Essence proportion customization method and device, storage medium and computer equipment

Abstract

The invention provides an essence proportion customization method and device, a storage medium and computer equipment. The method comprises the following steps: collecting taste related data of a target user; extracting taste preference characteristics in the taste related data, and parameterizing the taste preference characteristics to obtain a flavor parameter vector; constructing an essence proportioning model based on a neural network; and analyzing the flavor parameter vector by using the essence proportion model, and outputting a target essence proportion conforming to the taste of the target user. According to the method and the system, the taste preference of the user is digitally modeled, the neural network model is designed, and the flavor preference of the user is converted into a specific essence proportion, so that the beverage conforming to the preference of the user is customized for the user.

Description

Technical Field

[0001] The present invention relates to the field of intelligent cooking technology, and in particular to a flavor ratio customization method and device, a storage medium, and a computer device. Background Art

[0002] For existing prepared drinks, a drink generally uses a uniform flavor ratio, and users can only choose the sweetness of the drink when placing an order, such as "normal sugar, 70% sugar, 50% sugar, 30% sugar, sugar-free", etc. The sweetness cannot be customized in detail, nor can other taste dimensions such as acidity be adjusted. It is difficult to customize drinks that suit the user's taste based on their own needs. Summary of the invention

[0003] To this end, the present invention provides a flavor ratio customization method and device, storage medium, and computer equipment, aiming to solve the technical problem in the prior art that it is difficult to customize beverages that suit the user's taste according to the user's own needs.

[0004] To achieve the above objectives, the present invention adopts the following technical solutions:

[0005] According to a first aspect of the present invention, the present invention provides a method for customizing a flavor ratio, the method comprising:

[0006] Collect data about target users’ tastes;

[0007] Extracting taste preference features from the taste-related data, and parameterizing the taste preference features to obtain a flavor parameter vector;

[0008] Construct a flavor ratio model based on neural network;

[0009] The flavor parameter vector is analyzed using the flavor ratio model to output a target flavor ratio that meets the taste of the target user.

[0010] Optionally, the taste-related data includes at least one of historical order data, questionnaire data and self-selected data.

[0011] Optionally, parameterizing the taste preference characteristics to obtain a flavor parameter vector includes:

[0012] Defining basic taste dimensions of a flavor wheel; the basic taste dimensions include at least one of sour, sweet, bitter, spicy, salty, and umami;

[0013] Based on the basic taste dimension, the taste preference characteristics are parameterized into a flavor parameter vector, which is expressed as follows:

[0014] U=[u 1 ,u 2 ,...,u n]

[0015] Wherein, U is the flavor parameter vector of the target user; u i is the target user's preference for the i-th basic taste dimension, and its value range is [0, 1]; n is the number of basic taste dimensions.

[0016] Optionally, the construction of a flavor ratio model based on a neural network includes:

[0017] Construct a neural network model including an input layer, at least one hidden layer, an activation function, and an output layer;

[0018] The input layer is used to receive the flavor parameter vector of the target user; the hidden layer is used to learn the nonlinear relationship between flavor preference and flavor ratio; the activation function is used to realize the nonlinear transformation of data; the output layer is used to output the flavor ratio vector;

[0019] The neural network model is trained to obtain the trained flavor ratio model.

[0020] Optionally, the method further comprises:

[0021] The total amount of the flavor ratio vector is constrained to ensure that the total flavor ratio meets the actual production requirements. The formula is as follows:

[0022] R=[r 1 ,r 2 ,...,r m ]

[0023]

[0024] Wherein, R is the flavor ratio vector; r j is the jth flavor ratio vector, r j ≥0; m is the number of flavor types.

[0025] Optionally, the training of the neural network model to obtain the trained flavor ratio model includes:

[0026] Construct a training sample set containing multiple flavor parameter vectors and corresponding flavor ratio vectors. The formula is as follows:

[0027] (U (k) ,R (k) )k=1,2,...,N

[0028] Where N is the number of training samples;

[0029] The neural network model is iteratively trained using the training sample set, and the loss function is minimized using an optimization algorithm to obtain the flavor ratio model.

[0030] Optionally, the loss function is a MSE mean square error loss function, which is expressed as follows:

[0031]

[0032] Among them, L is the loss function value, r j (k) is the actual j-th flavor ratio in the k-th sample; The j-th flavor ratio in the k-th sample predicted by the neural network model.

[0033] According to a second aspect of the present invention, the present invention provides a flavor ratio customization device, the device comprising:

[0034] A data collection module, used to collect taste-related data of target users;

[0035] A feature extraction module, used to extract taste preference features from the taste-related data, and parameterize the taste preference features to obtain a flavor parameter vector;

[0036] Model building module, used to build flavor ratio model based on neural network;

[0037] The ratio analysis module is used to analyze the flavor parameter vector using the flavor ratio model and output a target flavor ratio that meets the taste of the target user.

