Dish seasoning recommendation method, device and equipment for intelligent kitchen range and storage medium

Abstract

The invention discloses a dish seasoning recommendation method and device for an intelligent kitchen range, equipment and a storage medium, relates to the technical field of digital intelligent kitchens and intelligent kitchen ranges, and can realize personalized and intelligent recommendation of seasonings. According to the scheme, the method comprises the steps that to-be-cooked dishes of a target intelligent kitchen range and regional information corresponding to the target intelligent kitchen range are acquired; obtaining target historical cooking data of the target intelligent kitchen range, wherein the target historical cooking data comprises historical dishes, historical seasoning information corresponding to the historical dishes and satisfaction degree information; according to the regional information and the to-be-cooked dish, recommended seasoning information corresponding to the to-be-cooked dish is acquired from a pre-stored data set, and the data set comprises a mapping relation among the regional information, the dish information and the seasoning information; and obtaining target seasoning information of the to-be-cooked dish according to the target historical cooking data and the recommended seasoning information.

Description

Technical Field

[0001] The present application relates to the technical field of digital smart kitchen smart stoves, and in particular to a method, device, equipment and storage medium for recommending seasonings for dishes of smart stoves. Background Art

[0002] With the popularization of Internet of Things technology, more and more smart homes appear in people's daily lives. Among them, smart stoves, as a part of daily smart homes, have also become popular in people's daily lives over time.

[0003] At present, the relevant technologies for smart stoves are still at the stage of being able to control cooking time and cooking reservations through mobile devices, and are unable to make targeted recommendations for intelligent cooking. For example, users can only cook the seasoning plan for a certain dish through their own searches or daily experience. As more and more young people lack basic cooking knowledge, the recommendation of cooking seasoning plans is becoming more and more meaningful. Summary of the invention

[0004] The present application provides a dish seasoning recommendation method, device, equipment and storage medium for a smart stove, which can realize personalized and intelligent recommendation of seasonings.

[0005] In order to achieve the above purpose, this application adopts the following technical solutions: In a first aspect of an embodiment of the present application, a method for recommending seasonings for a smart stove is provided, the method comprising: Obtaining dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove; Acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information; According to the region information and the dish to be cooked, obtaining recommended condiment information corresponding to the dish to be cooked from a pre-stored data set, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information; The target seasoning information of the dish to be cooked is obtained according to the target historical cooking data and the recommended seasoning information.

[0006] As a possible implementation manner, before obtaining the regional information corresponding to the smart stove, the method further includes: Acquire training samples, where the training samples include seasoning information of multiple smart stoves when cooking multiple different dishes, and each smart stove includes corresponding regional information; The preset initial decision model is trained by using the training samples until the optimization function of the initial decision model meets the preset conditions, thereby obtaining a target decision model and a target decision function, wherein the initial decision model is a model based on a support vector machine; After the target decision function is used to filter the input information, the data set is obtained, and the input information includes regional information, dish information and condiment information.

[0007] As a possible implementation manner, the obtaining of training samples includes: Obtaining an initial weight of the smart seasoning box when the smart stove starts cooking a dish, and an ending weight of the smart seasoning box when the cooking of the dish is finished; According to the initial weight and the final weight, seasoning information of the smart stove when cooking dishes is obtained.

[0008] As a possible implementation manner, the using the training samples to train a preset initial decision model includes: Extracting features from the information in the training sample to obtain corresponding feature vectors; The feature vector is input into the initial decision model, and the initial decision model is trained in an unsupervised training manner.

[0009] As a possible implementation manner, the using the target decision function to screen the input information includes: After inputting the input information into the target decision function, a corresponding decision value is obtained; The input information whose decision value is greater than a preset threshold is stored in the data set.

[0010] As a possible implementation, there are multiple pieces of recommended seasoning information corresponding to one dish information in the data set; The step of obtaining target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information includes: Obtaining data means of a plurality of recommended condiments corresponding to the dish to be cooked, wherein the data means include: an average value and a median; The data mean is corrected according to the target historical cooking data corresponding to the dish to be cooked to obtain the target seasoning information of the dish to be cooked.

