An oven intelligent control system, method and device

Abstract

The present application relates to the technical field of intelligent control, and discloses a kind of oven intelligent control system, method and device, its method includes: acquisition module: the regional taste preference data of each food category of each division area is collected;Drawing module: determine the category preference data of multiple taste labels of each division area of each food category and taste representation graph;Determination module: determine the feature approximation factor of each quantification feature of each taste label except original type and division label of each division area of each food category;Recommendation module: construct the taste recommendation model of each food category;Control module: determine recommended taste data, generate control instruction and realize cooking control. Can accurately adapt to the differentiated taste needs of different growth background and different regional population, weaken the adaptation deviation brought by regional diet difference, improve the pertinence and accuracy of oven taste recommendation, enhance the individualization and intelligent level of food processing control.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent control system, method and device for a baking oven. Background Technology

[0002] Early ovens were primarily mechanical, offering only basic temperature and timer control without any adaptation to specific food tastes. With advancements in electronics, electronic ovens became increasingly common, allowing for pre-programmed cooking programs for a limited number of fixed ingredients, but these settings remained generic. Recent rapid development of smart home appliances has led to the emergence of ovens with simple regional flavor profiles. However, these rely solely on general local statistics, failing to consider the differences in taste preferences resulting from users' life experiences. This crude segmentation and limited analytical dimensions cannot meet the refined taste preferences of mobile populations across regions. The current smart home appliance industry is rapidly shifting towards personalized customization, urgently requiring the development of smart kitchen appliances with special uses and suitability for specific groups. Existing oven technology struggles to meet this demand for high-end, personalized products.

[0003] Therefore, the present invention proposes an intelligent control system, method and apparatus for an oven. Summary of the Invention

[0004] This invention provides an intelligent control system, method, and apparatus for an oven. It sets flavor dimension data for each food category in the cooking category data of the target oven, collects regional flavor preference data for each divided area based on the flavor dimension data set for each food category, determines the category preference data of multiple flavor tags for each divided area of ​​each food category, draws a flavor representation map of each flavor tag for each divided area of ​​each food category, determines the feature proximity factor and division label of each quantitative feature of each flavor tag (excluding the original type) for each divided area of ​​each food category, constructs a flavor recommendation model for each food category, collects user data of the target users of the target oven, obtains data on the ingredients to be cooked and the selected flavors, determines the recommended flavor data, generates control commands, and realizes cooking control. This allows for precise quantitative discrimination of flavor characteristics, refined stratification of flavor attributes among different groups, accurate adaptation to the differentiated taste needs of people with different backgrounds and regions, weakens the adaptation bias caused by regional dietary differences, improves the targeting and accuracy of oven flavor recommendations, and enhances the personalization and intelligence level of food processing control.

[0005] This invention provides an intelligent control system for an oven, comprising: Data Acquisition Module: Sets flavor dimension data for each food category in the cooking category data of the target oven, and collects regional flavor preference data for each divided area based on the flavor dimension data set for each food category; The plotting module: Based on the regional taste preference data of each segmented region for each food category, it determines the category preference data of multiple taste labels for each segmented region for each food category, and plots the taste representation map of each taste label for each segmented region for each food category; Determine module: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantified feature of each flavor tag in each segmentation region of each food category, excluding the original type; Recommendation module: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), a flavor recommendation model is constructed for each food category. Control module: Collects user data of the target user of the target oven, obtains the ingredients to be cooked and the selected flavor data, determines the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generates control instructions based on the selected flavor data and the recommended flavor data and realizes cooking control.

[0006] Preferably, an intelligent control system for an oven includes a data acquisition module, comprising: Cooking category data unit: Based on the food attributes of the culinary ingredients of the target oven, determine the cooking category data of the target oven. The oven cooking category data includes at least meat, seafood, vegetables, pasta products, and fruits. Regional division: Based on administrative region data and food culture data, multiple regional divisions are determined; Setting Unit: Based on each food category in the cooking category data of the target oven, set the flavor dimension data, where the flavor dimension data includes multiple quantitative features and the color representing each quantitative feature; Regional taste preference data unit: Based on the taste dimension data set for each food category in the cooking category data of the target oven, regional taste preference data for each divided region is collected. The regional taste preference data includes individual growth data and taste preference data of multiple consumers. The individual growth data includes at least the primary residence for children aged 0-3 years, the primary residence for children aged 3-12 years, the age of migration to the divided region, and the length of stay. The taste preference data includes preference values ​​of multiple quantitative features.

[0007] Preferably, an intelligent control system for an oven includes a drawing module, comprising: Bottom-level taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of each consumer aged 0-3 years is determined by comparing it with all the regional divisions, and the regional division containing the primary residence of each consumer aged 0-3 years is determined as the bottom-level taste region of the consumer. Cultural taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of the 3-12-year-olds in the individual growth data is compared with all the regional divisions to determine the division containing the primary residence of the 3-12-year-olds as the cultural taste region of the consumer. Native type unit: If the underlying taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is consistent with the segmentation region, then the taste label of the consumer is determined to be native type; Bottom-level fusion type unit: If the bottom-level taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is inconsistent with the segmentation region, the taste label of the consumer is determined to be bottom-level fusion type. Culturally integrated unit: If the underlying taste region of each consumer in the regional taste preference data of each region of each food category is inconsistent with the region, but the cultural taste regions are consistent with the region, the taste label of the consumer is determined to be culturally integrated. Foreign type unit: If the underlying taste region and cultural taste region of each consumer in the regional taste preference data of each region of each food category are inconsistent with the region, the taste label of the consumer is determined to be foreign type. Category preference data unit: Based on the taste tags of all consumers in the regional taste preference data of each segmented region of each food category, the regional taste preference data of the segmented region is divided to determine the category preference data of multiple taste tags in each segmented region of each food category. The category preference data includes individual growth data and taste preference data of multiple consumers.

[0008] Preferably, in an intelligent control system for an oven, the drawing module further includes: Dot diameter unit: Based on the immigration age in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region of each food category, and the first adaptive mapping curve, determine the dot diameter of each consumer in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region. Color depth value unit: Based on the life duration in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category, the color of each quantitative feature in the taste dimension data and the second adaptive mapping curve are used to determine the color depth value of each consumer in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category based on each quantitative feature; Standard unit: The taste label of each segment of each food category is determined to be the standard diameter of the dot of each consumer in the native category preference data. At the same time, the color depth value of each quantized feature of each consumer in the native category preference data is determined to be the standard depth value of the taste label of each segment of each food category. The first representation dot unit: Based on the representation color and standard diameter and standard depth values ​​of each quantified feature in the taste dimension data, the taste label of each division region of each food category is determined as the representation dot of each quantified feature of all consumers in the original type; The second representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the culturally integrated category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the culturally integrated representation dot. The third representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the bottom-level fusion type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the bottom-level fusion type. The fourth representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the foreign type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the foreign type. Box plot unit: Based on the preference values ​​of all quantitative features in the category preference data of all consumers in the taste preference data of each flavor label in each segment of each food category, a box plot is drawn for each flavor label in each segment of each food category. The horizontal axis of the box plot represents all quantitative features, and the vertical axis represents the preference value of the quantitative features. Flavor representation map unit: Based on the preference value of each consumer's quantitative feature for each flavor label in each segmented region of each food category, the representation dots of the consumer's quantitative feature are plotted on the box plot to determine the flavor representation map for each flavor label in each segmented region of each food category, wherein the flavor representation map includes flavor representation sub-maps of multiple quantitative features.

[0009] Preferably, an intelligent control system for an oven includes a determining module comprising: The inclusion rate unit: For each region of each food category, the original flavor representation map of the flavor label is coaxially superimposed with the flavor representation map of each flavor label other than the original type. The number of representation dots between the upper and lower quartiles of the flavor representation sub-map of each quantitative feature of each flavor label other than the original type and the flavor representation sub-map of each quantitative feature of each flavor label of the original type are counted. Based on the number of representation dots and the number of consumers in the corresponding flavor label category preference data, the inclusion rate of each quantitative feature of each flavor label other than the original type in each region of each food category is calculated. Centroid offset value unit: Based on the median and box width of the flavor representation submap of each quantized feature of each flavor label in each partition region of each food category with the flavor label as the native type, and the median of the flavor representation submap of each quantized feature of each flavor label other than the native type, calculate the centroid offset value of each quantized feature of each flavor label in each partition region of each food category, excluding the native type. Feature proximity factor unit: Based on the fall rate and centroid offset value of each quantized feature of each flavor label in each partition region of each food category (excluding the original type), calculate the feature proximity factor of each quantized feature of each flavor label in each partition region of each food category (excluding the original type); Morphological labeling unit: For each flavor label in each segmented region of each food category, excluding the original type, a morphological judgment is performed on the flavor representation sub-image of each quantified feature. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label fall evenly and densely into the flavor representation sub-image of the quantified feature of the original type according to the dot diameter and color depth value, the morphological label of the quantified feature of the flavor label is determined to be uniformly close. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label with large dot diameter and deep color depth value fall densely into the flavor representation sub-image of the quantified feature of the original type, and the representation dots with small dot diameter and light color depth value are densely distributed at the edge of the flavor representation sub-image of the quantified feature of the original type, the morphological label of the quantified feature of the flavor label is determined to be gradient close. First labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is greater than the first preset threshold, and the morphological label is uniformly close, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Second labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is between the second preset threshold and the first preset threshold, and the morphological label is gradient proximity, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Third segmentation label unit: Otherwise, determine the segmentation label of the quantitative feature of the flavor label as a segmentation.

