Personalized nutrition matching management system and method based on AI generation

Through the AI-based nutrition matching management system, the basic information and nutritional needs of users are used to semantic code and fusion, and personalized recipes are generated, which solves the problem that traditional systems cannot meet individual differences and improves the intelligence and user experience of nutrition matching management.

CN118658588BActive Publication Date: 2024-11-12JILIN UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411149308.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-12
Estimated Expiration
2044-08-21

AI Technical Summary

Technical Problem

The traditional nutrition matching management system cannot fully consider individual differences, lacks dynamic adjustment and interactivity, cannot meet users' personalized needs, and lacks fun and experience.

Method used

Using an AI-based personalized nutrition matching management system, we use user basic information collection, semantic coding and fusion, combined with nutritional needs description to conduct semantic understanding, generate personalized recipes, and use 3D technology to preview the food production process to increase interactivity and fun.

Benefits of technology

It has realized the automatic generation of personalized recipes based on user basic situation and nutritional needs, improve the intelligent level of nutrition matching management, and enhance user experience and food calorie control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118658588B_ABST
    Figure CN118658588B_ABST
Patent Text Reader

Abstract

The present application discloses a personalized nutrition matching management system and method based on AI generation, which relates to the field of nutrition matching management. It collects the basic information of the user, including age, gender, weight, height, health status, allergies and nature of work, and supports the user to independently input the nutrition demand description text, and then introduces the data processing and semantic understanding algorithm based on artificial intelligence and natural language processing technology in the back end to analyze the user's basic information and nutrition demand description text, so as to capture the semantics of the user's basic information and the semantics of the user's demand description text, and then use the semantics of the user's basic information to constrain the expression of the semantics of the user's demand description text, so as to recommend a matching recipe for the user. In this way, the recommended user nutrition matching can be automatically generated according to the user's basic situation and nutritional needs, so as to generate personalized recipes in a more intelligent way.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of nutrition pairing management, and specifically to a personalized nutrition pairing management system and method based on AI generation. Background Art

[0002] Nutritional matching management refers to scientifically and rationally planning daily diets based on personal physical conditions, health conditions, lifestyles and other factors to achieve balanced nutrition, healthy weight loss or specific health goals. With the improvement of people's living standards and the enhancement of health awareness, more and more people are beginning to pay attention to how to improve their health and quality of life through a reasonable diet. However, traditional nutritional matching management systems are usually based on general standards or rules, and cannot fully take into account individual differences. In other words, everyone's physical condition, health needs and preferences are different, so universal solutions often cannot meet the personalized needs of users. In addition, traditional nutritional matching management systems are usually static and cannot dynamically adjust nutritional matching plans according to their own conditions. At the same time, it is difficult for users to interact with the system, lacking fun and experience.

[0003] Therefore, an optimized nutritional combination management program is desired. Summary of the invention

[0004] This application is made in consideration of the above problems. One purpose of this application is to provide a personalized nutrition matching management system and method based on AI generation.

[0005] The present application provides a personalized nutrition matching management system based on AI generation, which includes:

[0006] A user basic information collection module is used to obtain the user's basic information data, wherein the basic information data includes age, gender, weight, height, health status, allergies and job nature;

[0007] A user nutritional requirement description acquisition module, used to acquire a nutritional requirement description text input by the user;

[0008] A user basic information semantic encoding module, used to perform semantic encoding on the user's basic information data to obtain a sequence of user basic information semantic feature vectors;

[0009] A user basic information semantic fusion module, used for passing the sequence of user basic information semantic feature vectors through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector;

[0010] A nutrition requirement semantic understanding module, used for performing word granularity-based semantic coding on the nutrition requirement description text to obtain a sequence of word granularity semantic coding feature vectors of the nutrition requirement description text;

[0011] A module for expressing the semantics of nutritional needs under the constraints of basic information, which is used to express the sequence of the semantic encoding feature vectors of the nutritional needs description text word granularity based on the semantic fusion feature vector of the enhanced user basic information to obtain a semantic representation vector of the nutritional needs description text under the semantic constraints of the user basic information;

[0012] A recommended recipe generation module is used to generate a recommended recipe based on the semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information;

[0013] The user basic information semantic fusion module includes:

[0014] A mean vector calculation unit, used to calculate the mean vector of the sequence of the user basic information semantic feature vectors to obtain the user basic information semantic cluster center feature vector;

[0015] A semantic correlation coefficient calculation unit, used to calculate the semantic correlation coefficient between each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors and the user basic information semantic cluster center feature vector to obtain a sequence of user basic information cluster center semantic correlation coefficients;

[0016] A semantic anti-correlation unit, used for taking the inverse of the semantic correlation coefficient of each user basic information clustering center in the sequence of semantic correlation coefficients of the user basic information clustering center to obtain a sequence of semantic anti-correlation coefficients of the user basic information clustering center;

[0017] A semantic anti-correlation coefficient normalization unit, used to normalize the sequence of semantic anti-correlation coefficients of the user basic information clustering center using a Sigmoid function to obtain a sequence of semantic anti-correlation weight coefficients of the user basic information clustering center;

[0018] A weighting unit, configured to weight each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors by using each user basic information cluster center semantic anti-correlation weight coefficient in the sequence of user basic information cluster center semantic anti-correlation weight coefficients as a weight to obtain a sequence of suppressed user basic information semantic feature vectors;

[0019] A position difference unit, used for calculating the position difference between each group of corresponding user basic information semantic feature vectors and suppressed user basic information semantic feature vectors in the sequence of user basic information semantic feature vectors and the sequence of suppressed user basic information semantic feature vectors to obtain a sequence of enhanced user basic information semantic feature vectors;

[0020] The position-based mean unit is used to calculate the position-based mean vector of the sequence of the enhanced user basic information semantic feature vectors to obtain the enhanced user basic information semantic fusion feature vector.