[0011] As a possible implementation, the target smart stove includes an image acquisition device, and the step of obtaining the dishes to be cooked by the target smart stove includes: The image acquisition device is used to acquire images of ingredients to be cooked of the target smart stove, and at least one dish to be cooked is identified based on the images of ingredients to be cooked.

[0012] In a second aspect of the embodiment of the present application, a dish seasoning recommendation device for a smart stove is provided, the device comprising: A first acquisition module is used to acquire dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove; A second acquisition module is used to acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information; a third acquisition module, configured to acquire, according to the region information and the dish to be cooked, recommended condiment information corresponding to the dish to be cooked from a pre-stored data set, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information; The determination module is used to obtain the target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information.

[0013] According to a third aspect of an embodiment of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the method for recommending seasonings for a smart stove in the first aspect of an embodiment of the present application is implemented.

[0014] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for recommending seasonings for a smart stove described in the first aspect of an embodiment of the present application is implemented.

[0015] The beneficial effects brought by the technical solution provided by the embodiment of the present application include at least: The method for recommending seasonings for dishes of a smart stove provided in an embodiment of the present application is to obtain the dishes to be cooked of a target smart stove and the regional information corresponding to the target smart stove; obtain the target historical cooking data of the target smart stove, the target historical cooking data including: historical dishes, historical seasoning information corresponding to the historical dishes and satisfaction information; obtain the recommended seasoning information corresponding to the dishes to be cooked from a pre-stored data set according to the regional information and the dishes to be cooked, the data set including the mapping relationship between regional information, dish information and seasoning information; obtain the target seasoning information of the dishes to be cooked according to the target historical cooking data and the recommended seasoning information. In this way, it is no longer a standard and general recommendation, but a more personalized seasoning scheme recommendation can be made after a comprehensive analysis of the current user's region, habits, tastes, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1A process of a dish seasoning recommendation method for a smart stove provided in an embodiment of the present application Figure 1 ; Figure 2 A process of a dish seasoning recommendation method for a smart stove provided in an embodiment of the present application Figure 2 ; Figure 3 A structural diagram of a dish seasoning recommendation device for a smart stove provided in an embodiment of the present application; Figure 4 A schematic diagram of the internal structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present disclosure, unless otherwise specified, "plurality" means two or more.

[0019] Additionally, the use of “based on” or “according to” is meant to be open and inclusive, as a process, step, calculation, or other action “based on” or “according to” one or more conditions or values ​​may, in practice, be based on additional conditions or beyond values.

[0020] With the popularization of Internet of Things technology, more and more smart homes appear in people's daily lives. Among them, smart stoves, as a part of daily smart homes, have also become popular in people's daily lives over time.

[0021] At present, the relevant technologies for smart stoves are still at the stage of being able to control cooking time and cooking reservations through mobile devices, and are unable to make targeted recommendations for intelligent cooking. For example, users can only cook the seasoning plan for a certain dish through their own searches or daily experience. As more and more young people lack basic cooking knowledge, the recommendation of cooking seasoning plans is becoming more and more meaningful.

[0022] The existing condiment recommendation scheme is based on the standard recipe seasoning scheme recommendation, which is not very friendly to cooking in daily life. In daily life, not all seasoning ingredients may be available, or the cooked dishes may vary greatly. The existing condiment recommendation lacks a more humanized recommendation scheme and cannot accurately identify the current user's taste preferences. The existing technology has poor collection of condiment schemes, and cannot collect the differences in condiment schemes for the same dishes in various provinces or cities, and lacks personalized data.

[0023] In response to the above problems, the present application proposes a method for efficiently and quickly collecting seasoning schemes, and can perform targeted analysis on the collected seasoning schemes to obtain seasoning schemes corresponding to the same dishes in different regions, so as to achieve the purpose of accurately recommending seasoning schemes, so that users can get more accurate recommended schemes when cooking dishes and cook more delicious food.

[0024] It should be noted that the executor of the dish seasoning recommendation method for a smart stove provided in an embodiment of the present application is an electronic device, which may be a smart stove, a data processing device or chip in the smart stove, or a server communicating with the smart stove. The embodiment of the present application does not make any specific limitations on this.

[0025] The present application embodiment provides a method for recommending seasonings for dishes of a smart stove, such as Figure 1 As shown, the method includes: Step 101: Obtain dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove.

