A cooking step sequencing method and device based on multiple recipes
By identifying the dependencies between cooking steps in multiple recipes and optimizing the parameters, the problem of repeated and overlapping steps in multi-recipe cooking is solved, thus improving cooking and serving efficiency.
Patent Information
- Application Number
- CN202411118811.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-15
AI Technical Summary
In the process of cooking multiple recipes, the repeated and interspersed cooking steps cause users to be busy and flustered, making it impossible to effectively reduce the cooking time of multiple dishes and seriously affecting the overall serving efficiency.
By extracting a set of steps corresponding to cooking types from multiple recipes, identifying and sorting the dependencies between cooking steps, and optimizing the step sequence based on cooking parameters, the system utilizes a combination of mobile terminals and smart stoves to monitor cooking progress and environmental changes in real time, dynamically adjusting the step sequence and timing.
It achieves the correct order and optimization of cooking steps for multiple recipes, avoids conflicts and logical errors between steps, and improves cooking efficiency and serving efficiency.
Smart Images

Figure CN118643954B_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the technical field of intelligent devices, and in particular, relates to a cooking step sequencing method and device based on multiple recipes. Background Art
[0002] With the development of society and the increasingly faster pace of life, food safety issues have gradually become a focus of people's attention. The hygiene level and production methods of existing takeout can no longer meet people's health needs. More and more consumers are starting to go into the kitchen to learn how to cook, that is, to make healthier food by themselves.
[0003] Many users mainly refer to recipes when learning to cook in the kitchen, that is, they follow each step prompted by the recipe, and when preparing multiple dishes, they need to refer to multiple recipes accordingly. As a result, when users are cooking multiple recipes, they are often confused due to the repetition of multiple cooking steps. Not only does it fail to reduce the cooking time of multiple dishes, it also seriously affects the overall serving efficiency. Summary of the invention
[0004] In order to solve the above-mentioned technical defects that users often get flustered due to repeated interleaving of multiple cooking steps when cooking multiple recipes, which not only fails to reduce the cooking time of multiple dishes, but also seriously affects the overall serving efficiency, this application proposes a cooking step sorting method and device based on multiple recipes, and its technical solution is as follows:
[0005] In a first aspect, an embodiment of the present application provides a cooking step sequencing method based on multiple recipes, comprising:
[0006] Extracting a step set corresponding to the cooking type from at least two target recipes, and marking each cooking step in the step set;
[0007] Based on the identification result of each cooking step, the dependency relationship between any two cooking steps is determined, and all cooking steps are sorted according to all the dependency relationships;
[0008] At least two cooking parameters are identified from each cooking step, and the step sorting results are optimized according to all the cooking parameters to obtain step optimization results.
[0009] In an optional solution of the first aspect, before extracting the step set corresponding to the cooking type from the at least two target recipes, the method further includes:
[0010] Performing identification processing on each preset recipe in a preset recipe database to obtain corresponding at least two recipe parameters; wherein the preset recipe database includes at least two preset recipes;
[0011] Cleaning the preset recipe database according to all recipe parameters corresponding to each preset recipe;
[0012] At least two target recipes are determined from the preset recipe data after the cleaning process.
[0013] In yet another optional solution of the first aspect, before determining at least two target recipes from the preset recipe data after the cleaning process, the method further includes:
[0014] Extracting each preset recipe in the preset recipe data after the cleaning process to obtain corresponding at least two keywords;
[0015] According to all the keywords corresponding to each preset recipe, the cleaned preset recipe database is standardized;
[0016] At least two target recipes are determined from the preset recipe data after the cleaning process, including:
[0017] At least two target recipes are determined from the preset recipe data after the standardization process.
[0018] In another optional solution of the first aspect, extracting a step set corresponding to a cooking type from at least two target recipes comprises:
[0019] Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample;
[0020] Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample;
[0021] At least two cooking steps are extracted from the corresponding target recipe based on the classification result of each step, and a step set corresponding to the cooking type is obtained according to all the cooking steps.
[0022] In yet another optional solution of the first aspect, each cooking step in the step set is marked, including:
[0023] Identify each cooking step in the step set and extract the corresponding cooking action;
[0024] The cooking steps corresponding to each cooking action are numbered according to a preset numbering method.
[0025] In yet another optional solution of the first aspect, determining a dependency relationship between any two cooking steps based on the identification result of each cooking step includes:
[0026] Based on the arrangement order between any two adjacent cooking steps in each target recipe, determining the order relationship between the corresponding two cooking actions;
[0027] Matching all cooking actions based on a preset parallel-action database to obtain at least one set of cooking action sets; wherein the preset parallel-action database includes at least two sets of cooking action sets with a parallel relationship, and each set of cooking action sets includes two cooking actions;
[0028] According to all sequential relations and the two cooking steps corresponding to each sequential relation, and all parallel relations and the two cooking steps corresponding to each parallel relation, a dependency relation between any two cooking steps is obtained.
[0029] In yet another optional solution of the first aspect, all cooking steps are sorted according to all dependencies, including:
[0030] Construct a directed acyclic graph according to all sequential relations and two cooking steps corresponding to each sequential relation, all parallel relations and two cooking steps corresponding to each parallel relation, and the numbering result of each cooking step;
[0031] The directed acyclic graph is sorted based on a topological sorting algorithm to obtain a step sorting result.
[0032] In yet another alternative of the first aspect, the at least two cooking parameters include cooking time, cooking tools, and cooking methods;
[0033] The step sorting results are optimized according to all cooking parameters to obtain step optimization results, including:
[0034] The cooking time, cooking tools and cooking methods corresponding to each cooking step are scored respectively, and the three scores are weighted and summed based on the preset weights to obtain the corresponding priority score;
[0035] Based on the priority score corresponding to each cooking step, the step sorting results are sorted, and according to the cooking time and cooking tools corresponding to each cooking step, the sorted step sorting results are optimized to obtain step optimization results.
[0036] In another optional solution of the first aspect, after optimizing the step sorting results according to all cooking parameters to obtain the step optimization results, the method further includes:
[0037] Acquire at least two sensor data corresponding to the current step, and determine the step progress according to the cooking time of the current step and all sensor data; wherein the current step is any cooking step;
[0038] Show the user the progress of the steps; or
[0039] When an abnormality is detected in the step progress, at least one target step is selected from multiple cooking steps after the current step, and the step optimization result is updated according to all the target steps;
[0040] Show the user the optimization results of the updated steps.
[0041] In a second aspect, an embodiment of the present application provides a cooking step sequencing device based on multiple recipes, comprising:
[0042] A step identification module, used to extract a step set corresponding to a cooking type from at least two target recipes, and to identify each cooking step in the step set;
[0043] A step sorting module is used to determine the dependency between any two cooking steps based on the identification result of each cooking step, and sort all cooking steps according to all the dependency relationships;
[0044] The step optimization module is used to identify at least two cooking parameters from each cooking step, and optimize the step sorting results according to all the cooking parameters to obtain step optimization results.
[0045] In a third aspect, an embodiment of the present application further provides a mobile terminal, including a processor and a memory;
[0046] The processor is connected to the memory;
[0047] A memory for storing executable program codes;
[0048] The processor runs the program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the cooking step sorting method based on multiple recipes provided by the first aspect of the embodiment of the present application or any one of the implementations of the first aspect.
[0049] In a fourth aspect, an embodiment of the present application provides a computer storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the cooking step sorting method based on multiple recipes provided by the first aspect of the embodiment of the present application or any one of the implementation methods of the first aspect can be implemented.
