recommendation method provided in the above embodiments determines the first recommended dietary information based on the user's current target protein information, thereby focusing on the protein in the user's body. It also determines the second recommended dietary information based on the current dietary information and current metabolic dynamic information, thereby focusing on the user's eating habits and other physical indicators. Furthermore, by determining the recommended target dietary information based on the first and second recommended dietary information, it avoids situations where the interaction between foods eaten at the same time leads to less than expected results, thus improving the accuracy of dietary recommendations and enhancing the efficacy of the recommended target diet for the user.

[0056] Please see Figure 3 , Figure 3 This is a schematic diagram of a dietary recommendation device provided in an embodiment of this application. The dietary recommendation device can be installed in a computer device to perform the aforementioned dietary recommendation method.

[0057] like Figure 3 As shown, the dietary recommendation device 100 includes: an information acquisition module 110, an information prediction module 120, an information fusion module 130, and an information recommendation module 140.

[0058] Information acquisition module 110 is used to acquire the current user information of the target user, including current diet information, current target protein information in the body and current metabolic dynamic information; The information prediction module 120 is used to determine the first recommended dietary information based on the target dietary recommendation model and the dietary recommendation prediction network based on the current target protein information in the body, and to determine the second recommended dietary information based on the current dietary information and the current metabolic dynamic information. Information fusion module 130 is used to determine target dietary recommendation information based on the dietary fusion network of the target dietary recommendation model, according to the first dietary information to be recommended and the second dietary information to be recommended. The information recommendation module 140 pushes the target dietary recommendation information to the target user's target terminal.

[0059] For example, the information prediction module 120 includes a first information determination submodule, a first information adjustment submodule, and a second information determination submodule.

[0060] The first information determination submodule is used to determine telomerase activity information and the first recommended food based on the first information determination layer of the dietary recommendation prediction network and the current target protein information in the body.

[0061] The first information adjustment submodule is used to adjust the first recommended ratio between nucleic acid-containing foods and polyphenol-containing foods based on the first information adjustment layer of the dietary recommendation prediction network, when it is determined that the first recommended food includes nucleic acid-containing foods and polyphenol-containing foods.

[0062] The second information determination submodule is used to determine the first dietary information to be recommended based on the second information determination layer of the dietary recommendation prediction network, according to the nucleic acid-containing foods, the polyphenolic foods, and the recommended ratio.

[0063] For example, the information prediction module 120 includes a third information determination submodule, a second information adjustment submodule, and a fourth information determination submodule.

[0064] The third information determination submodule is used to determine, based on the third information determination layer of the dietary recommendation prediction network, a variety of second recommended foods and a second recommendation ratio between each of the second recommended foods according to the current metabolic dynamic information. The second information adjustment submodule is used to adjust the flavor of each of the second recommended foods and the second recommended ratio of each of the second recommended foods according to the current dietary information based on the second information adjustment layer of the dietary recommendation prediction network. The fourth information determination submodule is used to determine the second recommended dietary information based on the fourth information determination layer of the dietary recommendation prediction network, according to the second recommended food with adjusted flavor and the adjusted second recommendation ratio.

[0065] For example, the information fusion module 130 includes a condition judgment submodule and an information processing submodule.

[0066] The condition judgment submodule is used to determine, based on the condition judgment layer of the dietary fusion network, whether the first recommended dietary information and the second recommended dietary information satisfy the non-mutually exclusive condition.

[0067] The information processing submodule is configured to, based on the first information processing layer of the dietary fusion network, perform information splicing processing on the first dietary information to be recommended and the second dietary information to be recommended, under the condition of satisfying the non-mutual exclusion condition, to obtain the target dietary recommendation information; and to, based on the second information processing layer of the dietary fusion network, adjust the first dietary information to be recommended and / or adjust the second dietary information to be recommended, under the condition of not satisfying the non-mutual exclusion condition, and determine the target dietary recommendation information based on the adjusted dietary information to be recommended.

[0068] For example, the diet recommendation device 100 also includes a training data acquisition module, an update parameter determination module, and a model update module.

[0069] The training data acquisition module is used to acquire target training data, which includes historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, and historical dietary recommendation information.

[0070] The update parameter determination module is used to update the model based on preset parameters and determine the update parameters according to the target training data.

[0071] The model update module is used to update the parameters of the dietary recommendation model according to the update parameters to obtain the target dietary recommendation model.

[0072] For example, the update parameter determination module also includes a parameter acquisition submodule, a reward parameter determination submodule, and an update parameter determination submodule.

[0073] The parameter acquisition submodule is used to obtain learning parameters, discount parameters, and historical reward parameters.

[0074] The reward parameter determination submodule is used to determine the current reward parameter based on the historical metabolic dynamic information and the historical reward parameters.

[0075] The parameter update determination submodule is used to determine the update parameters based on the learning parameters, historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, historical dietary recommendation information, current reward parameters, and discount parameters.