[0026] Optionally, the above process may be: the target smart stove includes a display interface, and the user can input information about dishes to be cooked from the display interface, thereby obtaining dishes to be cooked by the target smart stove.

[0027] In addition, the user can also send information about dishes to be cooked to the target smart stove through the terminal.

[0028] Furthermore, the target smart stove includes an image acquisition device, and the step of obtaining dishes to be cooked by the target smart stove includes: The image acquisition device is used to acquire images of ingredients to be cooked of the target smart stove, and at least one dish to be cooked is identified based on the images of ingredients to be cooked.

[0029] It should be noted that the dish to be cooked identified based on the image of the food to be cooked may be one or more, and when the recommended condiment information corresponding to the dish to be cooked is subsequently obtained, the condiment information corresponding to all the dishes to be cooked may be obtained.

[0030] In addition, the geographical information corresponding to the target smart stove obtained in the above step 101 can be: the user who installed the smart stove registers the smart stove and reports the basic information of the smart stove such as the geographical location and IP to the electronic device. The electronic device can receive and store the basic information of all smart stoves, wherein the basic information includes but is not limited to the latitude and longitude information, the IP information of the smart stove, the city information to which the smart stove belongs, etc.

[0031] Step 102: Acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information.

[0032] Similarly, the target smart stove can obtain and record the seasoning information of each dish during each cooking process, and generate the target historical cooking data of the target smart stove.

[0033] Among them, the process of obtaining the seasoning information of each dish can be: obtaining the initial weight of the smart seasoning box of the target smart stove when it starts cooking the dish, and the final weight of the smart seasoning box when it finishes cooking the dish; based on the initial weight and the final weight, obtaining the seasoning information of the target smart stove when cooking the dish.

[0034] Step 103: According to the region information and the dish to be cooked, obtain recommended condiment information corresponding to the dish to be cooked from a pre-stored data set, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information.

[0035] Step 104: Obtain target seasoning information for the dish to be cooked based on the target historical cooking data and the recommended seasoning information.

[0036] Optionally, there are multiple pieces of recommended seasoning information corresponding to one dish information in the data set; The step of obtaining target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information includes: Obtaining data means of a plurality of recommended condiments corresponding to the dish to be cooked, wherein the data means include: an average value and a median; The data mean is corrected according to the target historical cooking data corresponding to the dish to be cooked to obtain the target seasoning information of the dish to be cooked.

[0037] Optional, such as Figure 2 As shown, before executing step 101 to obtain the regional information corresponding to the smart stove, the method further includes: Step 201: Obtain training samples, where the training samples include seasoning information of multiple smart stoves when cooking multiple different dishes, and each smart stove includes corresponding regional information.

[0038] Optionally, the process of obtaining training samples may be: Obtaining an initial weight of the smart seasoning box when the smart stove starts cooking a dish, and an ending weight of the smart seasoning box when the cooking of the dish is finished; According to the initial weight and the final weight, seasoning information of the smart stove when cooking dishes is obtained.

[0039] During the actual implementation process, the smart stove is associated with the smart seasoning box, the smart seasoning box initializes the initial value of the seasoning, the initial weight of each seasoning, and the smart stove initializes the initial value of the smart stove; the smart stove starts to cook the dish, and the user seasons the dish while cooking. The seasoning box encapsulates the seasoning plan according to the weight change of each seasoning, and reports the collected seasoning plan to the smart stove when the final dish is out of the pot. The smart seasoning box weight sensor is used to collect the weight change of each seasoning. This sensitive weight sensor has no technical restrictions and is mainly used to extract weight and pass relevant weight information to the smart seasoning box. The information collected here mainly includes but is not limited to the type of seasoning, the weight of the seasoning used, and the usage time of the seasoning bottle.

[0040] The smart stove reports the seasoning scheme information and regional information of the current dish to the service's Kafka, and sends them to different topics for each region and each dish, to achieve high-throughput data collection and reporting; during the cooking process, the smart stove will take high-definition photos of the dishes. After the dishes are cooked, the smart stove will report the current dish pictures and seasoning scheme information to the server. In addition to receiving the seasoning scheme information, the server can also identify the current dish type and dish name. The server receives the seasoning scheme and stores the collected seasoning schemes in ClickHouse, thus completing the collection of seasoning schemes for each dish.