[0021] For example, according to the personalized nutrition matching management system based on AI generation of the present application, the semantic correlation coefficient calculation unit is used to:

[0022] Calculating the norms of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector respectively to obtain a user basic information semantic coding norm representation value and a user basic information semantic cluster center norm representation value;

[0023] Calculate the product of the user basic information semantic code-norm representation value and the user basic information semantic cluster center-norm representation value to obtain the user basic information semantic-cluster center semantic correlation-norm representation value;

[0024] Calculate the multiplication between the transposed vector of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector to obtain the user basic information semantic-cluster center semantic correlation feature value;

[0025] The division between the user basic information semantics-cluster center semantic relevance feature value and the user basic information semantics-cluster center semantic relevance-norm representation value is calculated to obtain the user basic information cluster center semantic relevance coefficient.

[0026] For example, according to the AI-generated personalized nutrition matching management system of the present application, the nutrition requirement semantic understanding module is used to:

[0027] The nutritional requirement description text is word segmented and then passed through a semantic encoder including a word embedding layer to obtain a sequence of word granularity semantic encoding feature vectors of the nutritional requirement description text.

[0028] For example, according to the AI-generated personalized nutrition matching management system of the present application, the nutrition requirement semantic expression module under the basic information constraint is used to:

[0029] Taking the enhanced user basic information semantic fusion feature vector as the constraint vector, the constraint vector and the sequence of the word granularity semantic encoding feature vector of the nutritional need description text are passed through the nutritional need description text global word granularity semantic aggregation module based on unidirectional attention constraint expression to obtain the semantic representation vector of the nutritional need description text under the semantic constraint of the user basic information.

[0030] For example, according to the AI-generated personalized nutrition matching management system of the present application, the nutrition requirement semantic expression module under the basic information constraint includes:

[0031] A semantic similarity calculation unit, used for calculating the semantic similarity between the constraint vector and each nutrition requirement description text word granularity semantic encoding feature vector in the sequence of nutrition requirement description text word granularity semantic encoding feature vectors to obtain a sequence of nutrition requirement description text word granularity one-way matching semantic measurement coefficients;

[0032] A one-way matching semantic measurement coefficient normalization unit is used to normalize the sequence of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity to obtain a sequence of one-way matching semantic measurement weight values ​​of the nutritional requirement description text word granularity;

[0033] The weighted sum unit is used to calculate the weighted sum of the sequence of semantic encoding feature vectors of the nutritional need description text word granularity using the sequence of one-way matching semantic measurement weight values ​​of the nutritional need description text word granularity as weights to obtain the semantic representation vector of the nutritional need description text under the semantic constraints of the user's basic information.

[0034] For example, according to the personalized nutrition matching management system based on AI generation of the present application, the semantic similarity calculation unit is used to:

[0035] Cascading each of the constraint vectors and the sequence of the granular semantic encoding feature vectors of the nutrient requirement description text to obtain a set of query nutrient requirement description text granular semantic encoding joint feature vectors;

[0036] After respectively calculating the matrix product of each query nutritional requirement description text word granularity semantic coding joint feature vector in the set of the query nutritional requirement description text word granularity semantic coding joint feature vector and the weight matrix, the obtained feature vector and the bias vector are added by position to obtain a sequence of query nutritional requirement description text word granularity semantic coding joint bias feature vectors;

[0037] The sequence of the semantic encoding of the query nutritional requirement description text word granularity and the bias feature vector is input into the sigmoid activation function to obtain the sequence of the one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity.

[0038] For example, according to the AI-generated personalized nutrition matching management system of the present application, the one-way matching semantic metric coefficient normalization unit is used to:

[0039] Using each one-way matching semantic measurement coefficient of the nutritional requirement description text word granularity in the sequence of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity as an exponent of a natural constant to calculate an exponential function value with the natural constant as the base to obtain a sequence of logarithmic values ​​of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity;

[0040] Calculate the sum of the logarithmic values ​​of the one-way matching semantic metric coefficients of the nutritional requirement description text word granularity in the sequence of the logarithmic value of the one-way matching semantic metric coefficients of the nutritional requirement description text word granularity to obtain the total sum of the one-way matching semantic metric values ​​of the nutritional requirement description text word granularity;

[0041] Each logarithm of the one-way matching semantic metric coefficient of the nutritional need description text word granularity in the sequence of logarithm of the one-way matching semantic metric coefficient of the nutritional need description text word granularity is divided by the total sum of the one-way matching semantic metric of the nutritional need description text word granularity to obtain a sequence of the one-way matching semantic metric weight values ​​of the nutritional need description text word granularity.

[0042] For example, according to the AI-generated personalized nutrition matching management system of the present application, the recommended matching recipe generation module is used to:

[0043] The semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information is passed through a decoder-based recipe generator to obtain a recommended recipe.

[0044] The present application also provides a personalized nutrition matching management method based on AI generation, which includes:

[0045] Obtaining basic information data of the user, wherein the basic information data includes age, gender, weight, height, health status, allergies and job nature;

[0046] Acquire the nutritional requirement description text input by the user;

[0047] Performing semantic encoding on the basic information data of the user to obtain a sequence of semantic feature vectors of the basic information of the user;

[0048] The sequence of user basic information semantic feature vectors is passed through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector;

[0049] Performing word-granularity-based semantic coding on the nutrition requirement description text to obtain a sequence of word-granularity semantic coding feature vectors of the nutrition requirement description text;

[0050] Based on the enhanced user basic information semantic fusion feature vector, a sequence of word granularity semantic encoding feature vectors of the nutritional demand description text is semantically constrained to obtain a semantic representation vector of the nutritional demand description text under the semantic constraint of the user basic information;

[0051] Generate a recommended recipe based on the semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information;

[0052] The sequence of the user basic information semantic feature vectors is passed through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector, including:

[0053] Calculating the mean vector of the sequence of the user basic information semantic feature vectors to obtain the user basic information semantic cluster center feature vector;

[0054] Calculating the semantic correlation coefficient between each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors and the user basic information semantic cluster center feature vector to obtain a sequence of user basic information cluster center semantic correlation coefficients;

[0055] Taking the reciprocal of each user basic information cluster center semantic correlation coefficient in the sequence of user basic information cluster center semantic correlation coefficients to obtain a sequence of user basic information cluster center semantic anti-correlation coefficients;

[0056] Using a Sigmoid function to normalize the sequence of semantic anti-correlation coefficients of the user basic information clustering center to obtain a sequence of semantic anti-correlation weight coefficients of the user basic information clustering center;

[0057] Using each user basic information cluster center semantic anti-correlation weight coefficient in the sequence of user basic information cluster center semantic anti-correlation weight coefficients as a weight, weighting each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors to obtain a sequence of suppressed user basic information semantic feature vectors;

[0058] Calculating the positional difference between each group of corresponding user basic information semantic feature vectors and suppressed user basic information semantic feature vectors in the sequence of user basic information semantic feature vectors and the sequence of suppressed user basic information semantic feature vectors to obtain a sequence of enhanced user basic information semantic feature vectors;

[0059] The position-wise mean vector of the sequence of the enhanced user basic information semantic feature vectors is calculated to obtain the enhanced user basic information semantic fusion feature vector.