[0050] In an embodiment of the present application, when sorting the cooking steps of multiple recipes selected by a user, a step set corresponding to the cooking type can be extracted from the multiple recipes, and the dependency relationship between any two cooking steps can be determined based on the identification result of each cooking step in the step set, so that all the cooking steps can be preliminarily sorted according to the dependency relationship, which can not only ensure the correct order of step execution, but also avoid conflicts and logical errors between steps; secondly, the preliminary step sorting results can be optimized according to the various cooking parameters corresponding to each cooking step to obtain the final step optimization results, which can not only effectively solve the coordination and optimization problems of cooking steps of multiple recipes, but also improve the cooking efficiency and serving efficiency of multiple dishes. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0052] Figure 1 A schematic diagram of a system architecture of a cooking step sequencing method based on multiple recipes provided in an embodiment of the present application;
[0053] Figure 2 An overall flow chart of a cooking step sequencing method based on multiple recipes provided in an embodiment of the present application;
[0054] Figure 3 A schematic diagram of the structure of a parallel-action database provided in an embodiment of the present application;
[0055] Figure 4 A schematic diagram of the effect of a directed acyclic graph provided in an embodiment of the present application;
[0056] Figure 5 An overall flow chart of another cooking step sequencing method based on multiple recipes provided in an embodiment of the present application;
[0057] Figure 6 An overall flow chart of another cooking step sequencing method based on multiple recipes provided in an embodiment of the present application;
[0058] Figure 7 A schematic diagram of a cooking step sequencing device based on multiple recipes provided in an embodiment of the present application;
[0059] Figure 8 A schematic diagram of the structure of a mobile terminal provided in an embodiment of the present application. DETAILED DESCRIPTION
[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.
[0061] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present application, and different embodiments can be replaced or combined, so the present application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then the present application should also be considered to include embodiments containing one or more of all other possible combinations of A, B, C, and D, although the embodiment may not be clearly recorded in the following text.
[0062] The following description provides examples and does not limit the scope, applicability or examples set forth in the claims. Changes may be made to the functions and arrangements of the elements described without departing from the scope of the present application. Various processes or components may be appropriately omitted, substituted or added to each example. For example, the described method may be performed in an order different from the order described, and various steps may be added, omitted or combined. In addition, features described in some examples may be combined in other examples.
[0063] It should be noted that there are some processing methods for cooking control of multiple recipes in the prior art. For example, please refer to the method of a multi-dish cooking control method, control system and cooking utensils with application number CN202011025338.4, which receives the user's selection instruction, reads at least two navigation recipes selected by the user, and extracts the navigation data in the cooking process specified by at least two navigation recipes; receives the user's start instruction, and starts the cooking navigation of each navigation recipe one by one according to the user's instruction, and sends the following control signals to at least two controlled cooking utensils that respectively execute one of the recipes: ignition or start electric heater signal, fire power adjustment signal, cooking material delivery signal. However, the above scheme does not involve processing methods such as sorting or merging the cooking steps of multiple recipes, but only sends corresponding control signals according to the extracted cooking steps and user instructions. In essence, it does not solve the technical defects proposed in this application, such as the confusion caused by the repeated interweaving of multiple cooking steps, which not only cannot reduce the cooking time of multiple dishes, but also seriously affects the overall serving efficiency.
[0064] In summary, the present application will explain the ordering method of cooking steps in multiple recipes selected by the user in combination with one or more embodiments shown below.
[0065] See next Figure 1 , Figure 1 A schematic diagram of the system architecture of a cooking step sequencing method based on multiple recipes provided in an embodiment of the present application is shown.
[0066] like Figure 1 As shown, the system architecture of the cooking step sorting method based on multiple recipes may at least include a mobile terminal and a smart cooker. A communication connection may be established between the mobile terminal and the smart cooker. The mobile terminal is installed with a third-party application that facilitates users to select one or more recipes. After the user selects multiple recipes, the cooking steps of the multiple recipes are sorted, and the sorted cooking steps are displayed to the user through voice broadcast or text scrolling, so that the user can cook multiple recipes according to the sorted cooking steps.
[0067] It is understandable that when the user needs to perform multi-recipe cooking, the user can also, but is not limited to, issue an execution operation to the mobile terminal to remind the mobile terminal to display all the sorted cooking steps to the user, such as voice broadcasting of data such as cooking ingredients, cooking equipment, cooking methods and cooking time corresponding to the current step. If there are other cooking steps executed in parallel, the cooking ingredients, cooking equipment, cooking methods and cooking time corresponding to the current step and the cooking ingredients, cooking equipment, cooking methods and cooking time corresponding to other cooking steps executed in parallel can also be broadcast separately in a voice loop, and a data request can also be issued to the smart stove at the same time, so that the smart stove can obtain various types of sensor data when the user performs multi-recipe cooking, so that the mobile terminal can judge the execution status of the current step according to the various types of sensor data fed back by the smart stove, and can not only monitor the overall cooking progress and environmental changes in real time, but also dynamically adjust the execution order and time schedule of all remaining cooking steps according to the execution status of the current step, thereby improving the overall cooking efficiency and accuracy.
[0068] Here, before the user performs the multi-recipe cooking process, the mobile terminal can also extract all the preparation steps from the recipe selected by the user, and can also display all the preparation steps to the user through voice broadcast or text scrolling, so as to improve the user's preparation efficiency. Of course, in the embodiment of the present application, the mobile terminal can also sort all the preparation steps and display all the preparation efficiencies after sorting to the user, so as to further improve the user's preparation efficiency.
[0069] A smart stove can be understood as an IoT device that automatically cooks according to instructions input by the user, such as a smart gas stove or a smart integrated stove. The smart stove can be provided with various types of sensors, such as but not limited to sensor 1 (also known as a temperature sensor) for obtaining temperature data, sensor 2 (also known as a humidity sensor) for obtaining humidity data, and sensor 3 (also known as a weight sensor) for obtaining weight data, so that after receiving a data request from a mobile terminal, the smart stove can simultaneously control sensor 1, sensor 2, and sensor 3 to obtain corresponding sensor data respectively, and upload all sensor data to the mobile terminal. Here, the smart stove can also, but is not limited to, process the cooking image corresponding to the current step by image processing and other methods, and can feed back the image processing results to the mobile terminal, but is not limited to this.
[0070] Of course, the smart stove in the embodiment of the present application can also, but is not limited to, display the completion status interface of the current step to the user, so that the user can manually input or automatically select the completion status corresponding to the current step, and can upload the completion status corresponding to the current step to the mobile terminal, which is also not limited to this.
[0071] It can also be understood that when the mobile terminal sorts the cooking steps of multiple recipes selected by the user, it can extract a step set corresponding to the cooking type from the multiple recipes, and determine the dependency relationship between any two cooking steps based on the identification result of each cooking step in the step set, so as to perform preliminary sorting of all cooking steps according to the dependency relationship, which can not only ensure the correct order of step execution, but also avoid conflicts and logical errors between steps; secondly, the preliminary step sorting results can be optimized according to the various cooking parameters corresponding to each cooking step to obtain the final step optimization results, which can not only effectively solve the coordination and optimization problems of cooking steps of multiple recipes, but also improve the cooking efficiency and serving efficiency of multiple dishes.
[0072] See next Figure 2 , Figure 2 An overall flow chart of a cooking step sequencing method based on multiple recipes provided in an embodiment of the present application is shown.
[0073] like Figure 2 As shown, the cooking step sequencing method based on multiple recipes may at least include the following steps:
[0074] Step 202: extract a step set corresponding to the cooking type from at least two target recipes, and perform identification processing on each cooking step in the step set.