[0041] Step 202: train a preset initial decision model using the training samples until the optimization function of the initial decision model meets preset conditions, thereby obtaining a target decision model and a target decision function, wherein the initial decision model is a model based on a support vector machine.

[0042] Among them, the initial decision model is the OC-SVM model.

[0043] Optionally, using the training samples to train a preset initial decision model includes: Extract features from the information in the training samples to obtain corresponding feature vectors; input the feature vectors into the initial decision model, and train the initial decision model using an unsupervised training method.

[0044] In the actual execution process, the data loading and data feature construction of the smart stove, seasoning information and other information are completed. According to the address location information of the smart stove, the relevant seasoning scheme information and the corresponding dish feature data are read from the clickhouse data storage medium. At this time, the data may still exist in structured (such as row and column data in a table), semi-structured (such as key-value pair information in JSON) or other forms. The features contained in the data may include: geographical information (latitude and longitude coordinates, city or region ID), dish classification information (dish name, classification code), seasoning scheme information (seasoning type, seasoning weight, seasoning usage time). Data vectorization is performed based on the extracted data features. In order to conduct subsequent training and inference of machine learning algorithms (such as OC-SVM), the above raw or semi-structured data needs to be converted into a numerical vector form that can be input into the model. The vectorization process includes but is not limited to: Direct use of numerical features: such as condiment category, condiment weight, condiment usage time, etc. can be directly used as components of the numerical feature vector. Category feature encoding: Geographic information (city name, area code), dish category, etc. need to use One-Hot encoding or Embedding to numerically encode classification features.

[0045] The final feature vector can be expressed as: ,in, is the feature vector corresponding to the i-th data, and d is the dimension of the feature vector.

[0046] After vectorization, a set of training samples can be obtained: .

[0047] Among them, n is the total number of data. Based on the construction of the above data feature vector, we can vectorize the data. The above vectorized output will be used as the input for subsequent algorithm (such as OC-SVM) calculation.

[0048] Construct the OC-SVM model and build the data model. After preprocessing, the system uses the OC-SVM (One-Class Support Vector Machine) algorithm to calculate and analyze according to the corresponding behavioral feature data (i.e., the preprocessed vector data X).

[0049] Among them, OC-SVM description: Different from the traditional SVM (support vector machine) classifier, OC-SVM is a one-class classification method, mainly used for anomaly detection and novelty detection. The core idea is that given a set of one-class data samples (in this scenario, these data samples can be regarded as the conventional seasoning scheme data of a certain region or a certain dish), OC-SVM tries to find a hyperplane in the feature space to separate the data from the origin and maximize the interval between the data and the hyperplane, so that most normal data are on one side of the decision boundary, and outliers or abnormal data are on the other side or far away from the decision boundary. For the seasoning scheme analysis in this patent, the normal data samples are the baseline of the conventional seasoning ratio scheme we need to obtain, while the outliers may be extremely unreasonable seasoning methods (the outliers here may mainly appear in the following aspects, such as the user's location is inconsistent with the user's actual location, the user's cooking taste is special and the deviation of a certain seasoning is too high). The decision function obtained by OC-SVM through learning can determine whether a new sample deviates from the main data group when it arrives, thereby identifying outliers with high deviations. This is the denoising process in data analysis, and the denoised data is the baseline data we need to obtain.

[0050] OC-SVM model training is mainly based on the training sample data we get to build the model. Here the first step is to give the training data , OC-SVM attempts to solve the following optimization problem:

[0051] Among them, in the above formula is the offset of the decision function, is the hyperplane slope, is a slack variable, which is used to allow some samples to be on the other side of the decision boundary. , is the number of samples, and n is the total number of data in each sample. is a hyperparameter used to control the proportion of outliers and the tightness of the decision boundary. , n is the total number of data items.

[0052] The constraints of the OC-SVM model are: .

[0053] in, is the feature mapping function, the input number Mapping into a high-dimensional feature space. This mapping can be implicitly achieved through kernel functions (such as Gaussian kernel, linear kernel, RBF kernel, etc.).

[0054] After the initial decision model, that is, the initial OC-SVM model is trained, the target OC-SVM model is obtained. The target OC-SVM model includes: a corresponding optimization function and a decision function.