[0010] Preferably, an intelligent control system for an oven includes the following modules: Clustering Unit: Based on the quantitative features of each flavor tag (excluding the original type) for each partition region of each food category, and the preference values ​​of the quantitative features in the taste preference data of all consumers in the category preference data of the flavor tags, cluster analysis is performed on all consumers in the category preference data of the flavor tags to determine multiple feature clustering data of each flavor tag (excluding the original type) for each partition region of each food category. The feature clustering data includes the quantitative values ​​of multiple consumers and the characterization dots. Representation clustering graph unit: Based on the feature clustering data of each flavor label (excluding the original type) of each partition region of each food category, a representation clustering graph is drawn for each cluster label of each flavor label (excluding the original type) of each partition region of each food category, taking the quantified feature of each partition as the partition. Clustering proximity factor unit: Based on the clustering graph representing each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category, and the flavor representation subgraph of the quantification feature, calculate the clustering proximity factor of each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category. Flavor proximity data unit: Based on the feature proximity factors of all quantized features of flavor labels other than the original type for each region of each food category with no division, and the cluster proximity factors of all cluster labels with quantized features of division, the flavor proximity data of each flavor label other than the original type for each region of each food category is determined. Original quantization vector unit: Based on the original flavor representation map of each segmentation region of each food category, the original quantization vector of each segmentation region of each food category is determined. The original quantization vector includes the original quantization values ​​of multiple quantization features. Flavor Recommendation Model Unit: Based on the flavor proximity data of each flavor tag in all regions of each food category (excluding the original type), and the original quantized vector of the flavor tag as the original type, a flavor recommendation model is constructed for each food category.

[0011] Preferably, an intelligent control system for an oven includes a control module comprising: User data unit: Collects user data of the target users of the target oven, wherein the user data includes at least the address of use, the main residence of children aged 0-3 years, the main residence of children aged 3-12 years, the age of migration to the address of use, and the duration of residence; Unit Determination: The division of regions is determined based on the usage address in the user data of the target users; the basic taste region is determined based on the main residence of children aged 0-3 in the user data of the target users; and the cultural taste region is determined based on the main residence of children aged 3-12 in the user data of the target users. Recommended Flavor Data Unit: Obtain the ingredients to be cooked by the target user using the target oven, input the target user's regional segmentation, basic flavor region, cultural flavor region, age of migration, and duration of residence into the flavor recommendation model of the food category corresponding to the ingredients to be cooked, determine the recommended flavor data of the target user based on the ingredients to be cooked, and display it in the visualization area of ​​the target oven. The recommended flavor data includes recommended quantitative values ​​of multiple quantitative features. Flavor Selection Data Unit: Acquires the flavor selection data of the target user, which includes the selection quantification values ​​of multiple quantification features; Control unit: Inputs the target user's recommended taste data and selected taste data into the oven control model, generates control instructions based on the ingredients to be cooked for the target user, and executes the control instructions to achieve cooking control.

[0012] This invention provides an intelligent control method for an oven, used to execute any one of the intelligent control systems for an oven in Examples 1 to 7, comprising: S1: Set flavor dimension data for each food category in the cooking category data of the target oven, and collect regional flavor preference data for each divided area based on the flavor dimension data set for each food category; S2: Based on the regional taste preference data of each division of each food category, determine the category preference data of multiple taste labels in each division of each food category, and draw a taste representation map of each taste label in each division of each food category; S3: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantitative feature of each flavor tag in each segmentation region of each food category, excluding the original type; S4: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), construct a flavor recommendation model for each food category; S5: Collect user data of the target user of the target oven, obtain the ingredients to be cooked and the selected flavor data, determine the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generate control instructions based on the selected flavor data and the recommended flavor data, and realize cooking control.

[0013] The present invention provides an intelligent oven control device for executing any one of the intelligent oven control systems in Examples 1 to 7.

[0014] The beneficial effects of this invention compared to existing technologies are as follows: By setting flavor dimension data for each food category in the cooking category data of the target oven, collecting regional flavor preference data for each segmented area based on the flavor dimension data set for each food category, determining the category preference data of multiple flavor tags for each segmented area of ​​each food category, drawing a flavor representation map of each flavor tag for each segmented area of ​​each food category, determining the feature proximity factor and segmentation label of each quantitative feature of each flavor tag (excluding the original type) for each segmented area of ​​each food category, constructing a flavor recommendation model for each food category, collecting user data of the target users of the target oven, obtaining the ingredients to be cooked and the selected flavor data, determining the recommended flavor data, generating control commands, and realizing cooking control. This allows for precise quantitative discrimination of flavor features, refined stratification of flavor attributes for different groups, accurate adaptation to the differentiated flavor needs of people with different backgrounds and regions, weakening the adaptation bias caused by regional dietary differences, improving the targeting and accuracy of oven flavor recommendations, and enhancing the personalization and intelligence level of food processing control. This invention aligns with the development trend of smart home appliances, enabling personalized cooking control of the oven, meeting the exclusive dietary needs of special purposes and specific groups, and enriching the supply of high-end smart home appliances.

[0015] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of an intelligent control system for an oven according to an embodiment of the present invention; Figure 2 This is a flowchart of an intelligent oven control method according to an embodiment of the present invention. Detailed Implementation

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention. Example 1:

[0019] This invention provides an intelligent control system for an oven, as shown in the reference. Figure 1 ,include: Data Acquisition Module: Sets flavor dimension data for each food category in the cooking category data of the target oven, and collects regional flavor preference data for each divided area based on the flavor dimension data set for each food category; The plotting module: Based on the regional taste preference data of each segmented region for each food category, it determines the category preference data of multiple taste labels for each segmented region for each food category, and plots the taste representation map of each taste label for each segmented region for each food category; Determine module: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantified feature of each flavor tag in each segmentation region of each food category, excluding the original type; Recommendation module: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), a flavor recommendation model is constructed for each food category. Control module: Collects user data of the target user of the target oven, obtains the ingredients to be cooked and the selected flavor data, determines the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generates control instructions based on the selected flavor data and the recommended flavor data and realizes cooking control.

[0020] In this embodiment, the data acquisition module is responsible for constructing the basic data framework required for system operation. The target oven's cooking category data refers to the classification of all food types that the oven can handle. For each food category, the system sets a set of flavor dimension data, which includes multiple quantitative features and a color assigned to each feature. Each quantitative feature is preset with a color for subsequent visualization. Based on this, the data acquisition module collects a large amount of consumer feedback on each of the aforementioned food categories for each pre-constructed segmented region. This feedback strictly adheres to the pre-defined flavor dimensions, assigning preference scores.

[0021] In this embodiment, the division of regions involves spatial overlay analysis of administrative region data and food culture data. The administrative region data provides a spatial framework with clear boundaries at the provincial, municipal, and county levels, while the food culture data marks the mainstream seasoning styles and cooking tendencies of different regions based on the recognized culinary system divisions. The administrative culture basic unit obtained by taking the intersection of the two is the division of regions.

[0022] In this embodiment, the rendering module processes and visualizes the collected regional taste preference data. Within each region, it categorizes taste preferences into multiple tags based on the origin of the consumers: Native (referring to long-term residents whose birthplace and cultural taste development are consistent with the region); Immigrant (referring to groups who migrated from other food culture regions); Underlying Integration (referring to individuals aged 0-3 who lived in the region but whose tastes were formed elsewhere); and Cultural Integration (referring to individuals who spent their childhood elsewhere but whose tastes were formed in the region). For a given food category and region, preference data for each category is determined. Then, based on this data, a taste representation map for each taste tag is drawn.