[0076] For example, the reward parameter determination submodule is further configured to, when the ratio of coenzyme I to reduced coenzyme I included in the metabolic dynamics information is greater than or equal to a preset ratio threshold, multiply the historical reward parameter by a preset coefficient to obtain the current reward parameter; and when the ratio of coenzyme I to reduced coenzyme I is less than the preset ratio threshold, determine the historical reward parameter as the current reward parameter.

[0077] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the above-described apparatus and its modules and units can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0078] Please see Figure 4 , Figure 4 This is a schematic block diagram illustrating the structure of a computer device provided in an embodiment of this application. The computer device may be a server or a terminal.

[0079] like Figure 4 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0080] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any dietary recommendation method.

[0081] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0082] Internal memory provides an environment for the execution of computer programs stored in non-volatile storage media, which, when executed by a processor, enable the processor to perform any dietary recommendation method.

[0083] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0084] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0085] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: Obtain the target user's current user information, which includes current dietary information, current target protein information in the body, and current metabolic dynamic information; The dietary recommendation prediction network based on the target dietary recommendation model determines the first recommended dietary information based on the current target protein information in the body, and determines the second recommended dietary information based on the current dietary information and the current metabolic dynamic information. Based on the target diet recommendation model, the diet fusion network determines the target diet recommendation information according to the first diet information to be recommended and the second diet information to be recommended. The target dietary recommendation information is pushed to the target user's target terminal.

[0086] In one embodiment, when the processor implements a dietary recommendation prediction network based on the target dietary recommendation model and determines the first dietary information to be recommended based on the current in vivo target protein information, it is configured to: Based on the first information determination layer of the dietary recommendation prediction network, telomerase activity information and the first recommended food are determined according to the current target protein information in the body. Based on the first information adjustment layer of the dietary recommendation prediction network, when it is determined that the first recommended food includes nucleic acid-containing foods and polyphenol-containing foods, the first recommended ratio between nucleic acid-containing foods and polyphenol-containing foods is adjusted according to the telomerase activity information. Based on the second information determination layer of the dietary recommendation prediction network, the first dietary information to be recommended is determined according to the nucleic acid-containing foods, the polyphenolic foods, and the recommended proportions.

[0087] In one embodiment, when the processor implements a dietary recommendation prediction network based on the target dietary recommendation model and determines the second dietary information to be recommended based on the current dietary information and the current metabolic dynamic information, it is configured to: Based on the third information determination layer of the dietary recommendation prediction network, a variety of second recommended foods and a second recommendation ratio among the second recommended foods are determined according to the current metabolic dynamic information. Based on the second information adjustment layer of the dietary recommendation prediction network, the flavor of each of the second recommended foods is adjusted according to the current dietary information, and the second recommendation ratio of each of the second recommended foods is adjusted. Based on the fourth information determination layer of the dietary recommendation prediction network, the second recommended dietary information is determined according to the second recommended food with adjusted flavor and the adjusted second recommendation ratio.

[0088] In one embodiment, when the processor implements a dietary fusion network based on the target dietary recommendation model and determines target dietary recommendation information according to the first recommended dietary information and the second recommended dietary information, it is configured to: Based on the conditional decision layer of the dietary fusion network, it is determined whether the first dietary information to be recommended and the second dietary information to be recommended satisfy the non-mutually exclusive condition; Based on the first information processing layer of the dietary fusion network, under the condition of satisfying the non-mutually exclusive condition, the first dietary information to be recommended and the second dietary information to be recommended are spliced ​​together to obtain the target dietary recommendation information; Based on the second information processing layer of the dietary fusion network, if the non-mutually exclusive condition is not met, the first recommended dietary information and / or the second recommended dietary information are adjusted, and the target dietary recommendation information is determined based on the adjusted recommended dietary information.

[0089] In one embodiment, the processor, when implementing the dietary recommendation method, is further configured to implement: Acquire target training data, which includes historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, and historical dietary recommendation information; The model is updated based on preset parameters, and the update parameters are determined according to the target training data. The parameters of the dietary recommendation model are updated based on the updated parameters to obtain the target dietary recommendation model.

[0090] In one embodiment, when the processor determines the update parameters based on the target training data, it is configured to: Obtain learning parameters, discount parameters, and historical reward parameters; The current reward parameters are determined based on the historical metabolic dynamics information and the historical reward parameters. The update parameters are determined based on the learning parameters, historical dietary information, historical target protein information, historical metabolic dynamics information, historical dietary recommendations, current reward parameters, and discount parameters.

[0091] In one embodiment, when the processor determines the current reward parameter based on the historical metabolic dynamics information and the historical reward parameters, it is configured to: If the ratio of coenzyme I to reduced coenzyme I in the metabolic dynamics information is greater than or equal to a preset ratio threshold, the historical reward parameter is multiplied by a preset coefficient to obtain the current reward parameter. If the ratio of coenzyme I to reduced coenzyme I is less than the preset ratio threshold, the historical reward parameter is determined as the current reward parameter.

[0092] This application also provides a computer-readable storage medium storing a computer program, the computer program including program instructions, and the method implemented when the program instructions are executed can refer to various embodiments of the dietary recommendation method of this application.