[0060] According to the AI-generated personalized nutrition pairing management system and method of the present application, recommended user nutrition pairs can be automatically generated based on the user's basic situation and nutritional needs, so as to generate personalized recipes in a more intelligent way. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0062] Figure 1 A schematic diagram of the application architecture of a personalized nutrition matching management system generated based on AI in an embodiment of the present application is shown;

[0063] Figure 2 A structural schematic diagram of a personalized nutrition matching management system generated based on AI in an embodiment of the present application is shown;

[0064] Figure 3 It shows a schematic diagram of the structure of the user basic information semantic fusion module of the personalized nutrition matching management system generated based on AI in an embodiment of the present application;

[0065] Figure 4 A schematic diagram of the structure of a nutrition requirement semantic expression module under the basic information constraints of a personalized nutrition matching management system generated based on AI in an embodiment of the present application is shown;

[0066] Figure 5 A flowchart of a personalized nutrition matching management method based on AI generation in an embodiment of the present application is shown;

[0067] Figure 6 An application scenario diagram of a personalized nutrition matching management system generated based on AI in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0069] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0070] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0071] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0072] Figure 1 A schematic diagram of the application architecture of a personalized nutrition matching management system generated based on AI in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0073] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of local area network (LAN), metropolitan area network (MAN), wide area network (WAN), mobile, wired or wireless network, private network or virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0074] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0075] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the personalized nutrition matching management system generated based on AI in the embodiment of the present application.

[0076] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the personalized nutrition matching management methods generated based on AI in the embodiments of the present application according to the obtained program instructions.

[0077] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0078] The following is a non-limiting description of the AI-generated personalized nutrition matching management system provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0079] In view of the above technical problems, in the technical solution of the present application, a personalized nutrition matching management system based on AI generation is proposed, which can be applied to nutrition matching experience machines in various scenarios (such as scenic spots, parks, hospitals, nursing homes and schools). It can collect basic information of users, including age, gender, weight, height, health status, allergies and nature of work, and support users to independently input nutrition demand description text, and then introduce data processing and semantic understanding algorithms based on artificial intelligence and natural language processing technology in the back end to analyze the basic information of users and the description text of nutrition demand, so as to capture the semantics of user basic information and user demand description text, and then use the semantics of user basic information to constrain the expression of user demand description text semantics, so as to recommend matching recipes for the user. In this way, the recommended user nutrition matching can be automatically generated according to the basic situation and nutritional needs of the user, so as to generate personalized recipes in a more intelligent way. In addition, users can also preview the food production process and the final product through the system's 3D technology, which not only increases interactivity and fun, but also ensures that users make and consume food according to the food production process, and ensures that the calorie intake of food is within a controllable range.

[0080] Figure 2The structure diagram of the personalized nutrition matching management system 800 generated based on AI in the embodiment of the present application is shown. The personalized nutrition matching management system 800 generated based on AI includes: a user basic information acquisition module 810, which is used to obtain the basic information data of the user, wherein the basic information data includes age, gender, weight, height, health status, allergies and nature of work; a user nutrition requirement description acquisition module 820, which is used to obtain the nutrition requirement description text input by the user; a user basic information semantic encoding module 830, which is used to semantically encode the basic information data of the user to obtain a sequence of user basic information semantic feature vectors; a user basic information semantic fusion module 840, which is used to combine the sequence of user basic information semantic feature vectors through user basic information key semantics based on feature distribution reverse mask clustering. a fusion device to obtain a semantic fusion feature vector of enhanced user basic information; a nutritional requirement semantic understanding module 850, used to perform word-granularity-based semantic encoding on the nutritional requirement description text to obtain a sequence of word-granularity semantic encoding feature vectors of the nutritional requirement description text; a nutritional requirement semantic expression module 860 under basic information constraints, used to perform semantic constraint expression on the sequence of word-granularity semantic encoding feature vectors of the nutritional requirement description text based on the enhanced user basic information semantic fusion feature vector to obtain a semantic representation vector of the nutritional requirement description text under the semantic constraint of the user basic information; a recommended matching recipe generation module 870, used to generate a recommended matching recipe based on the semantic representation vector of the nutritional requirement description text under the semantic constraint of the user basic information.

[0081] Specifically, in the technical solution of the present application, first, the basic information data of the user is obtained, wherein the basic information data includes age, gender, weight, height, health status, allergies and nature of work, and the nutritional needs description text entered by the user is obtained. Next, considering that the basic information of the user (such as age, gender, weight, height, etc.) is usually in an unstructured form, such as text or numbers. Therefore, in order to be able to convert these user basic information data into a form that can be understood and processed by a computer, and to capture the semantic features in the user basic information data, in the technical solution of the present application, the basic information data of the user is semantically encoded to obtain a sequence of semantic feature vectors of the user basic information. Through the processing method of semantic encoding, these unstructured user basic information data can be converted into a structured vector form, and the semantic feature information of each user basic information item can be captured, which helps the system to understand the individual differences and needs of users more accurately.