[0075] Specifically, when sorting the cooking steps of multiple recipes selected by the user, the mobile terminal can, but is not limited to, filter out the target recipe corresponding to each recipe name according to the multiple recipe names selected by the user on the display interface. The target recipe may specifically include a recipe name, an ingredient list, one or more preparation steps and one or more cooking steps. Each preparation step also includes data such as corresponding ingredients, required equipment (such as a knife, pot or bowl, etc.) and preparation time. Each cooking step also includes corresponding ingredients, required equipment, cooking methods (such as temperature and heat requirements for steaming, boiling or frying) and cooking time, etc., and all cooking steps can be identified in the target recipe corresponding to each recipe name, so as to take the set of all cooking steps as a step set corresponding to the cooking type.
[0076] Furthermore, after obtaining the step set corresponding to the cooking type, the mobile terminal can also identify each cooking step in the step set. The identification processing method can be but is not limited to numbering all cooking steps according to different target recipes and the cooking step sorting method corresponding to each target recipe. For example, the target recipes can be represented as A and B respectively, the cooking steps corresponding to A include 3, and the cooking steps corresponding to B include 2, then the cooking step identification results in the step set can be A1, A2, A3, B1 and B2 respectively, and are not limited to this.
[0077] As an option of the embodiment of the present application, before extracting the step set corresponding to the cooking type from the at least two target recipes, the method further includes:
[0078] Performing identification processing on each preset recipe in a preset recipe database to obtain corresponding at least two recipe parameters; wherein the preset recipe database includes at least two preset recipes;
[0079] Cleaning the preset recipe database according to all recipe parameters corresponding to each preset recipe;
[0080] At least two target recipes are determined from the preset recipe data after the cleaning process.
[0081] In order to ensure the uniformity of the target recipes, as well as the accuracy and consistency of subsequent processing, the preset recipes in the preset recipe database may be cleaned before the user selects multiple target recipes to remove duplicate recipes or duplicate steps.
[0082] Specifically, before extracting the step set corresponding to the cooking type from the multiple target recipes, the mobile terminal can also identify each preset recipe in the preset recipe database to obtain at least two corresponding recipe parameters, and the type of the recipe parameter can be but is not limited to any at least two of the recipe name, ingredient list, all preparation steps, or all cooking steps. Here, the preset recipe database can store multiple preset recipes obtained through the Internet or big data statistics, and the preset recipe database can also be updated according to a preset time period to maximize the satisfaction of the user's cooking needs.
[0083] Next, after identifying the multiple recipe parameters corresponding to each preset recipe, the mobile terminal may also clean all the preset recipes included in the preset recipe database according to the multiple recipe parameters corresponding to each preset recipe, for example but not limited to, judging whether there are any two preset recipes with the same recipe name according to the recipe name corresponding to each preset recipe, and after determining that there are any two preset recipes with the same recipe name, performing similarity calculation processing on the preparation steps and cooking steps respectively included in the two preset recipes, so that when a high similarity is detected, the two preset recipes can be determined to be duplicate recipes, and the two preset recipes can be duplicated. Alternatively, according to all the preparation steps and all the cooking steps corresponding to each preset recipe, all the preparation steps and all the cooking steps can be converted into vectors respectively, and the similarity between any two vectors in all the vectors corresponding to all the preparation steps is calculated, and the similarity between any two vectors in all the vectors corresponding to all the cooking steps is calculated, so that when a high similarity is detected, the corresponding two vectors can be determined to be repeated steps, and then any one of the two preparation steps or any one of the two cooking steps corresponding to the two vectors can be cleaned.
[0084] Next, after the preset recipe database is cleaned, the mobile terminal can search for a target recipe corresponding to each recipe name in the preset recipe data after the cleaning process according to the multiple recipe names selected by the user. Here, when the preset recipe data after the cleaning process does not search for a target recipe corresponding to any recipe name, it indicates that the corresponding recipe has not been entered into the current preset recipe data, and then the recipe can be obtained through Internet query or other methods, and the recipe after the cleaning process is used as the target recipe.
[0085] As another option of the embodiment of the present application, before determining at least two target recipes from the preset recipe data after the cleaning process, the method further includes:
[0086] Extracting each preset recipe in the preset recipe data after the cleaning process to obtain corresponding at least two keywords;
[0087] According to all the keywords corresponding to each preset recipe, the cleaned preset recipe database is standardized;
[0088] At least two target recipes are determined from the preset recipe data after the cleaning process, including:
[0089] At least two target recipes are determined from the preset recipe data after the standardization process.
[0090] In order to further ensure the uniformity of the target recipes, as well as the accuracy and consistency of subsequent processing, the preset recipes in the preset recipe database may be standardized before the user selects multiple target recipes, so as to unify the formats of all preset recipes.
[0091] Specifically, after completing the cleaning process of the preset recipe database, the mobile terminal can also perform keyword extraction processing on each preset recipe in the preset recipe database after the cleaning process to obtain at least two corresponding keywords. The type of the keyword can be but is not limited to any at least two of the recipe name, ingredient list, preparation time of all preparation steps, preparation actions or cooking time of all cooking steps, cooking actions, and the corresponding preset recipes can be standardized according to the multiple keywords corresponding to each preset recipe.
[0092] Here, taking the keywords corresponding to each preset recipe including the recipe name, ingredient list, preparation time of all preparation steps, preparation actions, and cooking time and cooking actions of all cooking steps as an example, the mobile terminal can, but is not limited to, perform spelling recognition processing on the recipe name, ingredient list, preparation actions of all preparation steps, and cooking actions of all cooking steps corresponding to each preset recipe through an edit distance algorithm, a dictionary comparison, or a language processing model to determine whether there are spelling errors, and correct the corresponding spelling when it is determined that there are spelling errors; secondly, the mobile terminal can also unify the preparation time of all preparation steps and the cooking time of all cooking steps corresponding to each preset recipe. Secondly, the mobile terminal can also query the standard action corresponding to each food preparation action and the standard action corresponding to each cooking action in the preset action library. The preset action library may include at least two standard actions and at least one common action corresponding to each standard action. When it is detected that any food preparation action is inconsistent with the corresponding standard action, the food preparation action can be updated to the corresponding standard action, or when it is detected that any cooking action is inconsistent with the corresponding standard action, the cooking action can be updated to the corresponding standard action. For example, when any food preparation action is slicing and the corresponding standard action is slicing, slicing can be replaced by slicing.
[0093] Of course, the mobile terminal can also query the standard name corresponding to each ingredient name in the preset ingredient library, and when it is detected that any ingredient name is inconsistent with the corresponding standard name, the ingredient name is replaced with the corresponding standard name. For example, when any ingredient name is chopped green onion and the corresponding standard name is shallot, the chopped green onion can be replaced with shallot, and the present invention is not limited to this.
[0094] Next, after completing the standardization process on the preset recipe database, the mobile terminal can search for a target recipe corresponding to each recipe name in the preset recipe database after the standardization process according to the multiple recipe names selected by the user.
[0095] As another option of the embodiment of the present application, extracting a step set corresponding to a cooking type from at least two target recipes includes:
[0096] Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample;
[0097] Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample;
[0098] At least two cooking steps are extracted from the corresponding target recipe based on the classification result of each step, and a step set corresponding to the cooking type is obtained according to all the cooking steps.
[0099] In order to ensure the accuracy of the cooking steps extracted from each target recipe, natural language processing technology can also be used to analyze and classify each target recipe.