[0023] In this embodiment, the core task of the determination module is to quantify the degree of similarity in taste between non-native taste labels (such as those for foreign, bottom-level integration, or cultural integration) and the local native baseline, and to determine whether further segmentation of the group is necessary. It utilizes the native taste representation map generated by the drawing module and the taste representation maps of non-native taste labels, coaxially superimposing them on the same coordinate system. For each quantified feature, it calculates the proportion of the non-native group's representation dots falling within the core region between the upper and lower quartiles of the native box plot, obtaining the inclusion rate. This rate directly reflects the proportion of a foreign group whose taste preferences in a certain dimension are indistinguishable from those of the locals. Simultaneously, the determination module extracts the median of the native group for each quantified feature as the origin reference and extracts the box width of the native box plot as the distance scale unit, measuring the degree of deviation between the median of the non-native group and the median of the native group, obtaining the centroid offset value. The feature proximity factor is a comprehensive proximity score calculated by combining two complementary indicators: the fall rate and the centroid offset value. A high fall rate and a small centroid offset value indicate that the group is highly compatible with the native taste, resulting in a large feature proximity factor, and vice versa. The determination module also assists in the judgment by observing the spatial distribution of the representative dots of the non-native group. If the dots are evenly distributed and densely clustered inside the native box, it is judged as uniform proximity, indicating that the group has as a whole approached the local taste. If large-diameter dark dots are concentrated in the core area while small-diameter light dots are distributed in a remote area, it is judged as gradient proximity, indicating that the group is in a dynamic integration process of intergenerational succession. When the feature proximity factor is higher than the first preset threshold and the shape is uniform proximity, or when the factor falls between the first and second preset thresholds and the shape is gradient proximity, the classification label is determined as no classification; otherwise, the classification label is determined as classification.

[0024] In this embodiment, the control module is the execution center of the entire system, responsible for translating abstract flavor target values ​​into specific control commands for each actuator of the oven during actual use. It first collects user data of the target user for the target oven, determining the user's current region, basic flavor region, and cultural flavor region. Then, the system obtains the ingredients to be cooked by the user and the flavor selection data input through a touchscreen or other interface. It inputs five personalized parameters—the user's region, basic flavor region, cultural flavor region, age of arrival, and length of stay—along with the ingredients to be cooked into the corresponding food category flavor recommendation model. The model outputs recommended flavor data including multi-dimensional quantitative recommendation values. Based on the user's selected flavor data, the system makes necessary corrections or supplements to the recommended flavor data, generating a final multi-dimensional flavor target vector. Through a pre-established flavor control mapping table, the target vector is translated into control commands for each actuator of the oven, driving the heating element, hot air fan, and steam generator to work collaboratively to achieve precise cooking.

[0025] The beneficial effects of the above technology are as follows: By setting flavor dimension data for each food category in the cooking category data of the target oven, collecting regional flavor preference data for each segmented area based on the flavor dimension data set for each food category, determining the category preference data of multiple flavor tags for each segmented area of ​​each food category, drawing a flavor representation map of each flavor tag for each segmented area of ​​each food category, determining the feature proximity factor and segmentation label of each quantitative feature of each flavor tag for each segmented area of ​​each food category (excluding the original type), constructing a flavor recommendation model for each food category, collecting user data of the target users of the target oven, obtaining data on ingredients to be cooked and selected flavors, determining recommended flavor data, generating control commands, and realizing cooking control. This enables precise quantitative discrimination of flavor features, completes refined stratification of flavor attributes for different groups, accurately adapts to the differentiated flavor needs of people with different backgrounds and regions, weakens the adaptation bias caused by regional dietary differences, improves the targeting and accuracy of oven flavor recommendations, and enhances the personalization and intelligence level of food processing control. Example 2:

[0026] Based on Example 1, an intelligent control system for an oven includes a data acquisition module, comprising: Cooking category data unit: Based on the food attributes of the culinary ingredients of the target oven, determine the cooking category data of the target oven. The oven cooking category data includes at least meat, seafood, vegetables, pasta products, and fruits. Regional division: Based on administrative region data and food culture data, multiple regional divisions are determined; Setting Unit: Based on each food category in the cooking category data of the target oven, set the flavor dimension data, where the flavor dimension data includes multiple quantitative features and the color representing each quantitative feature; Regional taste preference data unit: Based on the taste dimension data set for each food category in the cooking category data of the target oven, regional taste preference data for each divided region is collected. The regional taste preference data includes individual growth data and taste preference data of multiple consumers. The individual growth data includes at least the primary residence for children aged 0-3 years, the primary residence for children aged 3-12 years, the age of migration to the divided region, and the length of stay. The taste preference data includes preference values ​​of multiple quantitative features.

[0027] In this embodiment, the target oven refers to the specific oven equipment adapted to this intelligent control system, and the culinary ingredients refer to all types of ingredients that the oven equipment can safely and stably bake and cook. Ingredient attributes refer to the inherent attributes of the ingredients themselves, such as their biological category, texture, baking process requirements, and flavor characteristics. The cooking category data is a standardized set of classifications based on the inherent attributes of the ingredients, used to distinguish different types of ingredients. It includes at least five core categories: meat, seafood, vegetables, pasta products, and fruits, and may also include nuts, poultry, and eggs.

[0028] In this embodiment, administrative region data refers to the boundary and attribution data related to the prescribed levels of administrative divisions. Food culture data refers to the cultural characteristics data related to food flavors, cooking habits, and ingredient preferences that have formed over a long period of time in different regions, such as the eight major cuisines or further subdivided regions like the Sichuan spicy region and the Cantonese light region. Multiple division regions are the final output of this unit, a standardized set of regional units that combine administrative boundaries and food culture attributes. By spatially overlaying the administrative region data and food culture data, the intersection of the precision of administrative boundaries and the fuzzy transitional characteristics of food culture boundaries is obtained to obtain a minimum common unit as the initial division region, such as Sichuan cuisine - spicy sub-region - Sichuan Province - Chengdu City, Cantonese cuisine - light sub-region - Guangdong Province - Chaoshan cities, etc.

[0029] In this embodiment, the flavor dimension data includes multiple quantitative features, which are key sensory indicators selected to fully reflect the quality of cooking and the eating experience of this type of food. For meat, quantitative features may include surface crispness, center tenderness, juiciness, caramelization intensity, and saltiness / umami perception. For seafood, these may include skin crispness, meat elasticity, moisture retention, sweetness, and fishy smell suppression. For vegetables, these may include edge crispness, body firmness, moisture loss, caramelization degree, and original flavor retention. For pasta products, these may include outer crispness, internal softness, color uniformity, wheat aroma intensity, and melt-in-your-mouth sensation. For fruits, these may include skin integrity, flesh softness, juice overflow, sweet-sour ratio shift, and charring degree. Each quantitative feature is also assigned a color for subsequent differentiation of dimensions in the visualization representation diagram.

[0030] In this embodiment, regional taste preference data is collected from real user data in each segmented region according to a template defined by the set unit. The collection process involves having a large number of consumers, whose geographical locations are marked, rate the food according to predefined dimensions through the target oven's cloud service, paired mobile applications, or offline survey channels. Each record contains the consumer's individual growth data and current taste preference data. Individual growth data includes at least the primary residence from 0 to 3 years old, corresponding to the period of basic human taste security and physiological acceptance of sour, sweet, bitter, and salty flavors; and the primary residence from 3 to 12 years old, corresponding to the period of cultural taste shaping formed through family dishes and regional foods during socialization. It also includes the age at which the consumer moved to the current segmented region and the length of time they have lived in that region. Current taste preference data is the consumer's self-rating value for the multidimensional quantitative characteristics of the corresponding food category at this moment.

[0031] The beneficial effects of the above technology are as follows: it sets flavor dimension data for each food category in the cooking category data of the target oven, and collects regional flavor preference data for each divided area based on the flavor dimension data set for each food category. It can take into account the regional division logic of administrative boundaries and the core of food culture, and realize the full-dimensional data capture of the underlying logic of user taste formation. Example 3:

[0032] Based on Example 1, an intelligent control system for an oven includes a drawing module, comprising: Bottom-level taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of each consumer aged 0-3 years is determined by comparing it with all the regional divisions, and the regional division containing the primary residence of each consumer aged 0-3 years is determined as the bottom-level taste region of the consumer. Cultural taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of the 3-12-year-olds in the individual growth data is compared with all the regional divisions to determine the division containing the primary residence of the 3-12-year-olds as the cultural taste region of the consumer. Native type unit: If the underlying taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is consistent with the segmentation region, then the taste label of the consumer is determined to be native type; Bottom-level fusion type unit: If the bottom-level taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is inconsistent with the segmentation region, the taste label of the consumer is determined to be bottom-level fusion type. Culturally integrated unit: If the underlying taste region of each consumer in the regional taste preference data of each region of each food category is inconsistent with the region, but the cultural taste regions are consistent with the region, the taste label of the consumer is determined to be culturally integrated. Foreign type unit: If the underlying taste region and cultural taste region of each consumer in the regional taste preference data of each region of each food category are inconsistent with the region, the taste label of the consumer is determined to be foreign type. Category preference data unit: Based on the taste tags of all consumers in the regional taste preference data of each segmented region of each food category, the regional taste preference data of the segmented region is divided to determine the category preference data of multiple taste tags in each segmented region of each food category. The category preference data includes individual growth data and taste preference data of multiple consumers.