[0093] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0094] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0095] It should also be understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. It should be noted that, herein, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0096] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above descriptions are merely specific implementations of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A dietary recommendation method, characterized in that, include: Obtain the target user's current user information, which includes current dietary information, current target protein information in the body, and current metabolic dynamic information; The dietary recommendation prediction network based on the target dietary recommendation model determines the first recommended dietary information based on the current target protein information in the body, and determines the second recommended dietary information based on the current dietary information and the current metabolic dynamic information. Based on the target diet recommendation model, the diet fusion network determines the target diet recommendation information according to the first diet information to be recommended and the second diet information to be recommended. The target dietary recommendation information is pushed to the target user's target terminal.

2. The dietary recommendation method as described in claim 1, characterized in that, The dietary recommendation prediction network based on the target dietary recommendation model determines the first recommended dietary information based on the current in vivo target protein information, including: Based on the first information determination layer of the dietary recommendation prediction network, telomerase activity information and the first recommended food are determined according to the current target protein information in the body. Based on the first information adjustment layer of the dietary recommendation prediction network, when it is determined that the first recommended food includes nucleic acid-containing foods and polyphenol-containing foods, the first recommended ratio between nucleic acid-containing foods and polyphenol-containing foods is adjusted according to the telomerase activity information. Based on the second information determination layer of the dietary recommendation prediction network, the first dietary information to be recommended is determined according to the nucleic acid-containing foods, the polyphenolic foods, and the first recommendation ratio.

3. The dietary recommendation method as described in claim 1, characterized in that, The dietary recommendation prediction network based on the target dietary recommendation model determines the second recommended dietary information according to the current dietary information and the current metabolic dynamic information, including: Based on the third information determination layer of the dietary recommendation prediction network, a variety of second recommended foods and a second recommendation ratio among the second recommended foods are determined according to the current metabolic dynamic information. Based on the second information adjustment layer of the dietary recommendation prediction network, the flavor of each of the second recommended foods is adjusted according to the current dietary information, and the second recommendation ratio of each of the second recommended foods is adjusted. Based on the fourth information determination layer of the dietary recommendation prediction network, the second recommended dietary information is determined according to the second recommended food with adjusted flavor and the adjusted second recommendation ratio.

4. The dietary recommendation method as described in claim 1, characterized in that, The dietary fusion network based on the target diet recommendation model determines target diet recommendation information according to the first diet information to be recommended and the second diet information to be recommended, including: Based on the conditional decision layer of the dietary fusion network, it is determined whether the first dietary information to be recommended and the second dietary information to be recommended satisfy the non-mutually exclusive condition; Based on the first information processing layer of the dietary fusion network, under the condition of satisfying the non-mutually exclusive condition, the first dietary information to be recommended and the second dietary information to be recommended are spliced ​​together to obtain the target dietary recommendation information; Based on the second information processing layer of the dietary fusion network, if the non-mutually exclusive condition is not met, the first recommended dietary information and / or the second recommended dietary information are adjusted, and the target dietary recommendation information is determined based on the adjusted recommended dietary information.

5. The dietary recommendation method according to any one of claims 1-4, characterized in that, The method further includes: Acquire target training data, which includes historical dietary information, historical in vivo target protein information, historical metabolic dynamics information, and historical dietary recommendation information; The model is updated based on preset parameters, and the update parameters are determined according to the target training data. The parameters of the dietary recommendation model are updated based on the updated parameters to obtain the target dietary recommendation model.

6. The dietary recommendation method as described in claim 5, characterized in that, The step of determining the update parameters based on the target training data includes: Obtain learning parameters, discount parameters, and historical reward parameters; The current reward parameters are determined based on the historical metabolic dynamics information and the historical reward parameters. The update parameters are determined based on the learning parameters, historical dietary information, historical target protein information, historical metabolic dynamics information, historical dietary recommendations, current reward parameters, and discount parameters.

7. The dietary recommendation method as described in claim 6, characterized in that, The step of determining the current reward parameter based on the historical metabolic dynamic information and the historical reward parameters includes: If the ratio of coenzyme I to reduced coenzyme I in the metabolic dynamics information is greater than or equal to a preset ratio threshold, the historical reward parameter is multiplied by a preset coefficient to obtain the current reward parameter. If the ratio of coenzyme I to reduced coenzyme I is less than the preset ratio threshold, the historical reward parameter is determined as the current reward parameter.

8. A dietary recommendation device, characterized in that, The dietary recommendation device includes: The information acquisition module is used to acquire the current user information of the target user, including current dietary information, current target protein information in the body, and current metabolic dynamic information. The information prediction module is used to determine the first recommended dietary information based on the current target protein information in the body and the second recommended dietary information based on the current dietary information and the current metabolic dynamic information, according to the dietary recommendation prediction network of the target dietary recommendation model. The information fusion module is used to determine the target diet recommendation information based on the diet fusion network of the target diet recommendation model and the first diet information to be recommended and the second diet information to be recommended. The information recommendation module pushes the target dietary recommendation information to the target user's target terminal.

9. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable by the processor, wherein when the computer program is executed by the processor, it implements the steps of the dietary recommendation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the steps of the dietary recommendation method as described in any one of claims 1 to 7.

Images