[0100] Specifically, when extracting a set of steps corresponding to a cooking type from at least two target recipes, the mobile terminal may also input all the preparation steps and all the cooking steps in each target recipe into a preset language processing model to obtain a high-dimensional text vector corresponding to all the preparation steps and all the cooking steps, and may input the high-dimensional text vector into a preset deep learning model to obtain a corresponding step classification result, which may be, but is not limited to, a predicted probability that each step corresponds to a cooking step. Here, the preset language processing model may be a conventional BERT language model, and may be trained by at least two step samples and a vector sample corresponding to each step sample; the preset deep learning model may be a neural network using supervised learning, which may be trained by at least two vector samples and a step classification label corresponding to each vector sample, and the step classification label may include a cooking label and a preparation label.
[0101] Next, after obtaining the step classification results corresponding to each target recipe, the mobile terminal can also filter out multiple cooking steps in the corresponding target recipe according to the step classification results, and can set all cooking steps as a step set corresponding to the cooking type.
[0102] As another option of the embodiment of the present application, each cooking step in the step set is marked, including:
[0103] Identify each cooking step in the step set and extract the corresponding cooking action;
[0104] The cooking steps corresponding to each cooking action are numbered according to a preset numbering method.
[0105] In order to ensure the logical order of subsequent sorting processing, each cooking step can also be analyzed and processed to obtain corresponding numbered processing results according to the cooking actions.
[0106] Specifically, when identifying each cooking step in the step set, the mobile terminal can also perform keyword extraction processing on each cooking step in the step set to obtain one or more corresponding cooking actions (for example, cutting onions, marinating beef, boiling water, and stir-frying vegetables, etc.). The keyword extraction processing method can be but not limited to referring to the language processing model or deep learning model mentioned above, which will not be elaborated here. The cooking steps corresponding to all cooking actions of each target recipe can be numbered according to a preset numbering method. The preset numbering method can be numbering from small to large according to the order of the steps in the target recipe, so that the numbering result of the cooking step corresponding to each cooking action is unique.
[0107] Here, a target recipe includes three cooking steps as an example. The numbering results corresponding to the three cooking steps can be represented as A1, A2 and A3 (A can be used to distinguish different target recipes, for example, another target recipe can be represented by B), or when the cooking actions included in the three cooking steps are cutting onions, marinating beef, and boiling water, respectively, the corresponding numbering results can also be represented as A1 cutting onions, A2 marinating beef, and A3 boiling water, and are not limited to this.
[0108] Step 204: Based on the identification result of each cooking step, determine the dependency relationship between any two cooking steps, and sort all the cooking steps according to all the dependency relationships.
[0109] Specifically, after obtaining the identification result of each cooking step, the mobile terminal can, but is not limited to, determine the dependency relationship between the two cooking steps with corresponding identification results according to the arrangement order between any two adjacent cooking steps in all target recipes. For example, the dependency relationship between the cooking step of heating the oil pan and the cooking step of frying onions until golden brown can be a sequential relationship, that is, heating the oil pan is executed before frying the onions until golden brown; or according to the preset cooking parallel logic, any two cooking steps that can be executed in parallel can be screened out from all cooking steps with identification results according to the preset cooking parallel logic. For example, the dependency relationship between the cooking step of boiling water and the cooking step of steaming corn can be a parallel relationship, that is, boiling water and steaming corn can be executed at the same time; or according to the preset cooking time logic, any two cooking steps arranged according to cooking time can be screened out from all cooking steps with identification results according to the preset cooking time logic. For example, the dependency relationship between the cooking step of pouring soy sauce and the cooking step of boiling for 5 minutes can also be a sequential execution relationship, that is, pouring soy sauce is executed before boiling for 5 minutes.
[0110] It can be understood that the dependency relationship can be but is not limited to the sequential relationship and parallel relationship mentioned above, and the preset cooking parallel logic can be but is not limited to being obtained by analyzing all preset recipes in the preset recipe database, and of course can also be obtained by analyzing multiple target recipes selected by the user, and the preset cooking parallel logic can include multiple groups of two cooking steps with a parallel relationship; the preset cooking time logic can also be but is not limited to being obtained by analyzing all preset recipes in the preset recipe database, and the preset cooking time logic can include multiple groups of two cooking steps with a sequential relationship.
[0111] In an embodiment of the present application, it may also include, but is not limited to, step premise logic and step result logic analyzed from all preset recipes in a preset recipe database, so as to determine the dependency relationship between two cooking steps with corresponding identification results according to the step premise logic and the step result logic, wherein the step premise logic may also include multiple groups of two cooking steps with a sequential relationship, such as heating an oil pan and frying onions in the pan until golden brown; the step result logic may also include multiple groups of two cooking steps with a sequential relationship, such as cutting onions and frying onions in the pan until golden brown, and is not limited to this.
[0112] Furthermore, after determining the dependency relationship between any two cooking steps with identification results, the mobile terminal can sort the identification results of all corresponding cooking steps according to all dependencies, and any two adjacent identification results can be connected by an arrow, which can be but is not limited to representing a sequential relationship or a parallel relationship, and a preliminary sorting result of all cooking steps is obtained.
[0113] As another option of the embodiment of the present application, based on the identification result of each cooking step, determining the dependency relationship between any two cooking steps includes:
[0114] Based on the arrangement order between any two adjacent cooking steps in each target recipe, determining the order relationship between the corresponding two cooking actions;
[0115] Matching all cooking actions based on a preset parallel-action database to obtain at least one set of cooking action sets; wherein the preset parallel-action database includes at least two sets of cooking action sets with a parallel relationship, and each set of cooking action sets includes two cooking actions;
[0116] According to all sequential relations and the two cooking steps corresponding to each sequential relation, and all parallel relations and the two cooking steps corresponding to each parallel relation, a dependency relation between any two cooking steps is obtained.
[0117] In order to further ensure the accuracy of the logical sequence between the cooking steps, a more accurate dependency relationship can be obtained by analyzing and processing the cooking actions corresponding to the cooking steps.
[0118] Specifically, when determining the dependency relationship between any two cooking steps, the mobile terminal can also, but is not limited to, identify the arrangement order of all cooking steps in each target recipe, and identify all two cooking actions that have a sequential relationship among all cooking actions. For example, if a target recipe contains cooking step A and cooking step B, and cooking step A is before cooking step B, then cooking action a corresponding to cooking step A and cooking action b corresponding to cooking step B can be identified among all cooking actions. At this time, cooking action a and cooking action b have a sequential relationship (i.e., a sequential relationship).
[0119] Next, the mobile terminal can also match all cooking actions based on the preset parallel-action database to identify all cooking action sets with parallel relationships (i.e., parallel relationships), and each cooking action set includes any two cooking actions from all cooking actions. Here, the preset parallel-action database can be, but is not limited to, obtained by analyzing all preset recipes in the preset recipe database mentioned above, and the preset parallel-action database includes at least two sets of cooking action sets with parallel relationships, and each set of cooking action sets includes two cooking actions.
[0120] Here, you can also see Figure 3 The structure diagram of a parallel-action database provided by an embodiment of the present application is shown as follows: Figure 3 As shown, the parallel-action database may include two sets of cooking action sets with a parallel relationship. The two cooking actions in the first set of cooking action sets may be boiling water and beating eggs, respectively, and the two cooking actions in the second set of cooking action sets may be steaming corn and slow-cooking over medium or low heat, respectively. In the embodiment of the present application, the number of cooking action sets is not limited.
[0121] Of course, in the embodiment of the present application, the mobile terminal can also, but is not limited to, match all cooking actions based on a preset sequence-action database to identify all cooking action sets with a sequential relationship. The sequence-action database is obtained by analyzing all preset recipes in the preset recipe database mentioned above, which includes all cooking action sets with a sequential relationship corresponding to the preset cooking parallel logic, step pre-logic and step result logic mentioned above, and is not limited to this.