[0038] According to the third aspect of the present invention, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the flavor ratio customization method as described in any one of the first aspect of the present invention.

[0039] According to the fourth aspect of the present invention, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for customizing the proportion of an essence as described in any one of the first aspect of the present invention is implemented.

[0040] The present invention adopts the above technical solution and has at least the following beneficial effects:

[0041] Through the scheme of the present invention, the taste-related data of the target user are collected; the taste preference features in the taste-related data are extracted, and the taste preference features are parameterized to obtain a flavor parameter vector; a flavor matching model based on a neural network is constructed; the flavor parameter vector is analyzed using the flavor matching model, and a target flavor matching that meets the taste of the target user is output. Thus, by digitally modeling the user's taste preferences, a neural network model is designed to convert the user's flavor preferences into specific flavor matching, thereby customizing a beverage that meets the user's preferences. It should be understood that the general description above and the detailed description below are merely exemplary and explanatory and do not limit the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0043] Figure 1 A schematic diagram showing a flow chart of a method for customizing a flavor ratio provided by an embodiment of the present invention;

[0044] Figure 2 A schematic diagram showing the structure of a flavor ratio customization device provided by an embodiment of the present invention is shown;

[0045] Figure 3 A schematic diagram of the physical structure of a computer device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0046] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0047] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0048] The embodiment of the present invention provides a method for customizing the proportion of flavors, such as Figure 1 As shown, at least the following steps S101 to S104 may be included:

[0049] Step S101, collecting taste-related data of target users.

[0050] In the embodiment of the present invention, the taste-related data may include historical order data, questionnaire survey data, and self-selection data, etc. The historical order data may include information such as the types, flavors, and flavor ratios of beverages purchased by the user in the past; the questionnaire survey data may be information about the user's preferences for different tastes obtained through questionnaires; and the self-selection data may include recording the user's choices and adjustments when customizing a beverage.

[0051] It should be noted that in the process of collecting taste-related data, it is necessary to ensure that the collected user data is accurate and representative, and it is also necessary to pay attention to protecting the user's personal privacy information.

[0052] Step S102, extracting taste preference features from the taste-related data, and parameterizing the taste preference features to obtain a flavor parameter vector.

[0053] Based on the taste-related data of the target user, features related to taste signals, such as sweetness preference, sourness preference, etc., are extracted as taste preference features. In order to make the taste preference features suitable for model input, the embodiment of the present invention converts the taste preference features into flavor parameter vectors.

[0054] Specifically, the basic taste dimensions of the flavor wheel are defined; the basic taste dimensions include sour, sweet, bitter, peppery, salty, umami, etc. Based on the basic taste dimensions, the taste preference characteristics are parameterized into a flavor parameter vector, which is expressed as follows:

[0055] U=[u1 ,u 2 ,...,u n ]

[0056] Among them, U is the flavor parameter vector of the target user; u i is the target user’s preference for the i-th basic taste dimension, and its value range is [0, 1]; n is the number of basic taste dimensions.

[0057] Step S103, constructing a flavor ratio model based on a neural network.

[0058] Specifically, a neural network model including an input layer, at least one hidden layer, an activation function and an output layer can be constructed first; and then the neural network model can be trained to obtain a trained flavor ratio model.

[0059] Among them, the input layer is used to receive the flavor parameter vector U of the target user; one or more hidden layers can be set to learn the nonlinear relationship between flavor preference and flavor ratio; the activation function is used to realize the nonlinear transformation of data, and ReLU, Sigmoid, etc. can be selected; the output layer is used to output the flavor ratio vector R, and the formula is as follows:

[0060] R=[r 1 ,r 2 ,...,r m ]

[0061] Among them, R is the flavor ratio vector; r j is the jth flavor ratio vector, r j ≥0; m is the number of flavor types.

[0062] Furthermore, when training the neural network model, a training sample set including multiple flavor parameter vectors and corresponding flavor ratio vectors can be constructed first, and then the neural network model can be iteratively trained using the training sample set, and the loss function can be minimized using an optimization algorithm to obtain a flavor ratio model.

[0063] The training sample set can be expressed as:

[0064] (U (k) ,R (k) )k=1,2,...,N

[0065] Among them, N is the number of training samples.

[0066] In an optional embodiment, the loss function may be MSE mean square error, which is expressed as follows:

[0067]

[0068] Among them, L is the loss function value, rj (k) is the actual j-th flavor ratio in the k-th sample; The jth flavor ratio in the kth sample predicted by the neural network model.

[0069] Furthermore, the embodiment of the present invention can use stochastic gradient descent (SGD) and Adam optimization algorithm to minimize the loss function L, and continuously adjust the network weights through back propagation until the loss function converges.