[0082] Then, considering that each user basic information semantic feature vector in the sequence of the user basic information semantic feature vector contains semantic feature information about each user's basic information item, and these semantic feature information have correlation relationships and interactive influences, they jointly determine the subsequent recipe matching recommendation. In addition, these different semantic features of user basic information items have different importance and contribution to the subsequent recipe matching recommendation task, and there will also be some interference and noise information irrelevant to the subsequent. Based on this, in the technical solution of the present application, the sequence of the user basic information semantic feature vector is further passed through the user basic information key semantic enhancement fusion device based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector. Specifically, the user basic information key semantic enhancement fusion device based on feature distribution reverse mask clustering can analyze the sequence of input user basic information semantic feature vectors through a clustering algorithm to identify the inherent structure in the features of each basic information item and extract a representative cluster center vector. Next, the semantic correlation coefficient between each user basic information semantic feature vector and the cluster center is calculated to quantify the similarity between them. By taking the inverse of these correlation coefficients, the semantic anti-correlation coefficients of the user's basic information are obtained to emphasize the semantic features of the user's basic information items that are significantly different from the cluster center. These features may reveal noise or key information in the user's basic information items. Subsequently, the anti-correlation coefficients are normalized using the Sigmoid function to ensure that the weight coefficients are within a reasonable range, providing a balanced weight distribution for the subsequent weighting of the semantic features of the user's basic information items. The weighting step generates a sequence of suppressed user basic information semantic feature vectors by applying the normalized weight coefficients to the original user's basic information semantic feature vectors. This process realizes the reverse screening of feature components to filter out feature components that are more likely to be interference components or noise distributions in the feature distribution. The feature components that are interference components or noise components in the semantic feature distribution of the user's basic information items are filtered out by calculating the difference between the user's basic information semantic feature vector and the suppressed user's basic information semantic feature vector. Finally, the enhanced user basic information semantic fusion feature vector is obtained by calculating the mean vector of the enhanced user basic information semantic feature vector sequence. This operation integrates all enhanced user basic information item semantic features to form a key feature vector that can represent the entire user basic information data set. This fused feature vector not only improves the discriminability of the semantic features of basic information of different users, but also provides a richer and more discriminative feature representation for the deep learning model, thereby enhancing the generalization ability of the model and the performance of the subsequent recipe generation system.

[0083] Accordingly, if Figure 3As shown, the user basic information semantic fusion module 840 includes: a mean vector calculation unit 841, which is used to calculate the mean vector of the sequence of the user basic information semantic feature vectors to obtain the user basic information semantic cluster center feature vector; a semantic correlation coefficient calculation unit 842, which is used to calculate the semantic correlation coefficient between each user basic information semantic feature vector in the sequence of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector to obtain a sequence of user basic information cluster center semantic correlation coefficients; a semantic anti-correlation unit 843, which is used to take the inverse of each user basic information cluster center semantic correlation coefficient in the sequence of the user basic information cluster center semantic correlation coefficient to obtain a sequence of user basic information cluster center semantic anti-correlation coefficients; a semantic anti-correlation coefficient normalization unit 844, which is used to use the Sigmoid function to normalize the sequence of the user basic information cluster center semantic anti-correlation coefficients To obtain a sequence of semantic anti-correlation weight coefficients of user basic information cluster centers; a weighting unit 845, used to use each semantic anti-correlation weight coefficient of the user basic information cluster center in the sequence of semantic anti-correlation weight coefficients of the user basic information cluster center as a weight, to weight each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors to obtain a sequence of suppressed user basic information semantic feature vectors; a position difference unit 846, used to calculate the position difference between the sequence of user basic information semantic feature vectors and each group of corresponding user basic information semantic feature vectors and suppressed user basic information semantic feature vectors in the sequence of suppressed user basic information semantic feature vectors to obtain a sequence of enhanced user basic information semantic feature vectors; a position mean unit 847, used to calculate the position mean vector of the sequence of enhanced user basic information semantic feature vectors to obtain the enhanced user basic information semantic fusion feature vector.

[0084] Among them, the semantic correlation coefficient calculation unit 842 is used to: respectively calculate the norm of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector to obtain the user basic information semantic coding norm representation value and the user basic information semantic cluster center norm representation value; calculate the product between the user basic information semantic coding norm representation value and the user basic information semantic cluster center norm representation value to obtain the user basic information semantic-cluster center semantic correlation norm representation value; calculate the multiplication between the transposed vector of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector to obtain the user basic information semantic-cluster center semantic correlation feature value; calculate the division between the user basic information semantic-cluster center semantic correlation feature value and the user basic information semantic-cluster center semantic correlation norm representation value to obtain the user basic information cluster center semantic correlation coefficient.

[0085] In a specific example, the user basic information semantic fusion module 840 is used to: process the sequence of the user basic information semantic feature vectors through the user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering with the following key semantic enhancement fusion formula to obtain the enhanced user basic information semantic fusion feature vector; wherein the key semantic enhancement fusion formula is:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] ;

[0092] ;

[0093] ;

[0094] in, represents a sequence of vectors, is a sequence of semantic feature vectors of the user's basic information, , , , are the first, second, and third in the sequence of the user basic information semantic feature vector and A user's basic information semantic feature vector, is the number of vectors in the sequence of the user basic information semantic feature vectors, is the feature vector of the semantic clustering center of the user’s basic information, is the one-norm of the vector, For the said The semantic correlation coefficient between the semantic feature vector of the user's basic information and the semantic cluster center feature vector of the user's basic information, For the said The semantic anti-correlation coefficient corresponding to the semantic feature vector of the user's basic information is For the said The semantic anti-correlation weight coefficient corresponding to the semantic feature vector of the user's basic information, For the first, second, and The suppressed user basic information semantic feature vector corresponding to the user basic information semantic feature vector is To suppress the sequence of semantic feature vectors of user basic information, The first, second, and third in the sequence of the enhanced user basic information semantic feature vectors and A semantic feature vector that enhances the user's basic information. To strengthen the sequence of semantic feature vectors of user basic information, It is the semantic fusion feature vector of the enhanced user basic information.

[0095] Furthermore, the nutritional requirement description text is segmented and then encoded in a semantic encoder including a word embedding layer to extract semantic feature information based on word granularity in the nutritional requirement description, thereby obtaining a sequence of word granularity semantic encoding feature vectors of the nutritional requirement description text.

[0096] Correspondingly, the nutritional requirement semantic understanding module 850 is used to: perform word segmentation processing on the nutritional requirement description text and then pass it through a semantic encoder including a word embedding layer to obtain a sequence of word granularity semantic encoding feature vectors of the nutritional requirement description text.