[0122] Next, after obtaining all sequential relationships and all parallel relationships between all cooking actions, the mobile terminal can sort the identification results of the cooking steps corresponding to all corresponding cooking actions according to all sequential relationships and all parallel relationships, and any two adjacent identification results can be connected by an arrow, which can be but is not limited to representing a sequential relationship or a parallel relationship, and obtain a preliminary sorting result of all cooking steps.
[0123] As another option of the embodiment of the present application, all cooking steps are sorted according to all dependencies, including:
[0124] Construct a directed acyclic graph according to all sequential relations and two cooking steps corresponding to each sequential relation, all parallel relations and two cooking steps corresponding to each parallel relation, and the numbering result of each cooking step;
[0125] The directed acyclic graph is sorted based on a topological sorting algorithm to obtain a step sorting result.
[0126] In order to ensure the accuracy of sorting all cooking steps according to the dependency relationship between any two cooking steps, a preliminary sorting result of all cooking steps can be generated by constructing a directed acyclic graph and performing a topological sorting process.
[0127] Specifically, when all cooking steps are sorted according to all dependencies, the mobile terminal can also, but is not limited to, use the numbering result of each cooking step as a node, and the sequential relationship or parallel relationship between any two cooking steps as a connecting arrow, and generate a directed acyclic graph by traversing all nodes and all connecting arrows.
[0128] Here, you can also see Figure 4 The effect schematic diagram of a directed acyclic graph provided by an embodiment of the present application is shown as follows: Figure 4 As shown, the directed acyclic graph may include six nodes, namely, A1, A2, A3, A4, B1 and B2, wherein: A1, A2, A3 and A4 nodes may correspond to the four cooking steps of target recipe A, and have a sequential relationship in sequence; B1 and B2 nodes may correspond to the two cooking steps of target recipe B, and have a sequential relationship in sequence. In addition, A2 and B1 nodes have a sequential relationship, and A3 and B1 nodes have a parallel relationship.
[0129] Next, after constructing the directed acyclic graph, the mobile terminal may further sort the directed acyclic graph based on a topological sorting algorithm. The topological sorting algorithm may be, but is not limited to, the Kahn algorithm well known in the art. The specific process may be as follows:
[0130] Add all nodes with in-degree 0 to the queue;
[0131] Process the nodes in the queue one by one, remove them from the graph, and reduce the in-degree of their adjacent nodes by 1; if the in-degree of an adjacent node is reduced to 0, add it to the queue;
[0132] Repeat the above steps until the queue is empty.
[0133] It is understandable that when it is detected that the number of nodes in the sorting result is consistent with the number of nodes in the directed acyclic graph, the sorting result can be used as the step sorting result of all cooking steps, for example but not limited to referring to the above Figure 4 The effect diagram of the directed acyclic graph is shown, and the step sequence can be expressed as A1-A2-A3-A4-B1-B2, or A1-A2-B1-B2-A3-A4.
[0134] It should be noted that when it is detected that the number of nodes in the sorting result is inconsistent with the number of nodes in the directed acyclic graph, or the queue is empty but the directed acyclic graph still has nodes, it indicates that there is a cycle (that is, the sorting fails), which may be caused by multiple nodes with circular dependencies. Then, the specific cycle can be found by but not limited to depth-first search (DFS) or other algorithms, such as using Tarjan's strongly connected component algorithm to detect and identify all cycles; then, all nodes corresponding to the cycle can be analyzed and processed, such as checking whether the dependency between any two nodes is really needed, or whether these dependencies can be broken by rearranging the cooking steps; then, the dependencies can be broken by reorganizing the cooking steps, such as adding additional cooking steps to resolve the dependencies, or redesigning the order of the cooking steps, until the number of nodes in the sorting result is consistent with the number of nodes in the directed acyclic graph.
[0135] Step 206: Identify at least two cooking parameters from each cooking step, and optimize the step sorting results according to all cooking parameters to obtain step optimization results.
[0136] Specifically, after obtaining the step ranking results of all cooking steps, the mobile terminal may, but is not limited to, identify multiple cooking parameters from each cooking step to determine the corresponding priority according to all cooking parameters corresponding to each cooking step, and may optimize the step ranking results in descending order of priority, that is, adjust the position of one or more cooking steps in the step ranking results, so that the final step optimization result can minimize the total cooking time, maximize resource utilization efficiency, minimize step repetition, and maximize parallelism as much as possible. Here, the type of cooking parameters may be, but is not limited to, any two of cooking time, cooking tools, and cooking methods.
[0137] See next Figure 5 , Figure 5 An overall flow chart of another cooking step sequencing method based on multiple recipes provided in an embodiment of the present application is shown.
[0138] like Figure 5 As shown, the cooking step sequencing method based on multiple recipes may at least include the following steps:
[0139] Step 502: extract a step set corresponding to the cooking type from at least two target recipes, and perform identification processing on each cooking step in the step set.
[0140] Specifically, step 502 may refer to step 202, which will not be described in detail here.
[0141] Step 504: Based on the identification result of each cooking step, determine the dependency relationship between any two cooking steps, and sort all the cooking steps according to all the dependency relationships.
[0142] Specifically, step 504 may refer to step 204, which will not be described in detail here.
[0143] Step 506: Identify the cooking time, cooking tools and cooking methods from each cooking step, assign scores to the cooking time, cooking tools and cooking methods corresponding to each cooking step, and perform weighted summation on the three scores based on preset weights to obtain corresponding priority scores.
[0144] Specifically, after obtaining the step ranking results of all cooking steps, the mobile terminal may also, but is not limited to, identify the cooking time, cooking tools and cooking methods from each cooking step, and respectively assign points to the cooking time, cooking tools and cooking methods according to the corresponding scoring rules. For example, taking the cooking time as an example, when the cooking time is 10 minutes, the corresponding score may be 70, and when the cooking time is 15 minutes, the corresponding score may be 75, and the present invention is not limited thereto. Here, the scoring rules corresponding to each cooking parameter may be, but are not limited to, obtained by manual analysis, so as to facilitate the subsequent calculation of the priority score.
[0145] Next, after determining the three scores corresponding to the three cooking parameters of each cooking step, the mobile terminal can also perform weighted sum calculation on the three scores of each cooking step according to the preset weights to obtain the corresponding priority score. The higher the priority score, the higher the ranking position of the corresponding cooking step, and the cooking step can be executed first; the lower the priority score, the later the ranking position of the corresponding cooking step, and the cooking step can be executed later.
[0146] Step 508: sort the step sorting results based on the priority score corresponding to each cooking step, and optimize the step sorting results after sorting according to the cooking time and cooking tools corresponding to each cooking step to obtain step optimization results.
[0147] Specifically, after calculating the priority score of each cooking step, the mobile terminal can also re-sort the step sorting results in descending order of the priority scores to ensure that the sorted results are more reliable, and can, but is not limited to, use a greedy algorithm to optimize the sorted results according to the cooking time and cooking tools corresponding to each cooking step, so that based on the sorted results, the cooking steps identified as having a parallel relationship can be adjusted, and the cooking steps with cooking tool conflicts can be adjusted (i.e., avoiding resource conflicts), and the cooking steps with idle cooking tools can be adjusted (i.e., avoiding idle resources), thereby generating a final step sorting result (i.e., step optimization result) that minimizes the total cooking time, maximizes resource utilization efficiency, minimizes step repetitions, and maximizes parallelism. It can be understood that the above-mentioned greedy algorithm can be a well-known technical means in the field of global optimization, and its technical principles will not be described in detail here.