[0070] It should also be noted that in order to meet the actual production requirements of the flavor ratio, the embodiment of the present invention also needs to standardize the flavor ratio, and the formula is as follows:

[0071]

[0072] That is, under the non-negative constraint r j ≥0, while ensuring that the total ratio of various flavors is 1.

[0073] It should be noted that, when training the neural network model, the embodiment of the present invention also needs to prevent overfitting to ensure the accuracy of the model's prediction for new users.

[0074] Step S104, analyzing the flavor parameter vector using the flavor ratio model, and outputting a target flavor ratio that meets the taste of the target user.

[0075] After obtaining the trained flavor matching model, the flavor parameter vector of the user to be customized can be input into the flavor matching model, and the flavor matching vector predicted by the model can be output. The final flavor matching can be provided to the user as a recipe for customized beverages. The recipe can also be provided to the beverage maker after obtaining the user's authorization to customize beverages that suit the user's taste.

[0076] In actual applications of the present invention, users can scan the code through the user-side application to authorize login and place an order for drinks. They can choose existing products / custom recipes, or use the present invention to obtain AI recommended recipes, and prepare drinks based on the flavor ratio obtained based on the neural network model analysis, thereby enhancing user experience and drink satisfaction.

[0077] The embodiment of the present invention provides a method for customizing flavor ratios, which collects taste-related data of target users; extracts taste preference features from the taste-related data, and parameterizes the taste preference features to obtain a flavor parameter vector; constructs a flavor ratio model based on a neural network; and uses the flavor ratio model to analyze the flavor parameter vector and output a target flavor ratio that meets the taste of the target user. Through the present invention, the user's taste preferences are digitally modeled, a neural network model is designed, and the user's flavor preferences are converted into specific flavor ratios, so as to customize a drink that meets the user's preferences.

[0078] Further, as Figure 1 The specific implementation of the present invention provides a flavor ratio customization device, such as Figure 2 As shown, the device may include: a data collection module 210, a feature extraction module 220, a model construction module 230 and a ratio analysis module 240.

[0079] The data collection module 210 may be used to collect taste-related data of target users;

[0080] The feature extraction module 220 can be used to extract taste preference features from the taste-related data and parameterize the taste preference features to obtain a flavor parameter vector;

[0081] A model building module 230 can be used to build a flavor ratio model based on a neural network;

[0082] The ratio analysis module 240 can be used to analyze the flavor parameter vector using the flavor ratio model and output a target flavor ratio that meets the taste of the target user.

[0083] Optionally, the feature extraction module 220 may also be used to define basic taste dimensions of the flavor wheel; the basic taste dimensions include at least one of sour, sweet, bitter, spicy, salty, and umami;

[0084] Based on the basic taste dimensions, the taste preference characteristics are parameterized into a flavor parameter vector, which is expressed as follows:

[0085] U=[u 1 ,u 2 ,...,u n ]

[0086] Among them, U is the flavor parameter vector of the target user; u i is the target user’s preference for the i-th basic taste dimension, and its value range is [0, 1]; n is the number of basic taste dimensions.

[0087] Optionally, the model construction module 230 may also be used to construct a neural network model including an input layer, at least one hidden layer, an activation function, and an output layer;

[0088] Among them, the input layer is used to receive the flavor parameter vector of the target user; the hidden layer is used to learn the nonlinear relationship between flavor preference and flavor ratio; the activation function is used to realize the nonlinear transformation of data; the output layer is used to output the flavor ratio vector;

[0089] The neural network model is trained to obtain a trained flavor ratio model.

[0090] Optionally, the model building module 230 can also be used to build a training sample set including multiple flavor parameter vectors and corresponding flavor ratio vectors, which is expressed as follows:

[0091] (U (k) ,R (k) )k=1,2,...,N

[0092] Where N is the number of training samples;

[0093] The neural network model is iteratively trained using the training sample set, and the optimization algorithm is used to minimize the loss function to obtain the flavor ratio model.

[0094] Optionally, the ratio analysis module 240 can also be used to constrain the total amount of the flavor ratio vector to ensure that the total flavor ratio meets the actual production requirements. The formula is as follows:

[0095] R=[r 1 ,r 2 ,...,r m ]

[0096]

[0097] Among them, R is the flavor ratio vector; r j is the jth flavor ratio vector, r j ≥0; m is the number of flavor types.

[0098] It should be noted that for other corresponding descriptions of the functional modules involved in the flavor ratio customization device provided in the embodiment of the present invention, reference can be made to Figure 1 The corresponding description of the method shown will not be repeated here.

[0099] Based on the above Figure 1 The method shown, accordingly, an embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the flavor ratio customization method described in any of the above embodiments are implemented.