[0097] It should be understood that when actually matching and recommending personalized recipes for users, it is necessary not only to pay attention to the user's demand information, but also to combine their personal basic information, such as age, gender, weight, height, health status, allergies and nature of work to generate personalized recipes. Based on this, in the technical solution of the present application, the semantic fusion feature vector of the enhanced user's basic information is further used as a constraint vector, and the sequence of the constraint vector and the word granularity semantic encoding feature vector of the nutritional demand description text is passed through a global word granularity semantic aggregation module of the nutritional demand description text based on the expression of unidirectional attention constraints to obtain the semantic representation vector of the nutritional demand description text under the semantic constraints of the user's basic information. Through the processing of the global word granularity semantic aggregation module of the nutritional demand description text based on the expression of unidirectional attention constraints, the semantics of the nutritional demand description text can be constrained and optimized according to the personalized basic information of the user. This personalized constraint can ensure that the recommended matching recipes generated subsequently are more in line with the individual differences and needs of the user. Moreover, through the global word-granularity semantic aggregation processing of the nutritional requirement description text expressed by unidirectional attention constraints, the system can comprehensively consider the semantic relevance and mutual influence between the user's basic information and the nutritional requirement description text, which helps the system to better generate personalized recipe recommendations for users and improve the accuracy and practicality of the recommendations.

[0098] Correspondingly, the module 860 for expressing the semantics of nutritional needs under the constraints of basic information is used to: use the semantic fusion feature vector of the enhanced user basic information as the constraint vector, and pass the sequence of the constraint vector and the word granularity semantic encoding feature vector of the nutritional needs description text through the global word granularity semantic aggregation module of the nutritional needs description text based on unidirectional attention constraint expression to obtain the semantic representation vector of the nutritional needs description text under the semantic constraints of the user basic information.

[0099] Specifically, Figure 4 As shown, the module 860 for expressing the semantics of nutritional needs under the constraint of basic information comprises: a semantic similarity calculation unit 861, which is used to calculate the semantic similarity between the constraint vector and each nutritional need description text word granularity semantic coding feature vector in the sequence of the nutritional need description text word granularity semantic coding feature vector to obtain a sequence of nutritional need description text word granularity one-way matching semantic measurement coefficients; a one-way matching semantic measurement coefficient normalization unit 862, which is used to normalize the sequence of nutritional need description text word granularity one-way matching semantic measurement coefficients to obtain a sequence of nutritional need description text word granularity one-way matching semantic measurement weight values; a weighted sum unit 863, which is used to calculate the weighted sum of the sequence of nutritional need description text word granularity semantic coding feature vectors using the sequence of nutritional need description text word granularity one-way matching semantic measurement weight values ​​as weights to obtain the nutritional need description text semantic representation vector under the semantic constraint of the user's basic information.

[0100] Among them, the semantic similarity calculation unit 861 is used to: cascade each nutrition requirement description text word granularity semantic coding feature vector in the sequence of the constraint vector and the nutrition requirement description text word granularity semantic coding feature vector to obtain a set of query nutrition requirement description text word granularity semantic coding joint feature vectors; calculate the matrix product of each query nutrition requirement description text word granularity semantic coding joint feature vector in the set of query nutrition requirement description text word granularity semantic coding joint feature vectors and the weight matrix, and then add the obtained feature vector and the bias vector by position to obtain a sequence of query nutrition requirement description text word granularity semantic coding joint bias feature vectors; input the sequence of query nutrition requirement description text word granularity semantic coding joint bias feature vectors into the sigmoid activation function to obtain a sequence of nutrition requirement description text word granularity one-way matching semantic measurement coefficients.

[0101] Among them, the one-way matching semantic measurement coefficient normalization unit 862 is used to: use each one-way matching semantic measurement coefficient of the nutritional need description text word granularity in the sequence of the one-way matching semantic measurement coefficient of the nutritional need description text word granularity as the exponent of the natural constant to calculate the exponential function value with the natural constant as the base to obtain a sequence of logarithmic values ​​of the one-way matching semantic measurement coefficient of the nutritional need description text word granularity; calculate the sum of the logarithmic values ​​of each one-way matching semantic measurement coefficient of the nutritional need description text word granularity in the sequence of the one-way matching semantic measurement coefficient of the nutritional need description text word granularity to obtain the total sum of the one-way matching semantic measurement of the nutritional need description text word granularity; divide each logarithmic value of the one-way matching semantic measurement coefficient of the nutritional need description text word granularity in the sequence of the one-way matching semantic measurement coefficient of the nutritional need description text word granularity by the total sum of the one-way matching semantic measurement of the nutritional need description text word granularity to obtain a sequence of one-way matching semantic measurement weight values ​​of the nutritional need description text word granularity.

[0102] In a specific example, the basic information constraint-based nutrition requirement semantic expression module 860 is used to: use the enhanced user basic information semantic fusion feature vector as the constraint vector, process the constraint vector and the sequence of the nutrition requirement description text word granularity semantic encoding feature vector through the nutrition requirement description text global word granularity semantic aggregation module based on unidirectional attention constraint expression with the following nutrition requirement semantic aggregation formula to obtain the nutrition requirement description text semantic representation vector under the user basic information semantic constraint; wherein, the nutrition requirement semantic aggregation formula is:

[0103] ;

[0104] ;

[0105] ;

[0106] in, is the constraint vector, The first word in the sequence of the granular semantic encoding feature vector of the nutritional requirement description text The word-granular semantic encoding feature vector of the nutritional requirement description text, It is cascade processing. The first word in the sequence of one-way matching semantic metric coefficients of the nutritional requirement description text word granularity The semantic metric coefficient of the one-way matching of the nutritional requirement description text word granularity is is the weight matrix, is the bias vector, is the sigmoid function, Indicated by natural constant The exponential function with base , is the number of feature vectors in the sequence of the feature vectors of the word-granular semantic encoding of the nutritional requirement description text, It is the first in the sequence of the semantic metric weight value of the granularity one-way matching of the nutritional requirement description text. The weight value of the semantic metric of the one-way matching of the nutritional requirement description text word granularity, It is the semantic representation vector of the nutritional demand description text under the semantic constraints of the user's basic information.