[0148] Of course, in the embodiment of the present application, the greedy algorithm can also be replaced by a genetic algorithm, that is, the genetic algorithm is used to optimize and sort all cooking steps in the step sorting result according to the parameters such as cooking time, cooking tools and cooking methods corresponding to each cooking step. Here, the implementation of the genetic algorithm can also be, but is not limited to, as follows:
[0149] Initial population generation: Based on all the cooking steps in the step sorting results, multiple step sortings are randomly generated;
[0150] Fitness function: Based on indicators such as total cooking time and resource utilization efficiency, the parameters such as cooking time, cooking tools and cooking methods corresponding to each cooking step are analyzed and processed to evaluate the pros and cons of each ranking;
[0151] Selection, crossover, and mutation: Generate new sorting by selecting the best individuals, crossover recombination, and mutation;
[0152] Iterative optimization: Repeat the above process and gradually optimize to generate the final step sorting result.
[0153] See next Figure 6 , Figure 6 An overall flow chart of another cooking step sequencing method based on multiple recipes provided in an embodiment of the present application is shown.
[0154] like Figure 6As shown, the cooking step sequencing method based on multiple recipes may at least include the following steps:
[0155] Step 602: extract a step set corresponding to the cooking type from at least two target recipes, and perform identification processing on each cooking step in the step set.
[0156] Specifically, step 602 may refer to step 202, which will not be described in detail here.
[0157] Step 604: Based on the identification result of each cooking step, determine the dependency relationship between any two cooking steps, and sort all the cooking steps according to all the dependency relationships.
[0158] Specifically, step 604 may refer to step 204, which will not be described in detail here.
[0159] Step 606: Identify at least two cooking parameters from each cooking step, and optimize the step sorting results according to all cooking parameters to obtain step optimization results.
[0160] Specifically, step 606 may refer to step 206, which will not be described in detail here.
[0161] Step 608: Obtain at least two sensor data corresponding to the current step, and determine the step progress according to the cooking time of the current step and all sensor data.
[0162] In the embodiment of the present application, the cooking progress can be monitored in real time and the order and timing of the cooking steps can be dynamically adjusted to improve the overall cooking efficiency and accuracy.
[0163] Specifically, when the user performs a multi-recipe cooking process, an execution operation may be issued on the display interface of the mobile terminal, but is not limited to, to remind the mobile terminal to display all the cooking steps that have been sorted and processed to the user. At the same time, the mobile terminal may also issue a data request to the smart cooker, so that the smart cooker can timely feed back the various sensor data collected in the current step to the mobile terminal. Here, the various sensor data may be, but is not limited to, at least two of temperature data, humidity data, or weight data.
[0164] Then, after obtaining the multiple sensor data corresponding to the current step, the mobile terminal can, but is not limited to, analyze the step time of the current step according to the multiple sensor data, and obtain the step progress of the current step by calculating the ratio between the step time and the cooking time corresponding to the current step. It can be understood that, taking the multiple sensor data including temperature data and humidity data, and the current step of simmering for 20 minutes on low heat as an example, when the mobile terminal analyzes the step time of the current step according to the two sensor data, it can specifically filter out part of the temperature data in the standard temperature range corresponding to the low heat from the acquired temperature data, and use the time interval corresponding to the part of the temperature data as the current execution time corresponding to the simmering cooking step; and it can also filter out part of the humidity data in the standard humidity range corresponding to the low heat from the acquired humidity data, and use the time interval corresponding to the part of the humidity data as the current execution time corresponding to the simmering cooking step; then, the two current execution times mentioned above can be averaged to obtain the step time of the simmering cooking step, but it is not limited to this.
[0165] Here, when it is detected that the step progress of the current step is close to the progress of the user actually executing the current step (or the difference between the two is within a preset difference interval), the mobile terminal can also send the step progress of the current step to the smart cooker, so that the smart cooker can show the step progress of the current step to the user, and the mobile terminal can continue to show the cooking ingredients, cooking equipment, cooking method and cooking time data corresponding to the current step to the user at the same time; if there are other cooking steps executed in parallel, the cooking ingredients, cooking equipment, cooking method and cooking time data corresponding to the current step and the cooking ingredients, cooking equipment, cooking method and cooking time data corresponding to other cooking steps executed in parallel can also be broadcasted in a voice loop. It can also be understood that the progress of the user actually executing the current step can be the ratio between the time spent by the user actually executing the current step and the cooking time corresponding to the current step.
[0166] Of course, in the embodiment of the present application, the user can also, but is not limited to, manually input or automatically select the completion status (i.e., step progress) corresponding to the current step in the current step completion status interface presented by the smart stove, and the smart stove will feed back the completion status to the mobile terminal, so that the mobile terminal can determine whether the completion status is close to the progress of the user's actual execution of the current step.
[0167] Step 610: When an abnormality is detected in the step progress, at least one target step is selected from the multiple cooking steps after the current step, and the step optimization results are updated according to all the target steps.
[0168] Specifically, when it is detected that the step progress is significantly different from the progress of the user actually executing the current step (or the difference between the two is not within a preset difference range), or when there is a data anomaly in any at least one sensor data, it indicates that there is an abnormality in the progress of the current step. Then, the mobile terminal can filter out one or more target steps from the multiple cooking steps after the current step according to the priority score or the step time, for example but not limited to filtering out one or more target steps with larger priority scores, or filtering out one or more target steps with shorter step times, and can update the step optimization results according to all the target steps so that all the target steps are before the current step, that is, the user executes the current step after executing all the target steps.
[0169] Of course, in an embodiment of the present application, the mobile terminal can also, but is not limited to, filter out one or more cooking steps that do not have any cooking equipment conflicts with the current step from multiple cooking steps with a parallel relationship as target steps, and can update the step optimization results based on all target steps so that all target steps are executed synchronously with the current step, and is not limited to this.
[0170] It should be noted that after updating the step optimization results, the mobile terminal can also reallocate time for each cooking step in the updated step optimization results according to the current time, so as to optimize the overall cooking efficiency and accuracy by dynamically adjusting the execution time of each cooking step.
[0171] Step 612: Display the step optimization results after the update processing to the user.
[0172] Specifically, after updating the step optimization results, the mobile terminal can also synchronously display to the user the updated step optimization results, as well as data such as cooking ingredients, cooking equipment, cooking methods and cooking time corresponding to the cooking steps at the current moment, so as to adjust the step optimization results and time schedule in real time according to the progress of the current step.
[0173] See also Figure 7 , Figure 7 A schematic structural diagram of a cooking step sequencing device based on multiple recipes provided in an embodiment of the present application is shown.
[0174] like Figure 7 As shown, the cooking step sequencing device based on multiple recipes may include at least a step identification module 701, a step sequencing module 702 and a step optimization module 703, wherein:
[0175] A step identification module 701 is used to extract a step set corresponding to a cooking type from at least two target recipes, and to identify each cooking step in the step set;
[0176] A step sorting module 702 is used to determine the dependency relationship between any two cooking steps based on the identification result of each cooking step, and sort all cooking steps according to all the dependency relationships;
[0177] The step optimization module 703 is used to identify at least two cooking parameters from each cooking step, and optimize the step sorting results according to all cooking parameters to obtain step optimization results.
[0178] In some possible embodiments, before extracting the step set corresponding to the cooking type from the at least two target recipes, the method further includes:
[0179] Performing identification processing on each preset recipe in a preset recipe database to obtain corresponding at least two recipe parameters; wherein the preset recipe database includes at least two preset recipes;
[0180] Cleaning the preset recipe database according to all recipe parameters corresponding to each preset recipe;
[0181] At least two target recipes are determined from the preset recipe data after the cleaning process.