[0055] In order to solve the above optimization problem, the decision function obtained by OC-SVM is:

[0056] After obtaining the corresponding decision function, for any new input x into the decision function, if , then it is judged as normal data (i.e., similar to the distribution of training samples), otherwise it is considered to be abnormal data.

[0057] Step 203: After using the target decision function to filter the input information, the data set is obtained, where the input information includes regional information, dish information, and condiment information.

[0058] Optionally, the above-mentioned screening of input information by using the target decision function includes: After inputting the input information into the target decision function, a corresponding decision value is obtained; The input information whose decision value is greater than a preset threshold is stored in the data set.

[0059] It is understandable that we use the OC-SVM model for single-class anomaly detection. Through the trained OC-SVM, for new seasoning samples, we can find the decision boundary that determines the normal data range in the feature space. For seasoning feature samples that fall outside the decision boundary, that is, seasoning schemes that are judged as abnormal by OC-SVM, they can be considered to be too special or extreme in the data. After completing the above data detection and analysis, we can get the following data results: a group of data samples (seasoning schemes) judged to be normal, these samples are relatively concentrated and stable in the feature space. A group of data samples judged to be abnormal (seasoning schemes with a large degree of deviation, data that deviates greatly from the current regional baseline data) can be selected to be eliminated or processed separately.

[0060] Abnormal samples are eliminated to obtain a "stable data set". The abnormal condiment data are filtered using the detection results of OC-SVM to remove samples that are obviously deviated from the main data distribution. For example, if a dish should use about 5 grams of salt, but there is a record of serious deviation (such as 100 grams of salt), the abnormal data will be eliminated. Through this cleaning process, a more "pure" and "stable" data set can be obtained, so that the actual distribution of the data can be more accurately reflected in the subsequent encoding and statistical processes, so that we can obtain a more reasonable stable data set, which is also the pre-stored data set in step 103.

[0061] After the data set is obtained, step 103 and step 104 may be performed.

[0062] In the actual implementation process, the user completes the preparation of relevant ingredients before cooking, and the smart stove completes the photo taking of the ingredients; the smart stove completes the photo taking of the initial state of the dish in the initial stage of cooking; the server performs corresponding image recognition based on the ingredients corresponding to the dish and the initial photo of the dish. The technology is not limited here to being able to recognize the ingredients and output the corresponding ingredient combination; after the server completes the dish recognition, it uses the stable seasoning data set baseline to propose the seasoning solution corresponding to the current dish, and extracts the quantitative reference value of the data. The following methods are mainly used to extract the quantitative reference value: Mean and median: Calculate the mean, median and other central tendency indicators for the usage of each condiment so that the recommendation plan can reflect the mainstream usage of most users. This method is to extract mainstream data.

[0063] Percentiles and ranges: Use percentiles (such as 25%, 50%, and 75%) to understand the upper and lower ranges of condiment usage, thereby providing users with a "recommended range of usage". This can not only provide single-point recommended values ​​(such as "5 grams of salt is recommended"), but also show users upper and lower limits (such as "3 grams to 7 grams of salt is recommended"). This method is mainly an extension of mainstream data, allowing users to make certain adjustments based on their actual tastes.

[0064] Regional information: Combined with the regional information of the current user, the seasoning scheme for the current regional dishes is obtained from the known regional data set. This seasoning scheme is mainly based on the common preferences of people in the same region, which can make the recommendation scheme more personalized.

[0065] User preference: Extract the seasoning scheme for the current dish of the current smart stove, extract the user's historical seasoning scheme data, and combine the above mean and quantile information to make the final output seasoning scheme more user-specific.

[0066] The seasoning solution information obtained from the final analysis is sent to the smart stove, and the user can complete the cooking of the dish according to the recommended seasoning solution.

[0067] The method for recommending seasonings for dishes of a smart stove provided in an embodiment of the present application is to obtain the dishes to be cooked of a target smart stove and the regional information corresponding to the target smart stove; obtain the target historical cooking data of the target smart stove, the target historical cooking data including: historical dishes, historical seasoning information corresponding to the historical dishes and satisfaction information; obtain the recommended seasoning information corresponding to the dishes to be cooked from a pre-stored data set according to the regional information and the dishes to be cooked, the data set including the mapping relationship between regional information, dish information and seasoning information; obtain the target seasoning information of the dishes to be cooked according to the target historical cooking data and the recommended seasoning information. In this way, it is no longer a standard and general recommendation, but a more personalized seasoning scheme recommendation can be made after a comprehensive analysis of the current user's region, habits, tastes, etc.