[0100] Based on the above Figure 1 The method shown and Figure 2 The embodiment of the device shown in the figure, the embodiment of the present invention also provides a physical structure diagram of a computer device, such as Figure 3 As shown, the computer device may include a communication bus, a processor, a memory and a communication interface, and may also include an input / output interface and a display device, wherein each functional unit may communicate with each other through the bus. The memory stores a computer program, and the processor is used to execute the program stored in the memory and execute the steps of the flavor ratio customization method described in the above embodiment.

[0101] Those skilled in the art can clearly understand that the specific working processes of the systems, devices, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and for the sake of brevity, they are not further described here.

[0102] In addition, the functional units in various embodiments of the present invention may be physically independent of each other, or two or more functional units may be integrated together, or all functional units may be integrated into one processing unit. The above integrated functional units may be implemented in the form of hardware, or in the form of software or firmware.

[0103] Those skilled in the art can understand that if the integrated functional unit is implemented in the form of software and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention can essentially or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, which includes a number of instructions to enable a computing device (such as a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention when running the instructions. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk and other media that can store program codes.

[0104] Alternatively, all or part of the steps of implementing the aforementioned method embodiments may be accomplished by hardware associated with program instructions (such as a computing device such as a personal computer, a server, or a network device), and the program instructions may be stored in a computer-readable storage medium. When the program instructions are executed by a processor of a computing device, the computing device executes all or part of the steps of the methods described in the embodiments of the present invention.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that within the spirit and principles of the present invention, the technical solutions described in the aforementioned embodiments can still be modified, or some or all of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate from the protection scope of the present invention.

Claims

1. A method for customizing flavor ratios, characterized in that: The method comprises: Collect data about target users’ tastes; Extracting taste preference features from the taste-related data, and parameterizing the taste preference features to obtain a flavor parameter vector; Construct a flavor ratio model based on neural network; The flavor parameter vector is analyzed using the flavor ratio model to output a target flavor ratio that meets the taste of the target user.

2. The method according to claim 1, characterized in that The taste-related data includes at least one of historical order data, questionnaire data, and self-selected data.

3. The method according to claim 1, characterized in that: The parameterizing the taste preference characteristics to obtain a flavor parameter vector includes: Defining basic taste dimensions of a flavor wheel; the basic taste dimensions include at least one of sour, sweet, bitter, spicy, salty, and umami; Based on the basic taste dimension, the taste preference characteristics are parameterized into a flavor parameter vector, which is expressed as follows: U=[u1,u2,...,u n ] Wherein, U is the flavor parameter vector of the target user; u i is the target user's preference for the i-th basic taste dimension, and its value range is [0, 1]; n is the number of basic taste dimensions.

4. The method according to claim 1, characterized in that The construction of the flavor ratio model based on the neural network includes: Construct a neural network model including an input layer, at least one hidden layer, an activation function, and an output layer; The input layer is used to receive the flavor parameter vector of the target user; the hidden layer is used to learn the nonlinear relationship between flavor preference and flavor ratio; the activation function is used to realize the nonlinear transformation of data; the output layer is used to output the flavor ratio vector; The neural network model is trained to obtain the trained flavor ratio model.

5. The method according to claim 4, characterized in that The method further comprises: The total amount of the flavor ratio vector is constrained to ensure that the total flavor ratio meets the actual production requirements. The formula is as follows: R＝[r1,r2,...,r m ] Wherein, R is the flavor ratio vector; r j is the jth flavor ratio vector, r j ≥0; m is the number of flavor types.

6. The method according to claim 4, characterized in that The training of the neural network model to obtain the trained flavor ratio model includes: Construct a training sample set containing multiple flavor parameter vectors and corresponding flavor ratio vectors. The formula is as follows: (U (k) ,R (k) )k=1,2,...,N Where N is the number of training samples; The neural network model is iteratively trained using the training sample set, and the loss function is minimized using an optimization algorithm to obtain the flavor ratio model.

7. The method according to claim 6, characterized in that The loss function is the MSE mean square error loss function, and the formula is as follows: Among them, L is the loss function value, is the actual j-th flavor ratio in the k-th sample; The j-th flavor ratio in the k-th sample predicted by the neural network model.

8. A flavor ratio customization device, characterized in that: The device comprises: A data collection module, used to collect taste-related data of target users; A feature extraction module, used to extract taste preference features from the taste-related data, and parameterize the taste preference features to obtain a flavor parameter vector; Model building module, used to build flavor ratio model based on neural network; The ratio analysis module is used to analyze the flavor parameter vector using the flavor ratio model and output a target flavor ratio that meets the taste of the target user.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the flavor ratio customization method according to any one of claims 1 to 7.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the flavor ratio customization method according to any one of claims 1 to 7 is implemented.

Images