[0182] In some possible embodiments, before determining at least two target recipes from the preset recipe data after the cleaning process, the method further includes:
[0183] Extracting each preset recipe in the preset recipe data after the cleaning process to obtain corresponding at least two keywords;
[0184] According to all the keywords corresponding to each preset recipe, the cleaned preset recipe database is standardized;
[0185] At least two target recipes are determined from the preset recipe data after the cleaning process, including:
[0186] At least two target recipes are determined from the preset recipe data after the standardization process.
[0187] In some possible embodiments, extracting a step set corresponding to a cooking type from at least two target recipes includes:
[0188] Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample;
[0189] Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample;
[0190] At least two cooking steps are extracted from the corresponding target recipe based on the classification result of each step, and a step set corresponding to the cooking type is obtained according to all the cooking steps.
[0191] In some possible embodiments, each cooking step in the step set is identified, including:
[0192] Identify each cooking step in the step set and extract the corresponding cooking action;
[0193] The cooking steps corresponding to each cooking action are numbered according to a preset numbering method.
[0194] In some possible embodiments, based on the identification result of each cooking step, determining the dependency relationship between any two cooking steps includes:
[0195] Based on the arrangement order between any two adjacent cooking steps in each target recipe, determining the order relationship between the corresponding two cooking actions;
[0196] Matching all cooking actions based on a preset parallel-action database to obtain at least one set of cooking action sets; wherein the preset parallel-action database includes at least two sets of cooking action sets with a parallel relationship, and each set of cooking action sets includes two cooking actions;
[0197] According to all sequential relations and the two cooking steps corresponding to each sequential relation, and all parallel relations and the two cooking steps corresponding to each parallel relation, a dependency relation between any two cooking steps is obtained.
[0198] In some possible embodiments, all cooking steps are sorted according to all dependencies, including:
[0199] Construct a directed acyclic graph according to all sequential relations and two cooking steps corresponding to each sequential relation, all parallel relations and two cooking steps corresponding to each parallel relation, and the numbering result of each cooking step;
[0200] The directed acyclic graph is sorted based on a topological sorting algorithm to obtain a step sorting result.
[0201] In some possible embodiments, the at least two cooking parameters include cooking time, cooking tools, and cooking methods;
[0202] The step sorting results are optimized according to all cooking parameters to obtain step optimization results, including:
[0203] The cooking time, cooking tools and cooking methods corresponding to each cooking step are scored respectively, and the three scores are weighted and summed based on the preset weights to obtain the corresponding priority score;
[0204] Based on the priority score corresponding to each cooking step, the step sorting results are sorted, and according to the cooking time and cooking tools corresponding to each cooking step, the sorted step sorting results are optimized to obtain step optimization results.
[0205] In some possible embodiments, after optimizing the step sorting results according to all cooking parameters to obtain the step optimization results, the method further includes:
[0206] Acquire at least two sensor data corresponding to the current step, and determine the step progress according to the cooking time of the current step and all sensor data; wherein the current step is any cooking step;
[0207] Show the user the progress of the steps; or
[0208] When an abnormality is detected in the step progress, at least one target step is selected from multiple cooking steps after the current step, and the step optimization result is updated according to all the target steps;
[0209] Show the user the optimization results of the updated steps.
[0210] See also Figure 8 , Figure 8 A schematic structural diagram of a mobile terminal provided in an embodiment of the present application is shown.
[0211] like Figure 8 As shown, the mobile terminal 800 may include at least one processor 801 , at least one network interface 804 , a user interface 803 , a memory 805 , and at least one communication bus 802 .
[0212] The communication bus 802 may be used to realize the connection and communication among the above-mentioned components.
[0213] The user interface 803 may include buttons, and the optional user interface may also include a standard wired interface or a wireless interface.
[0214] The network interface 804 may include, but is not limited to, a Bluetooth module, an NFC module, a Wi-Fi module, etc.
[0215] Among them, the processor 801 may include one or more processing cores. The processor 801 uses various interfaces and lines to connect various parts in the mobile terminal 800, and executes various functions and processes data of the routing mobile terminal 800 by running or executing instructions, programs, code sets or instruction sets stored in the memory 805, and calling data stored in the memory 805. Optionally, the processor 801 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 801 can integrate one or a combination of CPU, GPU, modem, etc. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 801, and it can be implemented separately through a chip.
[0216] The memory 805 may include RAM or ROM. Optionally, the memory 805 includes a non-transitory computer-readable medium. The memory 805 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 805 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 805 may optionally be at least one storage device located away from the aforementioned processor 801. As Figure 8 As shown, the memory 805 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a cooking step sequencing application based on multiple recipes.
[0217] Specifically, the processor 801 may be used to call a cooking step sequencing application based on multiple recipes stored in the memory 805, and specifically perform the following operations:
[0218] Extracting a step set corresponding to the cooking type from at least two target recipes, and marking each cooking step in the step set;
[0219] Based on the identification result of each cooking step, the dependency relationship between any two cooking steps is determined, and all cooking steps are sorted according to all the dependency relationships;
[0220] At least two cooking parameters are identified from each cooking step, and the step sorting results are optimized according to all the cooking parameters to obtain step optimization results.
[0221] In some possible embodiments, before extracting the step set corresponding to the cooking type from the at least two target recipes, the method further includes:
[0222] Performing identification processing on each preset recipe in a preset recipe database to obtain corresponding at least two recipe parameters; wherein the preset recipe database includes at least two preset recipes;
[0223] Cleaning the preset recipe database according to all recipe parameters corresponding to each preset recipe;
[0224] At least two target recipes are determined from the preset recipe data after the cleaning process.
[0225] In some possible embodiments, before determining at least two target recipes from the preset recipe data after the cleaning process, the method further includes:
[0226] Extracting each preset recipe in the preset recipe data after the cleaning process to obtain corresponding at least two keywords;
[0227] According to all the keywords corresponding to each preset recipe, the cleaned preset recipe database is standardized;
[0228] At least two target recipes are determined from the preset recipe data after the cleaning process, including:
[0229] At least two target recipes are determined from the preset recipe data after the standardization process.
[0230] In some possible embodiments, extracting a step set corresponding to a cooking type from at least two target recipes includes:
[0231] Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample;
[0232] Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample;
[0233] At least two cooking steps are extracted from the corresponding target recipe based on the classification result of each step, and a step set corresponding to the cooking type is obtained according to all the cooking steps.
[0234] In some possible embodiments, each cooking step in the step set is identified, including:
[0235] Identify each cooking step in the step set and extract the corresponding cooking action;
[0236] The cooking steps corresponding to each cooking action are numbered according to a preset numbering method.
[0237] In some possible embodiments, based on the identification result of each cooking step, determining the dependency relationship between any two cooking steps includes:
[0238] Based on the arrangement order between any two adjacent cooking steps in each target recipe, determining the order relationship between the corresponding two cooking actions;
[0239] Matching all cooking actions based on a preset parallel-action database to obtain at least one set of cooking action sets; wherein the preset parallel-action database includes at least two sets of cooking action sets with a parallel relationship, and each set of cooking action sets includes two cooking actions;
[0240] According to all sequential relations and the two cooking steps corresponding to each sequential relation, and all parallel relations and the two cooking steps corresponding to each parallel relation, a dependency relation between any two cooking steps is obtained.