[0033] In this embodiment, the underlying taste region unit is responsible for pinpointing the origin of each consumer's most fundamental taste base throughout their life. It utilizes individual growth data from regional taste preference data to extract key information such as the primary place of residence from ages 0 to 3. Scientific research has confirmed that this age group's place of residence is a crucial period for human taste security and basic taste acceptance; the flavors encountered at this time become lifelong familiar and safe taste references for the individual. The coordinates of this place of residence are then spatially compared with all previously constructed regional divisions to determine which geographical boundary the location falls into. Once a match is found, that region is identified as the consumer's underlying taste region. The underlying taste region identifies the most basic and instinctive part of the consumer's taste spectrum; regardless of where the person later moves, this underlying taste based on initial memories will not completely disappear.

[0034] In this embodiment, the cultural taste region unit is responsible for identifying the source of the cultural imprint shaped by experience in a consumer's taste preferences. It extracts information about the primary place of residence from the age of 3 to 12 from an individual's growth data. This period corresponds to the cultural acquisition period in which humans develop stable taste preferences through family cooking, school lunches, and regional dietary environments—the stage commonly known as the lifelong preference for a particular local cuisine. Similarly, through spatial affiliation analysis, this place of residence is compared with a designated region to determine its affiliation, identifying the region containing that location and marking it as the consumer's cultural taste region. This data provides a geographical source for understanding the formation period of a consumer's current explicit taste preferences; it represents the taste homeland that the consumer is most familiar with and most likely to actively identify with.

[0035] In this embodiment, the native unit uses the two regional labels mentioned above to determine which consumers are purely locals with no foreign taste genes. The logic is that for a consumer within a specific region, if their underlying taste region is consistent with that region, and their cultural taste region is also consistent with that region, it means that they have been nurtured by the food culture of that region from birth until their taste matures, without being profoundly shaped by flavors from other regions. Such consumers are assigned a native taste label. This group represents the most authentic traditional taste specimens of their region, and their preference data directly defines the authentic flavor baseline for a certain food category in that region.

[0036] In this embodiment, the underlying fusion type unit captures a special group of people who received local flavors in their childhood but were later reshaped by foreign cultures over a long period of time. If the consumer's underlying taste region is consistent with the current classification region, it means that their initial sense of taste security is rooted in the local area, but their cultural taste region is inconsistent with the current classification region. This means that they spent the most critical period of taste formation, from 3 to 12 years old, in another food culture region, forming a dominant preference different from that of the native inhabitants of this region. These people also feel familiar with local flavors, but their truly accustomed daily tastes come from elsewhere, putting them in a contradictory state of being rooted in the local area but with leaves in a foreign land, and thus they are given the underlying fusion type label.

[0037] In this embodiment, the culturally integrated unit follows the opposite path to the fundamentally integrated unit, capturing individuals who were immersed in the flavors of other places during childhood but have long been rooted in their local community as adults. When the consumer's fundamental taste region is inconsistent with the current segmented region, it indicates that their earliest taste imprint came from another place, but their cultural taste region is consistent with the current segmented region. This suggests that the critical period for the formation of their taste was completed in their current place of residence. Such individuals' outward taste expression is highly localized, and they can skillfully use local seasoning logic to measure deliciousness, but deep down, they still harbor an instinctive affinity for the original flavors of their hometown, thus being labeled as culturally integrated.

[0038] In this embodiment, the "outsider" unit is used to identify consumers who have no culinary connection to the current categorization area and belong to a purely outsider group. The criterion is that the consumer's underlying taste region and cultural taste region are inconsistent with the current categorization area. In other words, they have spent their lives in a different food culture area from birth to the maturity of their taste, and only settled here later due to migration. The taste system of such people has been completely solidified in their original region, and they lack a deep-rooted instinctive identification with the flavors of their current residence. Their preference data is a source of deviation from the standard model of their current residence, and they are given the "outsider" label.

[0039] In this embodiment, after determining the four categories of labels, the category preference data unit reorganizes and structures the messy individual data. Based on the four categories of taste labels—native, underlying fusion, cultural fusion, and foreign—already assigned to all consumers in each segmented area, it categorizes and consolidates the originally mixed regional taste preference data for that segmented area. In this way, instead of each area having only a vague average taste, there are multiple sets of category preference data clearly separated according to the taste formation path. Each set of category preference data contains the individual growth data and taste preference data of all consumers under that label.

[0040] The beneficial effects of the above technologies are as follows: Based on the regional taste preference data of each segmented region of each food category, the category preference data of multiple taste tags of each segmented region of each food category can be determined, which can provide a high-granular user segmentation basis for personalized taste adaptation and improve the accuracy of taste inference for users with complex migration experience. Example 4:

[0041] Based on Example 3, a smart oven control system, including a drawing module, further includes: Dot diameter unit: Based on the immigration age in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region of each food category, and the first adaptive mapping curve, determine the dot diameter of each consumer in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region. Color depth value unit: Based on the life duration in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category, the color of each quantitative feature in the taste dimension data and the second adaptive mapping curve are used to determine the color depth value of each consumer in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category based on each quantitative feature; Standard unit: The taste label of each segment of each food category is determined to be the standard diameter of the dot of each consumer in the native category preference data. At the same time, the color depth value of each quantized feature of each consumer in the native category preference data is determined to be the standard depth value of the taste label of each segment of each food category. The first representation dot unit: Based on the representation color and standard diameter and standard depth values ​​of each quantified feature in the taste dimension data, the taste label of each division region of each food category is determined as the representation dot of each quantified feature of all consumers in the original type; The second representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the culturally integrated category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the culturally integrated representation dot. The third representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the bottom-level fusion type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the bottom-level fusion type. The fourth representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the foreign type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the foreign type. Box plot unit: Based on the preference values ​​of all quantitative features in the category preference data of all consumers in the taste preference data of each flavor label in each segment of each food category, a box plot is drawn for each flavor label in each segment of each food category. The horizontal axis of the box plot represents all quantitative features, and the vertical axis represents the preference value of the quantitative features. Flavor representation map unit: Based on the preference value of each consumer's quantitative feature for each flavor label in each segmented region of each food category, the representation dots of the consumer's quantitative feature are plotted on the box plot to determine the flavor representation map for each flavor label in each segmented region of each food category, wherein the flavor representation map includes flavor representation sub-maps of multiple quantitative features.

[0042] In this embodiment, the first adaptive mapping curve is based on the critical period theory in developmental psychology, data on the stability of taste memory in sensory science, and a sampling survey of a large-scale immigrant population. By measuring the average deviation between the current taste and local taste of sample users who immigrated at different ages, a scatter plot of age deviation is plotted, and then a continuous curve is fitted using a regression method, thus obtaining the quantitative relationship between age and plasticity. The diameter of the dot is directly proportional to this plasticity value, thus forming an adaptive mapping with a larger diameter for earlier immigration. For example, for immigrants aged 0 to 3 years, the dot diameter is the largest and remains constant. This is because individuals are in their peak period of taste acceptance, capable of fully internalizing local flavors as their own core taste buds, exhibiting extremely high taste plasticity. For those migrating between 3 and 12 years old, the diameter of the taste buds decreases significantly with age, as individuals gradually complete taste socialization, and the window for accepting foreign flavors narrows year by year. With each year of delayed migration, their natural ability to accept local flavors weakens. For those migrating between 12 and 18 years old, the shrinking trend of the taste buds slows, the core taste structure is largely established, and openness to new flavors tends to stabilize but remains flexible. After the age of 18, the taste buds approach a very small, fixed value. Taste preferences are deeply solidified, and migration after adulthood hardly alters the core preference structure.

[0043] In this embodiment, the diameter unit of the dot determines the size of the individual dot for three types of consumers with migrating backgrounds: culturally integrated, fundamentally integrated, and foreign. It is based on the immigration age data in the individual's growth data, that is, at what age the consumer migrated to the current segmented region. Its determination logic follows the first adaptive mapping curve.

[0044] In this embodiment, the second adaptive mapping curve is determined based on the sociological model of cultural assimilation and the psychological theory of U-shaped adaptation. It tracks and surveys the self-rated taste data of migrants with different lengths of residence, extracts the mean cosine similarity or Euclidean distance between their tastes and local standard tastes, and plots a curve showing the relationship between length of residence and taste convergence. The color depth value strictly follows this cumulative adaptation curve, achieving an adaptive performance where the longer the residence period, the deeper the color. For example, within 0 to 1 year of residence, the color depth value increases the fastest. Newcomers experience rapid, active or passive adjustments in taste due to drastic changes in daily dining, social interactions, and other environmental factors, causing lighter colors to deepen rapidly. Within a lifespan of 1 to 10 years, the rate of increase in color depth gradually slows down, showing a smooth convergence trend. At this point, the active dominant taste changes have been basically completed, and what remains is mainly unconscious, slow deep immersion. The color steadily deepens but no longer increases sharply. After a lifespan of more than 10 years, the color depth gradually approaches but never exceeds the standard depth value of the native type. This means that even if an individual has lived in the area for decades, there may still be extremely subtle and incompletely eliminated genetic differences in their taste compared to the native type who has completed their entire life cycle purely in the local area.