[0107] Furthermore, the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information is passed through a decoder-based recipe generator to obtain a recommended recipe. In other words, the semantic features of the nutritional needs under the constraints of the user's basic information are used to generate recipes. In this way, the recommended user nutrition combination can be automatically generated according to the user's basic situation and nutritional needs. Specifically, daily dietary recommendations can be provided, including recommended food types and intakes, and personalized recipes can be generated, so as to generate personalized recipes in a more intelligent way. In addition, users can also preview the food production process and the final product through the system's 3D technology, which not only increases interactivity and fun, but also ensures that users make and consume food according to the food production process, ensuring that the food calorie intake is within a controllable range.

[0108] Correspondingly, the recommended recipe generation module 870 is used to: pass the semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information through a decoder-based recipe generator to obtain a recommended recipe.

[0109] Preferably, the present application takes into account that the sequence of the enhanced user basic information semantic fusion feature vector and the nutritional needs description text word granularity semantic encoding feature vector respectively express the encoded semantic features of the clustering key semantics enhancement of the user's basic information data and the nutritional needs description text and the encoded semantic features based on word granularity embedding. In this way, when the sequence of the enhanced user basic information semantic fusion feature vector and the nutritional needs description text word granularity semantic encoding feature vector is passed through the nutritional needs description text global word granularity semantic aggregation module based on unidirectional attention constraint expression, the obtained semantic representation vector of the nutritional needs description text under the semantic constraint of the user basic information will also cause outlier decoding regression inference mapping deviation due to the difference in constraint coverage attention weights, thereby affecting the accuracy of the recommended matching recipes obtained by the semantic representation vector of the nutritional needs description text under the semantic constraint of the user basic information through the decoder-based recipe generator.

[0110] Therefore, in a preferred example, the recommended combination recipe generation module 870 includes: an optimization unit, used to optimize the text semantic representation vector of the nutritional needs description under the semantic constraint of the user's basic information to obtain an optimized text semantic representation vector of the nutritional needs description under the semantic constraint of the user's basic information; a decoding unit, used to pass the optimized text semantic representation vector of the nutritional needs description under the semantic constraint of the user's basic information through the decoder-based recipe generator to obtain the recommended combination recipe.

[0111] Among them, the optimization unit is used to: calculate the feature mean of the semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information, and divide the feature mean by the difference between the maximum eigenvalue and the minimum eigenvalue of the semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information to obtain the semantic distribution representation value of the nutritional needs description text under the semantic constraints of the user's basic information; subtract the semantic distribution representation value of the nutritional needs description text under the semantic constraints of the user's basic information and divide it by the semantic distribution representation value of the nutritional needs description text under the semantic constraints of the user's basic information to obtain the semantic distribution modulation value of the nutritional needs description text under the semantic constraints of the user's basic information; activate the semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information through a probabilistic function to obtain a probabilistic semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information; perform dot-wise subtraction on the probabilistic semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information and the semantic distribution modulation value of the nutritional needs description text under the semantic constraints of the user's basic information , take the absolute value and calculate the negative of the logarithmic value with base 2 to obtain the semantic distribution modulation information vector of the nutrition needs description text under the semantic constraints of the user's basic information; divide the semantic distribution representation value of the nutrition needs description text under the semantic constraints of the user's basic information by one minus the difference of each eigenvalue of the semantic representation vector of the nutrition needs description text under the semantic constraints of the user's basic information, sum all eigenvalues ​​of the semantic representation vector of the nutrition needs description text under the semantic constraints of the user's basic information, and divide it by the length of the semantic representation vector of the nutrition needs description text under the semantic constraints of the user's basic information to obtain the semantic distribution modulation bias value of the nutrition needs description text under the semantic constraints of the user's basic information; perform point addition on the product of the semantic distribution modulation information vector of the nutrition needs description text under the semantic constraints of the user's basic information and the semantic distribution modulation bias value of the nutrition needs description text under the semantic constraints of the user's basic information and the weight as a hyperparameter to obtain the optimized semantic representation vector of the nutrition needs description text under the semantic constraints of the user's basic information.

[0112] In a specific example, the optimization unit is used to optimize the semantic representation vector of the nutritional requirement description text under the semantic constraint of the user's basic information by using the following optimization formula to obtain the optimized semantic representation vector of the nutritional requirement description text under the semantic constraint of the user's basic information; wherein the optimization formula is expressed as:

[0113] ;

[0114] And among them:

[0115] ;

[0116] in, is the semantic representation vector of the nutritional requirement description text under the probabilistic semantic constraint of the user's basic information, is the feature mean of the semantic representation vector of the nutritional demand description text under the semantic constraint of the user's basic information, and They respectively represent the maximum eigenvalue and the minimum eigenvalue of the semantic representation vector of the nutritional demand description text under the semantic constraint of the user's basic information, It represents the semantic distribution representation value of the nutritional demand description text under the semantic constraints of the user's basic information. It means to subtract by position point. It means summing by position. is the length of the semantic representation vector of the nutritional needs description text under the semantic constraints of the user's basic information, is the feature value of each position in the semantic representation vector of the nutritional demand description text under the probabilistic semantic constraint of the user's basic information, is the weight hyperparameter, It is the semantic representation vector of the nutritional demand description text under the optimized semantic constraints of the user's basic information.

[0117] That is, the probability information distribution planning of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information is carried out based on the eigenvalue through the Bernoulli probability modulation distribution of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information relative to the eigenvalue distribution, and the probability reverse mapping of the probability characteristics of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information is used as the extended coverage of the set mapping space of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information, so as to independently understand the interactive path between the intuitive probability information distribution and the abstract probability space mapping of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information, so as to improve the accuracy of the recommended collocation recipes obtained by the recipe generator based on the decoder by avoiding the counterfactual reasoning mapping of the outlier feature distribution of the semantic representation vector of the description text of the nutritional needs under the semantic constraints of the user's basic information. In this way, personalized collocation recipes can be generated more accurately according to the basic situation and nutritional needs of the user, so as to improve the intelligent level of the nutritional collocation management system.