[0068] The present invention proposes a new method for recommending seasoning solutions for smart stoves based on kafka+clickhouse+big data model analysis, which can capture the seasoning solutions of any recipe cooked by the smart stove, and obtain people's preferences for dishes in fixed areas based on the geographical location of the smart stove. It supports data statistics and analysis functions for seasoning solutions for dishes, which are mainly reflected in data denoising and aggregation analysis, and can efficiently extract unified dishes and seasoning solutions for different regions from massive data. The regional seasoning solution recommendation technology is no longer a standard and general recommendation, but a more personalized seasoning solution recommendation based on a comprehensive analysis of the current user's region, habits, tastes, etc.

[0069] The present application embodiment provides a dish seasoning recommendation device for a smart stove, such as Figure 3 As shown, the device comprises: The first acquisition module 11 is used to acquire dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove; The second acquisition module 12 is used to acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information; A third acquisition module 13 is used to acquire recommended condiment information corresponding to the dish to be cooked from a pre-stored data set according to the region information and the dish to be cooked, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information; The determination module 14 is used to obtain the target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information.

[0070] In one embodiment, the device further comprises a processing module 15, wherein the processing module 15 is configured to: Acquire training samples, where the training samples include seasoning information of multiple smart stoves when cooking multiple different dishes, and each smart stove includes corresponding regional information; The preset initial decision model is trained by using the training samples until the optimization function of the initial decision model meets the preset conditions, thereby obtaining a target decision model and a target decision function, wherein the initial decision model is a model based on a support vector machine; After the target decision function is used to filter the input information, the data set is obtained, and the input information includes regional information, dish information and condiment information.

[0071] In one embodiment, the processing module 15 is specifically used for: Obtaining an initial weight of the smart seasoning box when the smart stove starts cooking a dish, and an ending weight of the smart seasoning box when the cooking of the dish is finished; According to the initial weight and the final weight, seasoning information of the smart stove when cooking dishes is obtained.

[0072] In one embodiment, the processing module 15 is specifically used for: Extracting features from the information in the training sample to obtain corresponding feature vectors; The feature vector is input into the initial decision model, and the initial decision model is trained in an unsupervised training manner.

[0073] In one embodiment, the processing module 15 is specifically used for: After inputting the input information into the target decision function, a corresponding decision value is obtained; The input information whose decision value is greater than a preset threshold is stored in the data set.

[0074] In one embodiment, there are multiple pieces of recommended seasoning information corresponding to one dish information in the data set; the determination module 14 is specifically used to: Obtaining data means of a plurality of recommended condiments corresponding to the dish to be cooked, wherein the data means include: an average value and a median; The data mean is corrected according to the target historical cooking data corresponding to the dish to be cooked to obtain the target seasoning information of the dish to be cooked.

[0075] In one embodiment, the target smart stove includes an image acquisition device, a first acquisition module, which is used to: The image acquisition device is used to acquire images of ingredients to be cooked of the target smart stove, and at least one dish to be cooked is identified based on the images of ingredients to be cooked.

[0076] The dish seasoning recommendation device for the smart stove provided in the embodiment of the present application can execute the above-mentioned dish seasoning recommendation method embodiment of the smart stove, and its implementation principle and technical effect are similar, which will not be elaborated here.

[0077] For the specific definition of the dish seasoning recommendation device of the smart stove, please refer to the definition of the dish seasoning recommendation method of the smart stove above, which will not be repeated here. Each module in the dish seasoning recommendation device of the above-mentioned smart stove can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor of the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0078] The executor of the dish seasoning recommendation method for the smart stove provided in the embodiment of the present application may be an electronic device, which may be a smart stove, a data processing device or chip in the smart stove, or a server or server cluster that communicates with the smart stove, etc. The embodiment of the present application does not specifically limit this.

[0079] Figure 4 This is a schematic diagram of the internal structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, the electronic device includes a processor and a memory connected via a system bus. The processor is used to provide computing and control capabilities. The memory may include a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program can be executed by the processor to implement the steps of the dish seasoning recommendation method for the smart stove provided in each of the above embodiments. The internal memory provides a cache operating environment for the operating system and the computer program in the non-volatile storage medium.