[0045] In this embodiment, the color depth value unit also targets these three types of consumers with migrating backgrounds, determining the intensity of their representative color across different taste dimensions for each individual. Its input is the length of life lived in the individual's growth data, i.e., how many years the consumer has resided in the currently defined area. Its determination rule follows the second adaptive mapping curve.

[0046] In this embodiment, the standard unit sets an invariant set of reference benchmarks for the native population. It directly sets the dot diameter of each native consumer to the standard diameter, a fixed constant representing a pure, unmixed, original state. Similarly, the color depth values ​​of the native population across all quantified characteristics are also uniformly set to standard depth values. This standard depth value is generally the most saturated level, meaning that these individuals prefer colors that are rich, pure, and undiluted; they themselves are the taste definers for this food category in this region. After establishing this benchmark, the dot size and color depth of all hybrid or foreign types will be referenced to the native type, allowing for a direct comparison of how far a foreigner's taste deviates from the local native flavor.

[0047] In this embodiment, the first characterization dot unit is responsible for drawing graphical dots representing each quantified characteristic of each consumer in the original group. It directly uses the characterization color of each quantified characteristic in the taste dimension data, such as purple corresponding to center tenderness, and then combines it with standard diameter and standard depth values ​​to generate dots. Each dot is a pure expression of a taste, with rich color and uniform size.

[0048] In this embodiment, the second representation dot unit specifically generates visual dots for culturally integrated groups. These individuals are those whose basic tastes originated elsewhere but have become ingrained in their current cultural context. During rendering, preset quantitative features from the taste dimension data are used to represent color, but the dot diameter is derived from a value calculated individually by the dot diameter unit, and the color depth is taken from a personalized depth calculated based on the individual's length of stay by the color depth value unit. Therefore, the representation dot of a culturally integrated consumer will exhibit characteristics such as a larger dot due to early migration and a medium color depth due to being in the middle stage of integration. This set of dots vividly depicts an intermediate state where roots are in a foreign land but the roots have integrated into the local culture.

[0049] In this embodiment, the third representation dot unit generates dots for the underlying assimilation group. These individuals have their basic tastes rooted in their current place of residence, but their cultural preferences developed elsewhere. The same representation color is used, but the individualized dot diameter and color depth values ​​are taken from these individuals. Since the underlying assimilation group typically migrates to other areas and settles into their current culture at a younger age, their dot diameter may be larger, while the color depth is determined by their actual length of time living in the local area. These dots exhibit a subtle tension in the representation map; their base is local color, but their explicit preferences have been washed by external culture, allowing observers to identify the foreign cultural impurities hidden within the local background of this group.

[0050] In this embodiment, the fourth characterizing dot unit is used to draw the dots for a completely immigrant group. These individuals come from other regions and have fully developed their tastes in their homeland before migrating to their current location. The diameter and color depth of their dots are entirely determined by their age of arrival and length of stay. An immigrant who migrated as an adult is typically represented by a very light-colored dot in the diagram, located far from the standard dot of the native group; their tastes stubbornly retain their native characteristics and have not yet been influenced by the culture of their current location.

[0051] In this embodiment, the box plot unit constructs the statistical background framework for the entire taste representation map. For a specific taste label, such as the native type, within a defined region, it collects all preference scores from all consumers under that label for a certain quantitative characteristic, such as crispness. These scores are used to calculate five statistical measures: the minimum, first quartile, median, third quartile, and maximum value, which are then plotted into a standard box plot structure. The horizontal axis of the plot sequentially arranges all the quantitative characteristics defined for that food category, while the vertical axis represents the scale of preference values. The boxes in the box plot show the concentration range of preferences in the middle 50% of the population, the lines at both ends show the extreme cases of preference dispersion, and the median indicates the typical level of the group.

[0052] In this embodiment, the taste representation map unit synthesizes all the dot plots and box plots mentioned above into a complete visual analysis tool. Specifically, on the drawn box plot, the representation dots for each consumer under the corresponding quantitative feature in the divided region and for that taste label are overlaid and drawn. Each dot is placed at a corresponding height on the vertical axis according to the user's specific preference value for that feature. Thus, a taste representation map is composed of multiple sub-maps representing the taste of quantitative features. At first glance, the box plot provides the macroscopic structure of the group, while the dots of varying shades and sizes scattered around the box each tell a unique story of an individual's origin and evolution.

[0053] The beneficial effects of the above technologies are: to draw a flavor representation map of each flavor tag in each segment of each food category, realizing the leap from static group classification to dynamic individual evolution tracking, and providing an interpretable basis for precise migration-based flavor recommendations. Example 5:

[0054] Based on Example 4, an intelligent control system for an oven includes a determining module, comprising: The inclusion rate unit: For each region of each food category, the original flavor representation map of the flavor label is coaxially superimposed with the flavor representation map of each flavor label other than the original type. The number of representation dots between the upper and lower quartiles of the flavor representation sub-map of each quantitative feature of each flavor label other than the original type and the flavor representation sub-map of each quantitative feature of each flavor label of the original type are counted. Based on the number of representation dots and the number of consumers in the corresponding flavor label category preference data, the inclusion rate of each quantitative feature of each flavor label other than the original type in each region of each food category is calculated. Centroid offset value unit: Based on the median and box width of the flavor representation submap of each quantized feature of each flavor label in each partition region of each food category with the flavor label as the native type, and the median of the flavor representation submap of each quantized feature of each flavor label other than the native type, calculate the centroid offset value of each quantized feature of each flavor label in each partition region of each food category, excluding the native type. Feature proximity factor unit: Based on the fall rate and centroid offset value of each quantized feature of each flavor label in each partition region of each food category (excluding the original type), calculate the feature proximity factor of each quantized feature of each flavor label in each partition region of each food category (excluding the original type); Morphological labeling unit: For each flavor label in each segmented region of each food category, excluding the original type, a morphological judgment is performed on the flavor representation sub-image of each quantified feature. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label fall evenly and densely into the flavor representation sub-image of the quantified feature of the original type according to the dot diameter and color depth value, the morphological label of the quantified feature of the flavor label is determined to be uniformly close. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label with large dot diameter and deep color depth value fall densely into the flavor representation sub-image of the quantified feature of the original type, and the representation dots with small dot diameter and light color depth value are densely distributed at the edge of the flavor representation sub-image of the quantified feature of the original type, the morphological label of the quantified feature of the flavor label is determined to be gradient close. First labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is greater than the first preset threshold, and the morphological label is uniformly close, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Second labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is between the second preset threshold and the first preset threshold, and the morphological label is gradient proximity, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Third segmentation label unit: Otherwise, determine the segmentation label of the quantitative feature of the flavor label as a segmentation.

[0055] In this embodiment, the inclusion rate unit is used to quantify the actual overlap between non-native taste labels (such as those from outside the group, those with underlying integration, or those with cultural integration) and the local native benchmark in each specific taste dimension. It first coaxially superimposes the taste representation maps of the native and non-native labels on the same coordinate system, ensuring complete overlap of the taste representation maps and representation dots. Then, for each quantified feature, such as crispness, it identifies the core region between the upper and lower quartiles of the box plot corresponding to the native feature's taste representation map, and calculates how many representation dots from the non-native label group fall into this region. This number of falling dots is then divided by the total number of consumers in that non-native label group for that feature, yielding the inclusion rate. The inclusion rate directly reflects the proportion of people in an outside group whose taste preferences are indistinguishable from those of native residents.

[0056] In this embodiment, the centroid offset value unit is used to measure the magnitude of the shift in the taste centroid of the non-native group relative to the local native benchmark. It uses the median of each quantized feature in the box plot corresponding to the native taste representation map as the origin reference for that dimension, and extracts the box width (interquartile range) of the box plot corresponding to the native taste representation map as the distance unit. Then, it extracts the median of the non-native group for that quantized feature. Based on the absolute value of the difference between the median of a certain quantized feature in the box plot corresponding to the taste representation map of a flavor label (excluding the native group) and the median of a certain quantized feature in the box plot corresponding to the native taste representation map, divided by the box width of that quantized feature in the box plot corresponding to the native box plot, the centroid offset value of that quantized feature for that flavor label is obtained. The core significance of this value is that it not only tells the system whether the tastes of the foreign group are different, but also quantifies, with an intuitive scale, how significant this difference is.

[0057] In this embodiment, the feature proximity factor unit integrates the two complementary metrics, the fall rate and the centroid offset value, into a comprehensive proximity score. A high fall rate and a small centroid offset value indicate that the group's taste is highly aligned with the native type, resulting in a large proximity factor. Groups with low fall rates or large centroid offset values, creating a two-way squeeze, will receive a smaller proximity factor. The feature proximity factor for each quantitative feature of each flavor tag in each segmentation region of each food category (excluding the native type) is calculated by dividing the fall rate by the centroid offset value plus 1. The feature proximity factor ranges from 0 to 1.

