Food recommendation method, apparatus, and pet robot

Abstract

The application provides a food recommendation method, device and pet robot. The method comprises the following steps: determining a recommendation score of each type of food based on a first adaptation degree between each type of food and a real-time state of the pet robot, and / or a second adaptation degree between each type of food and a living scene in which the pet robot is located; and displaying food recommendation information of the pet robot based on the recommendation score of each type of food. The method, device and robot provided by the application can simulate a real pet to actively express a preference and demand for specific food, thereby guiding a user to preferentially feed food that is more suitable for the current state of the pet robot and / or the living scene in which the pet robot is located, greatly improving the feeding interaction experience between the user and the pet robot, and making the feeding interaction of the pet robot more real, rich and interesting.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a food recommendation method, apparatus, and pet robot. Background Technology

[0002] To enhance the realism of pet robots, most current pet robots are equipped with feeding functions, allowing users to interact with them by feeding them food, water, and other actions, thereby increasing the possibility of interaction between users and pet robots.

[0003] However, in actual feeding interactions, pet robots can only proactively express hunger to users, leaving the feeding entirely to the user's choice of food based on their preferences. Unlike real pets, they do not choose their preferred foods or refuse foods they dislike. This mechanical, one-way feeding process fails to provide users with a realistic pet feeding experience. Summary of the Invention

[0004] This invention provides a food recommendation method, apparatus, and pet robot to address the shortcomings of related technologies in terms of poor user experience due to the mechanical and one-way feeding interaction of pet robots.

[0005] This invention provides a food recommendation method, comprising: Based on the first degree of fit between various foods and the real-time status of the pet robot, and / or the second degree of fit between various foods and the living scenario of the pet robot, the recommendation score of each type of food is determined. Based on the recommendation scores for each type of food, the pet robot displays food recommendation information.

[0006] According to a food recommendation method provided by the present invention, the real-time status includes at least one of hunger level, mood value and health status; the life scenario includes seasonal scenario and / or cultural event scenario.

[0007] According to a food recommendation method provided by the present invention, determining a recommendation score for each type of food based on a first degree of fit between various types of food and the real-time status of a pet robot, and / or a second degree of fit between various types of food and the living environment of the pet robot, includes: Based on the first fit between the nutritional properties of each type of food and at least one of the hunger and mood values ​​of the pet robot, the mood coefficient of each type of food is determined. Based on the first degree of fit between the health intervention attributes of the various types of food and the health status of the pet robot, the health coefficient of the various types of food is determined. Based on the second degree of fit between the suitable temperature attribute of the various types of food and the seasonal scene in which the pet robot is located, the seasonal bonus coefficient of the various types of food is determined. Based on the second degree of fit between the various types of food and the cultural event scenario in which the pet robot is located, the event bonus coefficient of the various types of food is determined; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient.

[0008] According to a food recommendation method provided by the present invention, determining a recommendation score for each type of food based on at least one of a mood coefficient, a health coefficient, a seasonal bonus coefficient, and an event bonus coefficient includes: Based on the number of consecutive feedings of each type of food, the continuous penalty coefficient for each type of food is determined; Randomly generate random perturbation coefficients for various types of food; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient, as well as the continuous penalty coefficient and / or random perturbation coefficient of each type of food.

[0009] A food recommendation method provided by the present invention further includes: In response to a food selection operation, the actual food to be fed to the pet robot is determined; Based on the actual food fed to the pet robot, determine the feedback behavior information of the pet robot in response to the actual food fed to the pet robot; Display the feedback behavior information.

[0010] A food recommendation method provided by the present invention further includes: The real-time status of the pet robot is updated based on the actual food fed to it.

[0011] According to a food recommendation method provided by the present invention, updating the real-time status of the pet robot based on the actual food fed includes: Based on the differences in food attributes between the actual feeding food and the target feeding food, the recommendation deviation of the actual feeding food is determined; the target feeding food is the food with the highest recommendation score among the various food categories. The real-time status of the pet robot is updated based on the recommended deviation.