[0080] Those skilled in the art will understand that Figure 4 The internal structure diagram of the electronic device shown in the figure is only a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0081] In another embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the dish seasoning recommendation method for the smart stove in the embodiment of the present application are implemented.

[0082] In another embodiment of the present application, a computer program product is also provided. The computer program product includes computer instructions. When the computer instructions are run on the dish seasoning recommendation device of the smart stove, the dish seasoning recommendation device of the smart stove executes each step of the dish seasoning recommendation method of the smart stove in the method flow shown in the above method embodiment.

[0083] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer execution instruction is loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instruction can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instruction can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that can be accessed by the computer or a data storage device such as a server, data center, etc. that contains one or more servers that can be integrated with the medium. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0084] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0085] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for recommending seasoning for dishes of an intelligent stove, characterized in that: The method comprises: Obtaining dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove; Acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information; According to the region information and the dish to be cooked, obtaining recommended condiment information corresponding to the dish to be cooked from a pre-stored data set, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information; The target seasoning information of the dish to be cooked is obtained according to the target historical cooking data and the recommended seasoning information.

2. The method according to claim 1, characterized in that: Before obtaining the regional information corresponding to the smart stove, the method further includes: Acquire training samples, where the training samples include seasoning information of multiple smart stoves when cooking multiple different dishes, and each smart stove includes corresponding regional information; The preset initial decision model is trained by using the training samples until the optimization function of the initial decision model meets the preset conditions, thereby obtaining a target decision model and a target decision function, wherein the initial decision model is a model based on a support vector machine; After the target decision function is used to filter the input information, the data set is obtained, and the input information includes regional information, dish information and condiment information.

3. The method according to claim 2, characterized in that The obtaining of training samples comprises: Obtaining an initial weight of the smart seasoning box when the smart stove starts cooking a dish, and an ending weight of the smart seasoning box when the cooking of the dish is finished; According to the initial weight and the final weight, seasoning information of the smart stove when cooking dishes is obtained.

4. The method according to claim 2, characterized in that: The using the training samples to train the preset initial decision model includes: Extracting features from the information in the training sample to obtain corresponding feature vectors; The feature vector is input into the initial decision model, and the initial decision model is trained in an unsupervised training manner.

5. The method according to claim 2, characterized in that: The using the target decision function to screen the input information includes: After inputting the input information into the target decision function, a corresponding decision value is obtained; The input information whose decision value is greater than a preset threshold is stored in the data set.

6. The method according to claim 1, characterized in that There are multiple pieces of recommended seasoning information corresponding to one piece of dish information in the data set; The step of obtaining target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information includes: Obtaining data means of a plurality of recommended condiments corresponding to the dish to be cooked, wherein the data means include: an average value and a median; The data mean is corrected according to the target historical cooking data corresponding to the dish to be cooked to obtain the target seasoning information of the dish to be cooked.

7. The method according to claim 1, characterized in that The target smart stove includes an image acquisition device, and the step of obtaining dishes to be cooked by the target smart stove includes: The image acquisition device is used to acquire images of ingredients to be cooked of the target smart stove, and at least one dish to be cooked is identified based on the images of ingredients to be cooked.

8. A dish seasoning recommendation device for an intelligent stove, characterized in that: The device comprises: A first acquisition module is used to acquire dishes to be cooked by a target smart stove and regional information corresponding to the target smart stove; A second acquisition module is used to acquire target historical cooking data of the target smart stove, wherein the target historical cooking data includes: historical dishes, historical condiment information corresponding to the historical dishes, and satisfaction information; a third acquisition module, configured to acquire, according to the region information and the dish to be cooked, recommended condiment information corresponding to the dish to be cooked from a pre-stored data set, wherein the data set includes a mapping relationship between the region information, the dish information and the condiment information; The determination module is used to obtain the target seasoning information of the dish to be cooked according to the target historical cooking data and the recommended seasoning information.

9. An electronic device, characterized in that: It includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the dish seasoning recommendation method for the smart stove according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the dish seasoning recommendation method for the smart stove according to any one of claims 1 to 7 is implemented.

Images