[0058] In this embodiment, the morphological labeling unit adds a visual distribution pattern judgment criterion in addition to numerical calculation to identify the dynamic pattern of group integration. It judges the visual pattern of the taste representation sub-graph of a certain non-native taste label on a certain quantitative feature. If the representation dots in the graph are evenly distributed and densely clustered inside the box plot corresponding to the native taste representation graph in terms of dot diameter and color depth, the morphological label is judged as uniformly close. For example, large-diameter dark dots (early immigrants and long-term residents) and small-diameter light dots (late immigrants and short-term residents) are evenly and densely clustered. However, if large-diameter and dark-colored dots are densely clustered in the native core area, but small-diameter and light-colored dots are sparsely distributed at the edge of the core area or even outside, the morphological label is judged as gradient close. This means that the early immigrants have been assimilated while the new immigrants have not yet been assimilated, and the group is in a dynamic integration process of intergenerational succession.

[0059] In this embodiment, the first segmentation labeling unit decides to maintain the status quo for groups with large feature proximity factors and uniform and stable morphology. It judges a certain quantitative feature of the flavor label; if its feature proximity factor is higher than a first preset threshold (meaning high similarity), and the morphological label is judged as uniformly similar (meaning that representation dots of similar diameter and color depth fall evenly and densely within the original flavor representation map), then the segmentation label for that quantitative feature is determined to be unclassified. This indicates that this flavor dimension of the group has been completely integrated with the local original group and no longer needs to be processed separately as an independent flavor subgroup. The first preset threshold can be set to 0.7.

[0060] In this embodiment, the second segmentation label unit handles subtle cases where the numerical values ​​are already moderately close, but the internal structure exhibits gradient evolution. A feature proximity factor of a quantified feature falling between a second preset threshold and a first preset threshold represents a moderate proximity, and a gradient proximity morphological label indicates that the group is in a process of transition between old and new forms. In this case, the segmentation label is also set to "no segmentation." The underlying logic is that although the group has not yet fully merged, it has shown a strong fusion inertia and will inevitably continue to move closer to the original form in the future. Maintaining no segmentation for the time being is more in line with its evolutionary trend. The second preset threshold can be set to 0.4.

[0061] In this embodiment, as long as the determination conditions of the first two units are not met, such as the feature proximity factor being very low and far from the local taste, or although the proximity factor is moderate but the morphology is neither gradient nor uniform, the division label will be determined as a division.

[0062] The beneficial effects of the above technology are as follows: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, the feature proximity factor and segmentation label of each quantitative feature of each flavor tag in each segmentation region of each food category (excluding the native type) are determined. This ensures that the oven's flavor recommendation model adapts and evolves with the real population across generations, achieving accurate capture and forward-looking service of taste changes in a mobile society. Example 6:

[0063] Based on Example 5, a recommended module for an intelligent oven control system includes: Clustering Unit: Based on the quantitative features of each flavor tag (excluding the original type) for each partition region of each food category, and the preference values ​​of the quantitative features in the taste preference data of all consumers in the category preference data of the flavor tags, cluster analysis is performed on all consumers in the category preference data of the flavor tags to determine multiple feature clustering data of each flavor tag (excluding the original type) for each partition region of each food category. The feature clustering data includes the quantitative values ​​of multiple consumers and the characterization dots. Representation clustering graph unit: Based on the feature clustering data of each flavor label (excluding the original type) of each partition region of each food category, a representation clustering graph is drawn for each cluster label of each flavor label (excluding the original type) of each partition region of each food category, taking the quantified feature of each partition as the partition. Clustering proximity factor unit: Based on the clustering graph representing each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category, and the flavor representation subgraph of the quantification feature, calculate the clustering proximity factor of each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category. Flavor proximity data unit: Based on the feature proximity factors of all quantized features of flavor labels other than the original type for each region of each food category with no division, and the cluster proximity factors of all cluster labels with quantized features of division, the flavor proximity data of each flavor label other than the original type for each region of each food category is determined. Original quantization vector unit: Based on the original flavor representation map of each segmentation region of each food category, the original quantization vector of each segmentation region of each food category is determined. The original quantization vector includes the original quantization values ​​of multiple quantization features. Flavor Recommendation Model Unit: Based on the flavor proximity data of each flavor tag in all regions of each food category (excluding the original type), and the original quantized vector of the flavor tag as the original type, a flavor recommendation model is constructed for each food category.

[0064] In this embodiment, the clustering unit is responsible for a deeper analysis of the flavor tag that has been determined to be segmented. Its input consists of the quantitative feature that previously determined the segmentation and the specific preference values ​​of all consumers on that quantitative feature from the flavor tag category preference data. The unit treats these consumers as individual samples and uses only their preference values ​​on this single quantitative feature as the clustering variable to perform unsupervised clustering analysis. Clustering automatically groups people with similar tastes on this feature into several smaller, more consistent groups, thereby identifying multiple latent subgroups within the tag. The output is multiple feature clustering data, each representing a new latent subgroup, containing the specific quantitative value of each consumer on that feature and the previously generated representation dots for that person.

[0065] In this embodiment, the characterization clustering graph unit constructs a visual taste graph for each latent subgroup that emerges after clustering. For each cluster label generated under the quantified feature of each partitioning label, its feature clustering data is extracted, and a new graph is drawn. This graph is structurally similar to the previous taste characterization subgraph, but its data scope is limited to consumers within this single cluster. The graph also uses dot diameter and color depth to encode each member's age of arrival and length of local residence, and these dots are scattered along the vertical axis of preference values. This characterization clustering graph vividly demonstrates the taste distribution pattern and migration background composition within the partitioned subgroups.

[0066] In this embodiment, the cluster proximity factor unit measures the proximity of a subgroup within a non-native flavor label to a local native benchmark on a single quantitative feature. Its calculation method is completely consistent with the feature proximity factor. This unit uses the representation clustering map of a specific cluster label under a quantitative feature determined to require segmentation as the analysis object, while simultaneously introducing a native flavor representation submap on the same quantitative feature as a reference. Upon determination, the representation clustering map and the native flavor representation submap are coaxially superimposed. The number of representation dots in the subgroup falling within the core region between the upper and lower quartiles of the box plot corresponding to the native flavor representation submap is counted. This number is divided by the total number of subgroup members to obtain the inclusion rate. Simultaneously, the median and box width of the box plot corresponding to the native flavor representation submap are extracted as distance scales, and the offset of the subgroup median relative to the native median is calculated to obtain the centroid offset value. Subsequently, the inclusion rate and centroid offset value are fused to obtain a comprehensive cluster proximity factor.

[0067] In this embodiment, the taste proximity data unit summarizes all feature proximity factors of all quantized features marked as undivided within the non-native taste labels, and also summarizes the cluster proximity factors of all cluster labels determined by clustering within all quantized features marked as divided, to form taste proximity data.

[0068] In this embodiment, the original quantization vector unit extracts a set of baseline vectors specifically for the native population, serving as the absolute origin for all recommendations. It extracts data directly from the native taste representation map. Since the map is visual, the underlying values ​​are the median or mode. Arranging these medians according to the order of quantization features constitutes an original quantization vector. Each original quantization value in the vector represents the ideal value of the most standard and authentic taste of that food category in that region on the corresponding dimension. This vector is the digital expression of the taste gene in that region.

[0069] In this embodiment, the taste recommendation model uses the original quantization vector as the absolute benchmark and achieves a precise mapping from group statistics to individual matching by structurally interpreting the taste proximity data of the fusion group. The model first embeds the original quantization vector of the original group for each segmented region. Simultaneously, the model stratifies the taste proximity data of each non-original taste tag into two states: segmented and unsegmented. For quantization features determined to be unsegmented, the overall feature proximity factor of the group is directly used as the fusion coefficient for that dimension. For quantization features determined to be segmented, the clustering proximity factors of multiple latent subgroups within it are retained, describing the independent proximity degree between different subgroups and the original benchmark. When recommending to a target user, the model determines the taste tag based on their migration background, further matching the specific clustering subgroups corresponding to their migration age and duration of residence in the segmentation dimension, extracting the corresponding clustering proximity factors, and combining them with the overall feature proximity factors of the unsegmented dimension to form a set of personalized multi-dimensional fusion coefficient vectors. Finally, using this vector as the weight, a dimension-wise weighted interpolation is performed between the original quantized vector and the user's native taste vector. The higher the fusion coefficient, the closer the recommended value is to the local native taste; otherwise, it is reverted to the native taste, thereby generating recommended data that matches the degree of fusion with the individual's taste.