[0118] Based on the above embodiments, see Figure 5 As shown in FIG. 1 , a flowchart of a personalized nutrition matching management method based on AI generation in an embodiment of the present application is shown. For example, the personalized nutrition matching management method based on AI generation can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 5 As shown, according to the AI-generated personalized nutrition matching management method of the embodiment of the present application, the steps include: S510, obtaining basic information data of the user, wherein the basic information data includes age, gender, weight, height, health status, allergies and nature of work; S520, obtaining a nutritional requirement description text input by the user; S530, semantically encoding the basic information data of the user to obtain a sequence of semantic feature vectors of the user's basic information; S540, encoding the sequence of semantic feature vectors of the user's basic information through a key semantic feature vector of the user's basic information based on feature distribution reverse mask clustering A fusion device is used to obtain a semantic fusion feature vector of enhanced user basic information; S550, the nutrition requirement description text is semantically encoded based on word granularity to obtain a sequence of word granularity semantic encoding feature vectors of the nutrition requirement description text; S560, based on the enhanced user basic information semantic fusion feature vector, a semantic constraint expression is performed on the sequence of word granularity semantic encoding feature vectors of the nutrition requirement description text to obtain a semantic representation vector of the nutrition requirement description text under the semantic constraint of the user basic information; S570, based on the semantic representation vector of the nutrition requirement description text under the semantic constraint of the user basic information, a recommended matching recipe is generated.

[0119] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned personalized nutrition matching management method based on AI generation have been referred to above. Figures 2 to 4 It has been introduced in detail in the description of the AI-generated personalized nutrition pairing management system 800, and therefore, its repeated description will be omitted.

[0120] Figure 6 FIG. 1 is an application scenario diagram of a personalized nutrition matching management system generated based on AI according to an embodiment of the present application. Figure 6 As shown, in this application scenario, first, the user's basic information data (for example, Figure 6 D1 as shown in the figure), the nutritional requirement description text entered by the user (for example, Figure 6 D2 as shown in the figure), wherein the basic information data includes age, gender, weight, height, health status, allergies and nature of work, and then the basic information data of the user and the nutritional needs description text are input into a server deployed with a personalized nutrition matching management algorithm generated based on AI (for example, Figure 6 In S) as shown in , the server is capable of using the personalized nutrition matching management algorithm generated based on AI to process the basic information data of the user and the nutrition requirement description text to generate a recommended matching recipe.

[0121] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the personalized nutrition matching management method based on AI generation in the above embodiments when executing the program.

[0122] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0123] The embodiments of the present application also provide a computer-readable storage medium having computer executable instructions stored thereon. When the computer executable instructions are executed by a processor, the personalized nutrition matching management method based on AI generation according to the embodiments of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0124] The embodiments of the present application also provide a computer program product or a computer program, which includes computer executable instructions, which are stored in a computer-readable storage medium. The processor of the computer device reads the computer executable instructions from the computer-readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the personalized nutrition matching management method based on AI generation according to the embodiments of the present application.

[0125] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various modifications and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0126] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0127] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0128] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0129] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A personalized nutrition matching management system based on AI generation, characterized in that: include: A user basic information collection module is used to obtain the user's basic information data, wherein the basic information data includes age, gender, weight, height, health status, allergies and job nature; A user nutritional requirement description acquisition module, used to acquire a nutritional requirement description text input by the user; A user basic information semantic encoding module, used to perform semantic encoding on the user's basic information data to obtain a sequence of user basic information semantic feature vectors; A user basic information semantic fusion module, used for passing the sequence of user basic information semantic feature vectors through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector; A nutrition requirement semantic understanding module, used for performing word granularity-based semantic coding on the nutrition requirement description text to obtain a sequence of word granularity semantic coding feature vectors of the nutrition requirement description text; A module for expressing the semantics of nutritional needs under the constraints of basic information, which is used to express the sequence of the semantic encoding feature vectors of the nutritional needs description text word granularity based on the semantic fusion feature vector of the enhanced user basic information to obtain a semantic representation vector of the nutritional needs description text under the semantic constraints of the user basic information; A recommended recipe generation module is used to generate a recommended recipe based on the semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information; The user basic information semantic fusion module includes: A mean vector calculation unit, used to calculate the mean vector of the sequence of the user basic information semantic feature vectors to obtain the user basic information semantic cluster center feature vector; A semantic correlation coefficient calculation unit, used to calculate the semantic correlation coefficient between each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors and the user basic information semantic cluster center feature vector to obtain a sequence of user basic information cluster center semantic correlation coefficients; A semantic anti-correlation unit, used for taking the inverse of the semantic correlation coefficient of each user basic information clustering center in the sequence of semantic correlation coefficients of the user basic information clustering center to obtain a sequence of semantic anti-correlation coefficients of the user basic information clustering center; A semantic anti-correlation coefficient normalization unit, used to normalize the sequence of semantic anti-correlation coefficients of the user basic information clustering center using a Sigmoid function to obtain a sequence of semantic anti-correlation weight coefficients of the user basic information clustering center; A weighting unit, configured to weight each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors by using each user basic information cluster center semantic anti-correlation weight coefficient in the sequence of user basic information cluster center semantic anti-correlation weight coefficients as a weight to obtain a sequence of suppressed user basic information semantic feature vectors; A position difference unit, used for calculating the position difference between each group of corresponding user basic information semantic feature vectors and suppressed user basic information semantic feature vectors in the sequence of user basic information semantic feature vectors and the sequence of suppressed user basic information semantic feature vectors to obtain a sequence of enhanced user basic information semantic feature vectors; A position-based mean unit, used for calculating the position-based mean vector of the sequence of the enhanced user basic information semantic feature vectors to obtain the enhanced user basic information semantic fusion feature vector; The module for expressing the semantics of nutritional requirements under the constraints of basic information is used to: Taking the enhanced user basic information semantic fusion feature vector as a constraint vector, the constraint vector and the sequence of the word granularity semantic encoding feature vector of the nutritional requirement description text are passed through a nutritional requirement description text global word granularity semantic aggregation module based on unidirectional attention constraint expression to obtain a semantic representation vector of the nutritional requirement description text under the semantic constraint of the user basic information; The module for expressing the semantics of nutritional requirements under the constraints of basic information includes: A semantic similarity calculation unit, used for calculating the semantic similarity between the constraint vector and each nutrition requirement description text word granularity semantic encoding feature vector in the sequence of nutrition requirement description text word granularity semantic encoding feature vectors to obtain a sequence of nutrition requirement description text word granularity one-way matching semantic measurement coefficients; A one-way matching semantic measurement coefficient normalization unit is used to normalize the sequence of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity to obtain a sequence of one-way matching semantic measurement weight values ​​of the nutritional requirement description text word granularity; A weighted sum unit, used for calculating the weighted sum of the sequence of semantic encoding feature vectors of the nutritional requirement description text word granularity using the sequence of one-way matching semantic measurement weight values ​​of the nutritional requirement description text word granularity as weights to obtain a semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information; Wherein, the semantic similarity calculation unit is used to: Cascading each of the constraint vectors and the sequence of the granular semantic encoding feature vectors of the nutrient requirement description text to obtain a set of query nutrient requirement description text granular semantic encoding joint feature vectors; After respectively calculating the matrix product of each query nutritional requirement description text word granularity semantic coding joint feature vector in the set of the query nutritional requirement description text word granularity semantic coding joint feature vector and the weight matrix, the obtained feature vector and the bias vector are added by position to obtain a sequence of query nutritional requirement description text word granularity semantic coding joint bias feature vectors; Inputting the sequence of the semantic encoding of the query nutritional requirement description text word granularity combined with the bias feature vector into the sigmoid activation function to obtain the sequence of the one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity; The one-way matching semantic metric coefficient normalization unit is used to: Using each one-way matching semantic measurement coefficient of the nutritional requirement description text word granularity in the sequence of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity as an exponent of a natural constant to calculate an exponential function value with the natural constant as the base to obtain a sequence of logarithmic values ​​of one-way matching semantic measurement coefficients of the nutritional requirement description text word granularity; Calculate the sum of the logarithmic values ​​of the one-way matching semantic metric coefficients of the nutritional requirement description text word granularity in the sequence of the logarithmic value of the one-way matching semantic metric coefficients of the nutritional requirement description text word granularity to obtain the total sum of the one-way matching semantic metric values ​​of the nutritional requirement description text word granularity; Each logarithm of the one-way matching semantic metric coefficient of the nutritional need description text word granularity in the sequence of logarithm of the one-way matching semantic metric coefficient of the nutritional need description text word granularity is divided by the total sum of the one-way matching semantic metric of the nutritional need description text word granularity to obtain a sequence of the one-way matching semantic metric weight values ​​of the nutritional need description text word granularity.