[0241] In some possible embodiments, all cooking steps are sorted according to all dependencies, including:
[0242] Construct a directed acyclic graph according to all sequential relations and two cooking steps corresponding to each sequential relation, all parallel relations and two cooking steps corresponding to each parallel relation, and the numbering result of each cooking step;
[0243] The directed acyclic graph is sorted based on a topological sorting algorithm to obtain a step sorting result.
[0244] In some possible embodiments, the at least two cooking parameters include cooking time, cooking tools, and cooking methods;
[0245] The step sorting results are optimized according to all cooking parameters to obtain step optimization results, including:
[0246] The cooking time, cooking tools and cooking methods corresponding to each cooking step are scored respectively, and the three scores are weighted and summed based on the preset weights to obtain the corresponding priority score;
[0247] Based on the priority score corresponding to each cooking step, the step sorting results are sorted, and according to the cooking time and cooking tools corresponding to each cooking step, the sorted step sorting results are optimized to obtain step optimization results.
[0248] In some possible embodiments, after optimizing the step sorting results according to all cooking parameters to obtain the step optimization results, the method further includes:
[0249] Acquire at least two sensor data corresponding to the current step, and determine the step progress according to the cooking time of the current step and all sensor data; wherein the current step is any cooking step;
[0250] Show the user the progress of the steps; or
[0251] When an abnormality is detected in the step progress, at least one target step is selected from multiple cooking steps after the current step, and the step optimization result is updated according to all the target steps;
[0252] Show the user the optimization results of the updated steps.
[0253] The embodiment of the present specification also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a computer or a processor, the computer or the processor executes the above Figure 2 or Figure 5 or Figure 6 One or more steps in the illustrated embodiment. If the components of the electronic device described above are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.
Claims
1. A cooking step sequencing method based on multiple recipes, characterized in that: include: Extracting a step set corresponding to a cooking type from at least two target recipes, and marking each cooking step in the step set; Based on the identification result of each cooking step, determining the dependency relationship between any two cooking steps, and sorting all the cooking steps according to all the dependency relationships; Identifying at least two cooking parameters from each of the cooking steps, and optimizing the step sorting results according to all the cooking parameters to obtain step optimization results; The step of extracting a set of steps corresponding to the cooking type from at least two target recipes includes: Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample; Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample; wherein the step classification label includes a cooking label and a preparation label; extracting at least two cooking steps from the corresponding target recipe based on each step classification result, and obtaining a step set corresponding to the cooking type according to all the cooking steps; Wherein, at least two cooking parameters include cooking time, cooking tools and cooking method; The step sorting results are optimized according to all the cooking parameters to obtain step optimization results, including: The cooking time, the cooking tools and the cooking method corresponding to each cooking step are scored respectively, and the three scores are weighted summed based on preset weights to obtain corresponding priority scores; Based on the priority score corresponding to each cooking step, the step sorting results are sorted, and according to the cooking time and the cooking tools corresponding to each cooking step, the step sorting results after sorting are optimized to obtain step optimization results.
2. The method according to claim 1, characterized in that Before extracting a set of steps corresponding to the cooking type from at least two target recipes, the method further includes: Performing identification processing on each preset recipe in a preset recipe database to obtain corresponding at least two recipe parameters; wherein the preset recipe database includes at least two preset recipes; The preset recipe database is cleaned according to all the recipe parameters corresponding to each preset recipe; and at least two target recipes are determined from the preset recipe database after the cleaning process.
3. The method according to claim 2, characterized in that Before determining at least two target recipes from the preset recipe data after the cleaning process, the method further includes: Extracting each of the preset recipes in the preset recipe data after the cleaning process to obtain corresponding at least two keywords; According to all the keywords corresponding to each of the preset recipes, the cleaned preset recipe database is standardized; The step of determining at least two target recipes from the preset recipe data after the cleaning process includes: At least two target recipes are determined from the preset recipe data after the standardization process.
4. The method according to claim 1, characterized in that: The step of marking each cooking step in the step set includes: Identify each cooking step in the step set and extract corresponding cooking actions; The cooking steps corresponding to each cooking action are numbered according to a preset numbering method.
5. The method according to claim 4, characterized in that The step of determining the dependency relationship between any two cooking steps based on the identification result of each cooking step includes: Based on the arrangement order between any two adjacent cooking steps in each of the target recipes, determining the order relationship between the corresponding two cooking actions; Matching all the cooking actions based on a preset parallel-action database to obtain at least one set of cooking action sets; wherein the preset parallel-action database includes at least two sets of cooking action sets with a parallel relationship, and each set of the cooking action sets includes two cooking actions; According to all the sequential relationships and the two cooking steps corresponding to each sequential relationship, and all the parallel relationships and the two cooking steps corresponding to each parallel relationship, a dependency relationship between any two cooking steps is obtained.
6. The method according to claim 5, characterized in that The step of sorting all the cooking steps according to all the dependencies comprises: Constructing a directed acyclic graph according to all the sequential relationships and the two cooking steps corresponding to each of the sequential relationships, all the parallel relationships and the two cooking steps corresponding to each of the parallel relationships, and the numbering result of each of the cooking steps; The directed acyclic graph is sorted based on a topological sorting algorithm to obtain a step sorting result.
7. The method according to claim 1, characterized in that After optimizing the step sorting results according to all the cooking parameters to obtain the step optimization results, the method further includes: Acquire at least two sensor data corresponding to the current step, and determine the step progress according to the cooking time of the current step and all the sensor data; wherein the current step is any one of the cooking steps; Show the user the progress of said steps; or When an abnormality is detected in the step progress, at least one target step is selected from the plurality of cooking steps after the current step, and the step optimization result is updated according to all the target steps; The optimization results of the steps after the update process are displayed to the user.
8. A cooking step sequencing device based on multiple recipes, characterized in that: include: A step identification module, used to extract a step set corresponding to a cooking type from at least two target recipes, and to identify each cooking step in the step set; a step sorting module, configured to determine a dependency relationship between any two cooking steps based on an identification result of each cooking step, and sort all the cooking steps according to all the dependency relationships; a step optimization module, used for identifying at least two cooking parameters from each of the cooking steps, and optimizing the step sorting results according to all the cooking parameters to obtain a step optimization result; The step of extracting a set of steps corresponding to the cooking type from at least two target recipes includes: Input all steps included in each target recipe into a preset language processing model to obtain a corresponding step vector; wherein the preset language processing model is trained by at least two step samples and a vector sample corresponding to each step sample; Inputting the step vector of each target recipe into a preset deep learning model to obtain a corresponding step classification result; wherein the preset deep learning model is trained by at least two vector samples and a step classification label corresponding to each vector sample; wherein the step classification label includes a cooking label and a preparation label; extracting at least two cooking steps from the corresponding target recipe based on each step classification result, and obtaining a step set corresponding to the cooking type according to all the cooking steps; Wherein, at least two cooking parameters include cooking time, cooking tools and cooking method; The step sorting results are optimized according to all the cooking parameters to obtain step optimization results, including: The cooking time, the cooking tools and the cooking method corresponding to each cooking step are scored respectively, and the three scores are weighted summed based on preset weights to obtain corresponding priority scores; Based on the priority score corresponding to each cooking step, the step sorting results are sorted, and according to the cooking time and the cooking tools corresponding to each cooking step, the step sorting results after sorting are optimized to obtain step optimization results.
Citation Information
Patent Citations
Multi-dish cooking control method, multi-dish control system and cooking utensil
CN112369896A
Automatic cooking method and system, intelligent cooking equipment and intelligent oven
CN105981999A
Intelligent menu adaptation method and device, computer equipment and storage medium
CN118035429A
Cooking task management method and system and computer program product
CN118428705A