[0070] The beneficial effects of the above technology are as follows: based on the category preference data, flavor representation map, feature proximity factor, and division label of each flavor tag in each region of each food category (excluding the original type), a flavor recommendation model for each food category is constructed, realizing a leap from fuzzy regional group inference to accurate personal migration history matching for personalized cooking recommendations. Example 7:

[0071] Based on Example 1, an intelligent control system for an oven includes a control module comprising: User data unit: Collects user data of the target users of the target oven, wherein the user data includes at least the address of use, the main residence of children aged 0-3 years, the main residence of children aged 3-12 years, the age of migration to the address of use, and the duration of residence; Unit Determination: The division of regions is determined based on the usage address in the user data of the target users; the basic taste region is determined based on the main residence of children aged 0-3 in the user data of the target users; and the cultural taste region is determined based on the main residence of children aged 3-12 in the user data of the target users. Recommended Flavor Data Unit: Obtain the ingredients to be cooked by the target user using the target oven, input the target user's regional segmentation, basic flavor region, cultural flavor region, age of migration, and duration of residence into the flavor recommendation model of the food category corresponding to the ingredients to be cooked, determine the recommended flavor data of the target user based on the ingredients to be cooked, and display it in the visualization area of ​​the target oven. The recommended flavor data includes recommended quantitative values ​​of multiple quantitative features. Flavor Selection Data Unit: Acquires the flavor selection data of the target user, which includes the selection quantification values ​​of multiple quantification features; Control unit: Inputs the target user's recommended taste data and selected taste data into the oven control model, generates control instructions based on the ingredients to be cooked for the target user, and executes the control instructions to achieve cooking control.

[0072] In this embodiment, the user data unit is responsible for collecting key life trajectory information of the user when the oven is put into use. The collected data items include the user's current geographical location (address) when using the oven, the long-term residence during the period from 0 to 3 years old, the long-term residence during the period from 3 to 12 years old, the age at which the user moved to the area, and the cumulative duration of residence in the area corresponding to the address.

[0073] In this embodiment, the determining unit maps the discrete address information collected by the user data unit to taste geographic tags within the system. It performs a spatial query in the spatial database of the segmented areas using the usage address to obtain the segmented area to which the user currently belongs. Simultaneously, it uses the primary residence data of children aged 0 to 3 years to determine their affiliation within the same segmented area system. Once a segmented area containing that location is matched, it is marked as the user's underlying taste area. The same method is used to determine the primary residence of children aged 3 to 12 years to obtain their cultural taste area.

[0074] In this embodiment, the recommended flavor data unit first obtains the ingredients the user intends to cook in the oven, such as a steak or a plate of fries. Then, it inputs five personalized parameters—the user's parsed regional classification, underlying flavor region, cultural flavor region, age of arrival, and length of residence—into the flavor recommendation model for the corresponding food category. Based on the input data, the model first determines which flavor tag type the user belongs to locally, and then, within that type, determines the recommended quantitative values ​​for each quantitative feature based on the user's arrival date and length of residence. This multi-dimensional recommendation value is then displayed on the oven's visualization screen.

[0075] In this embodiment, the taste data selection unit grants users the freedom to exercise their current preferences based on system recommendations. After displaying the recommended taste data, it allows users to set the quantification values ​​for each quantitative feature through interactive methods such as touchscreens.

[0076] In this embodiment, the control unit is responsible for translating abstract flavor target values ​​into a control language that the oven hardware can understand. It inputs both recommended flavor data and user-defined flavor selection data into the oven control model. The model prioritizes the user-defined quantified values ​​in its mapping logic, supplementing dimensions not manually adjusted by the user with recommended quantified values, forming a final flavor target vector. The oven control model then translates this target vector into a set of control instructions using the correspondence stored in the flavor control mapping table, such as the heating element's set temperature, the steam generator's injection duration, and the hot air fan's speed. Finally, the actuators work collaboratively according to these instructions, completing a closed-loop control process from analyzing the geographical migration background, group statistical analysis, and fine-tuning based on individual needs, to ultimately achieving precise physical cooking operations.

[0077] The beneficial effects of the above technologies are as follows: collecting user data of the target users of the target oven, obtaining the ingredients to be cooked and the selected flavor data, determining the recommended flavor data based on the user data, the ingredients to be cooked, and the flavor recommendation model of all food categories, generating control instructions based on the selected flavor data and the recommended flavor data, and realizing cooking control, thereby achieving customized flavor closed-loop control that varies from person to person and changes with the individual. Example 8:

[0078] This invention provides an intelligent control method for an oven, used to execute any one of the intelligent control systems for an oven in Examples 1 to 7, with reference to... Figure 2 ,include: S1: Set flavor dimension data for each food category in the cooking category data of the target oven, and collect regional flavor preference data for each divided area based on the flavor dimension data set for each food category; S2: Based on the regional taste preference data of each division of each food category, determine the category preference data of multiple taste labels in each division of each food category, and draw a taste representation map of each taste label in each division of each food category; S3: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantitative feature of each flavor tag in each segmentation region of each food category, excluding the original type; S4: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), construct a flavor recommendation model for each food category; S5: Collect user data of the target user of the target oven, obtain the ingredients to be cooked and the selected flavor data, determine the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generate control instructions based on the selected flavor data and the recommended flavor data, and realize cooking control.

[0079] The beneficial effects of the above technology are as follows: By setting flavor dimension data for each food category in the cooking category data of the target oven, collecting regional flavor preference data for each segmented area based on the flavor dimension data set for each food category, determining the category preference data of multiple flavor tags for each segmented area of ​​each food category, drawing a flavor representation map of each flavor tag for each segmented area of ​​each food category, determining the feature proximity factor and segmentation label of each quantitative feature of each flavor tag for each segmented area of ​​each food category (excluding the original type), constructing a flavor recommendation model for each food category, collecting user data of the target users of the target oven, obtaining data on ingredients to be cooked and selected flavors, determining recommended flavor data, generating control commands, and realizing cooking control. This enables precise quantitative discrimination of flavor features, completes refined stratification of flavor attributes for different groups, accurately adapts to the differentiated flavor needs of people with different backgrounds and regions, weakens the adaptation bias caused by regional dietary differences, improves the targeting and accuracy of oven flavor recommendations, and enhances the personalization and intelligence level of food processing control. Example 9:

[0080] The present invention provides an intelligent oven control device for executing any one of the intelligent oven control systems in Examples 1 to 7.

[0081] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An intelligent control system for an oven, characterized in that, include: Data Acquisition Module: Sets flavor dimension data for each food category in the cooking category data of the target oven, and collects regional flavor preference data for each divided area based on the flavor dimension data set for each food category; The plotting module: Based on the regional taste preference data of each segmented region for each food category, it determines the category preference data of multiple taste labels for each segmented region for each food category, and plots the taste representation map of each taste label for each segmented region for each food category; Determine module: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantified feature of each flavor tag in each segmentation region of each food category, excluding the original type; Recommendation module: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), a flavor recommendation model is constructed for each food category. Control module: Collects user data of the target user of the target oven, obtains the ingredients to be cooked and the selected flavor data, determines the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generates control instructions based on the selected flavor data and the recommended flavor data and realizes cooking control.

2. The intelligent control system for an oven as claimed in claim 1, wherein, The data acquisition module includes: Cooking category data unit: Based on the food attributes of the culinary ingredients of the target oven, determine the cooking category data of the target oven. The oven cooking category data includes at least meat, seafood, vegetables, pasta products, and fruits. Regional division: Based on administrative region data and food culture data, multiple regional divisions are determined; Setting Unit: Based on each food category in the cooking category data of the target oven, set the flavor dimension data, where the flavor dimension data includes multiple quantitative features and the color representing each quantitative feature; Regional taste preference data unit: Based on the taste dimension data set for each food category in the cooking category data of the target oven, regional taste preference data for each divided region is collected. The regional taste preference data includes individual growth data and taste preference data of multiple consumers. The individual growth data includes at least the primary residence for children aged 0-3 years, the primary residence for children aged 3-12 years, the age of migration to the divided region, and the length of stay. The taste preference data includes preference values ​​of multiple quantitative features.

3. The intelligent control system for an oven as claimed in claim 1, wherein, The drawing module includes: Bottom-level taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of each consumer aged 0-3 years is determined by comparing it with all the regional divisions, and the regional division containing the primary residence of each consumer aged 0-3 years is determined as the bottom-level taste region of the consumer. Cultural taste region unit: For each food category, in the regional taste preference data of each consumer, the primary residence of the 3-12-year-olds in the individual growth data is compared with all the regional divisions to determine the division containing the primary residence of the 3-12-year-olds as the cultural taste region of the consumer. Native type unit: If the underlying taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is consistent with the segmentation region, then the taste label of the consumer is determined to be native type; Bottom-level fusion type unit: If the bottom-level taste region of each consumer in the regional taste preference data of each segmentation region of each food category is consistent with the segmentation region, and the cultural taste region is inconsistent with the segmentation region, the taste label of the consumer is determined to be bottom-level fusion type. Culturally integrated unit: If the underlying taste region of each consumer in the regional taste preference data of each region of each food category is inconsistent with the region, but the cultural taste regions are consistent with the region, the taste label of the consumer is determined to be culturally integrated. Foreign type unit: If the underlying taste region and cultural taste region of each consumer in the regional taste preference data of each region of each food category are inconsistent with the region, the taste label of the consumer is determined to be foreign type. Category preference data unit: Based on the taste tags of all consumers in the regional taste preference data of each segmented region of each food category, the regional taste preference data of the segmented region is divided to determine the category preference data of multiple taste tags in each segmented region of each food category. The category preference data includes individual growth data and taste preference data of multiple consumers.