2. The personalized nutrition matching management system based on AI generation according to claim 1, characterized in that: The semantic correlation coefficient calculation unit is used to: Calculating the norms of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector respectively to obtain a user basic information semantic coding norm representation value and a user basic information semantic cluster center norm representation value; Calculate the product of the user basic information semantic code-norm representation value and the user basic information semantic cluster center-norm representation value to obtain the user basic information semantic-cluster center semantic correlation-norm representation value; Calculate the multiplication between the transposed vector of the user basic information semantic feature vector and the user basic information semantic cluster center feature vector to obtain the user basic information semantic-cluster center semantic correlation feature value; The division between the user basic information semantics-cluster center semantic relevance feature value and the user basic information semantics-cluster center semantic relevance-norm representation value is calculated to obtain the user basic information cluster center semantic relevance coefficient.

3. The personalized nutrition matching management system based on AI generation according to claim 2, characterized in that: The nutritional requirement semantic understanding module is used to: The nutritional requirement description text is word segmented and then passed through a semantic encoder including a word embedding layer to obtain a sequence of word granularity semantic encoding feature vectors of the nutritional requirement description text.

4. The personalized nutrition matching management system based on AI generation according to claim 3 is characterized in that: The recommended recipe generation module is used to: The semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information is passed through a decoder-based recipe generator to obtain a recommended recipe.

5. A personalized nutrition matching management method based on AI generation, using the personalized nutrition matching management system based on AI generation according to claim 1, characterized in that: include: Obtaining basic information data of the user, wherein the basic information data includes age, gender, weight, height, health status, allergies and job nature; Acquire the nutritional requirement description text input by the user; Performing semantic encoding on the basic information data of the user to obtain a sequence of semantic feature vectors of the basic information of the user; The sequence of user basic information semantic feature vectors is passed through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector; Performing word-granularity-based semantic coding on the nutrition requirement description text to obtain a sequence of word-granularity semantic coding feature vectors of the nutrition requirement description text; Based on the enhanced user basic information semantic fusion feature vector, a sequence of word granularity semantic encoding feature vectors of the nutritional demand description text is semantically constrained to obtain a semantic representation vector of the nutritional demand description text under the semantic constraint of the user basic information; Generate a recommended recipe based on the semantic representation vector of the nutritional requirement description text under the semantic constraints of the user's basic information; The sequence of the user basic information semantic feature vectors is passed through a user basic information key semantic enhancement fuser based on feature distribution reverse mask clustering to obtain an enhanced user basic information semantic fusion feature vector, including: Calculating the mean vector of the sequence of the user basic information semantic feature vectors to obtain the user basic information semantic cluster center feature vector; Calculating the semantic correlation coefficient between each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors and the user basic information semantic cluster center feature vector to obtain a sequence of user basic information cluster center semantic correlation coefficients; Taking the reciprocal of each user basic information cluster center semantic correlation coefficient in the sequence of user basic information cluster center semantic correlation coefficients to obtain a sequence of user basic information cluster center semantic anti-correlation coefficients; Using a Sigmoid function to normalize the sequence of semantic anti-correlation coefficients of the user basic information clustering center to obtain a sequence of semantic anti-correlation weight coefficients of the user basic information clustering center; Using each user basic information cluster center semantic anti-correlation weight coefficient in the sequence of user basic information cluster center semantic anti-correlation weight coefficients as a weight, weighting each user basic information semantic feature vector in the sequence of user basic information semantic feature vectors to obtain a sequence of suppressed user basic information semantic feature vectors; Calculating the positional difference between each group of corresponding user basic information semantic feature vectors and suppressed user basic information semantic feature vectors in the sequence of user basic information semantic feature vectors and the sequence of suppressed user basic information semantic feature vectors to obtain a sequence of enhanced user basic information semantic feature vectors; The position-wise mean vector of the sequence of the enhanced user basic information semantic feature vectors is calculated to obtain the enhanced user basic information semantic fusion feature vector.