4. The intelligent control system for an oven according to claim 3, characterized in that, The drawing module also includes: Dot diameter unit: Based on the immigration age in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region of each food category, and the first adaptive mapping curve, determine the dot diameter of each consumer in the category preference data of each food category with the taste label of cultural integration, basic integration or foreign type in each division region. Color depth value unit: Based on the life duration in the individual growth data of all consumers in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category, the color of each quantitative feature in the taste dimension data and the second adaptive mapping curve are used to determine the color depth value of each consumer in the category preference data of each food category with the taste label of cultural integration, low-level integration or foreign type in each division region of each food category based on each quantitative feature; Standard unit: The taste label of each segment of each food category is determined to be the standard diameter of the dot of each consumer in the native category preference data. At the same time, the color depth value of each quantized feature of each consumer in the native category preference data is determined to be the standard depth value of the taste label of each segment of each food category. The first representation dot unit: Based on the representation color and standard diameter and standard depth values ​​of each quantified feature in the taste dimension data, the taste label of each division region of each food category is determined as the representation dot of each quantified feature of all consumers in the original type; The second representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the culturally integrated category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the culturally integrated representation dot. The third representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the bottom-level fusion type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the bottom-level fusion type. The fourth representation dot unit: Based on the representation color of each quantitative feature in the taste dimension data and the taste label of each division region of each food category as the dot diameter of each consumer in the foreign type category preference data and the color depth value of each quantitative feature, the taste label of each division region of each food category is determined as the representation dot of each quantitative feature of each consumer in the foreign type. Box plot unit: Based on the preference values ​​of all quantitative features in the category preference data of all consumers in the taste preference data of each flavor label in each segment of each food category, a box plot is drawn for each flavor label in each segment of each food category. The horizontal axis of the box plot represents all quantitative features, and the vertical axis represents the preference value of the quantitative features. Flavor representation map unit: Based on the preference value of each consumer's quantitative feature for each flavor label in each segmented region of each food category, the representation dots of the consumer's quantitative feature are plotted on the box plot to determine the flavor representation map for each flavor label in each segmented region of each food category, wherein the flavor representation map includes flavor representation sub-maps of multiple quantitative features.

5. The intelligent control system for an oven according to claim 4, characterized in that, The module to be determined includes: The inclusion rate unit: For each region of each food category, the original flavor representation map of the flavor label is coaxially superimposed with the flavor representation map of each flavor label other than the original type. The number of representation dots between the upper and lower quartiles of the flavor representation sub-map of each quantitative feature of each flavor label other than the original type and the flavor representation sub-map of each quantitative feature of each flavor label of the original type are counted. Based on the number of representation dots and the number of consumers in the corresponding flavor label category preference data, the inclusion rate of each quantitative feature of each flavor label other than the original type in each region of each food category is calculated. Centroid offset value unit: Based on the median and box width of the flavor representation submap of each quantized feature of each flavor label in each partition region of each food category with the flavor label as the native type, and the median of the flavor representation submap of each quantized feature of each flavor label other than the native type, calculate the centroid offset value of each quantized feature of each flavor label in each partition region of each food category, excluding the native type. Feature proximity factor unit: Based on the fall rate and centroid offset value of each quantized feature of each flavor label in each partition region of each food category (excluding the original type), calculate the feature proximity factor of each quantized feature of each flavor label in each partition region of each food category (excluding the original type); Morphological labeling unit: For each flavor label in each segmented region of each food category, excluding the original type, a morphological judgment is performed on the flavor representation sub-image of each quantified feature. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label fall evenly and densely into the flavor representation sub-image of the quantified feature of the original type according to the dot diameter and color depth value, the morphological label of the quantified feature of the flavor label is determined to be uniformly close. If the representation dots in the flavor representation sub-image of the quantified feature of the flavor label with large dot diameter and deep color depth value fall densely into the flavor representation sub-image of the quantified feature of the original type, and the representation dots with small dot diameter and light color depth value are densely distributed at the edge of the flavor representation sub-image of the quantified feature of the original type, the morphological label of the quantified feature of the flavor label is determined to be gradient close. First labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is greater than the first preset threshold, and the morphological label is uniformly close, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Second labeling unit: If the feature proximity factor of the quantitative feature of the flavor label is between the second preset threshold and the first preset threshold, and the morphological label is gradient proximity, then the labeling of the quantitative feature of the flavor label is determined to be unlabeled. Third segmentation label unit: Otherwise, determine the segmentation label of the quantitative feature of the flavor label as a segmentation.

6. The intelligent control system for an oven according to claim 5, characterized in that, Recommendation modules include: Clustering Unit: Based on the quantitative features of each flavor tag (excluding the original type) for each partition region of each food category, and the preference values ​​of the quantitative features in the taste preference data of all consumers in the category preference data of the flavor tags, cluster analysis is performed on all consumers in the category preference data of the flavor tags to determine multiple feature clustering data of each flavor tag (excluding the original type) for each partition region of each food category. The feature clustering data includes the quantitative values ​​of multiple consumers and the characterization dots. Representation clustering graph unit: Based on the feature clustering data of each flavor label (excluding the original type) of each partition region of each food category, a representation clustering graph is drawn for each cluster label of each flavor label (excluding the original type) of each partition region of each food category, taking the quantified feature of each partition as the partition. Clustering proximity factor unit: Based on the clustering graph representing each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category, and the flavor representation subgraph of the quantification feature, calculate the clustering proximity factor of each clustering label of each flavor label (excluding the original type) as the quantification feature of each region of each food category. Flavor proximity data unit: Based on the feature proximity factors of all quantized features of flavor labels other than the original type for each region of each food category with no division, and the cluster proximity factors of all cluster labels with quantized features of division, the flavor proximity data of each flavor label other than the original type for each region of each food category is determined. Original quantization vector unit: Based on the original flavor representation map of each segmentation region of each food category, the original quantization vector of each segmentation region of each food category is determined. The original quantization vector includes the original quantization values ​​of multiple quantization features. Flavor Recommendation Model Unit: Based on the flavor proximity data of each flavor tag in all regions of each food category (excluding the original type), and the original quantized vector of the flavor tag as the original type, a flavor recommendation model is constructed for each food category.

7. The intelligent control system for an oven according to claim 1, characterized in that, The control module includes: User data unit: Collects user data of the target users of the target oven, wherein the user data includes at least the address of use, the main residence of children aged 0-3 years, the main residence of children aged 3-12 years, the age of migration to the address of use, and the duration of residence; Unit Determination: The division of regions is determined based on the usage address in the user data of the target users; the basic taste region is determined based on the main residence of children aged 0-3 in the user data of the target users; and the cultural taste region is determined based on the main residence of children aged 3-12 in the user data of the target users. Recommended Flavor Data Unit: Obtain the ingredients to be cooked by the target user using the target oven, input the target user's regional segmentation, basic flavor region, cultural flavor region, age of migration, and duration of residence into the flavor recommendation model of the food category corresponding to the ingredients to be cooked, determine the recommended flavor data of the target user based on the ingredients to be cooked, and display it in the visualization area of ​​the target oven. The recommended flavor data includes recommended quantitative values ​​of multiple quantitative features. Flavor Selection Data Unit: Acquires the flavor selection data of the target user, which includes the selection quantification values ​​of multiple quantification features; Control unit: Inputs the target user's recommended taste data and selected taste data into the oven control model, generates control instructions based on the ingredients to be cooked for the target user, and executes the control instructions to achieve cooking control.

8. A method for intelligent control of an oven, characterized in that, An intelligent control system for an oven as described in any one of claims 1 to 7, comprising: S1: Set flavor dimension data for each food category in the cooking category data of the target oven, and collect regional flavor preference data for each divided area based on the flavor dimension data set for each food category; S2: Based on the regional taste preference data of each division of each food category, determine the category preference data of multiple taste labels in each division of each food category, and draw a taste representation map of each taste label in each division of each food category; S3: Based on the flavor representation map of all flavor tags in each segmentation region of each food category, determine the feature proximity factor and segmentation label of each quantitative feature of each flavor tag in each segmentation region of each food category, excluding the original type; S4: Based on the category preference data, flavor representation map, feature proximity factor, and partition label of each flavor tag in each region of each food category (excluding the original type), construct a flavor recommendation model for each food category; S5: Collect user data of the target user of the target oven, obtain the ingredients to be cooked and the selected flavor data, determine the recommended flavor data based on the user data, the ingredients to be cooked and the flavor recommendation model of all food categories, generate control instructions based on the selected flavor data and the recommended flavor data, and realize cooking control.

9. An intelligent control device for an oven, characterized in that, An intelligent control system for an oven as described in any one of claims 1 to 7.

Images