[0012] According to a food recommendation method provided by the present invention, the food attributes include energy intake value and / or mood optimization value.

[0013] The present invention also provides a food recommendation device, comprising: The calculation unit is used to determine the recommendation score of each type of food based on the first degree of fit between the real-time status of each type of food and the pet robot, and / or the second degree of fit between each type of food and the living scenario in which the pet robot is located. The recommendation unit is used to display food recommendation information for the pet robot based on the recommendation scores of the various types of food.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the food recommendation method as described above.

[0015] The present invention also provides a pet robot, including a pet robot body, a memory, a processor, and a computer program stored in the memory and running on the processor disposed inside the pet robot body, wherein the processor executes the computer program to implement any of the above-described food recommendation methods.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the food recommendation method as described above.

[0017] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the food recommendation method as described above.

[0018] The food recommendation method, device, and pet robot provided by this invention calculate recommendation scores for various foods based on a first degree of fit between various foods and the real-time state of the pet robot and / or a second degree of fit between various foods and the living environment of the pet robot. Based on the recommendation scores, the device displays food recommendation information to the user, enabling the pet robot to simulate a real pet actively expressing its preferences and needs for specific foods. This guides the user to prioritize feeding foods that are more suitable for the pet robot's current state and / or living environment, significantly improving the feeding interaction experience between the user and the pet robot, and making the feeding interaction with the pet robot more realistic, rich, and interesting. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1This is a flowchart illustrating the food recommendation method provided by the present invention.

[0021] Figure 2 This is a schematic diagram of the food recommendation device provided by the present invention.

[0022] Figure 3 This is a structural schematic diagram of the pet robot provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0024] All actions involving the acquisition of signal information or data in this invention are carried out in compliance with the relevant data protection laws and policies of the country where the device is located, and with the authorization granted by the owner of the device.

[0025] Child companion robots are intelligent interactive devices designed specifically for children aged 0-12. Through functions such as voice dialogue, facial expression interaction, motion sensing, and content playback, they provide emotional companionship, early education, and safety monitoring. Typically, these robots integrate technologies such as voice recognition, sensors, and cameras, enabling them to tell stories, teach knowledge, remind parents of schedules, provide simple emotional responses, and connect with parents' mobile phones for remote monitoring and interaction.

[0026] Pet robots, as a type of companion robot for children, are intelligent interactive devices capable of mimicking the appearance and behavior of pets. They can provide emotional companionship, interactive entertainment, and other services to children or other users. For example, pet robots can be dog-shaped, cat-shaped, panda-shaped, etc.

[0027] To enhance the realism of pet robots and make them more like real pets, most current pet robots have feeding functions. Users can interact with their pet robots by feeding them, giving them water, taking them to the vet, bathing them, and so on, thus creating more possibilities for interaction between users and pet robots.

[0028] However, in actual feeding interactions, pet robots can only proactively express basic physiological needs such as hunger to users. Users then choose food entirely based on their own preferences. In the real world, however, real pets have clear food preferences and typically refuse to eat foods they don't like. Current pet robots, on the other hand, cannot choose their preferred food like real pets, nor can they express their preferences or needs for specific foods to users. This mechanical, one-way feeding process makes pet robots lack a sense of life and fails to provide users with a realistic pet feeding experience.

[0029] In response, this invention provides a food recommendation method aimed at optimizing the user experience of feeding a pet robot. Figure 1 This is a flowchart illustrating the food recommendation method provided by the present invention. Figure 1 As shown, the method includes: Step 110: Determine the recommendation score for each type of food based on the first degree of fit between the various types of food and the real-time status of the pet robot, and / or the second degree of fit between the various types of food and the living scenario in which the pet robot is located.

[0030] Specifically, the various types of food refer to a variety of virtual foods pre-established in the pet robot's food database. These foods can be everyday foods, such as rice, mangoes, and hamburgers, or they can be virtual foods for pet robots, such as hydrogen batteries. This embodiment of the invention does not specifically limit these.

[0031] When recommending food for pet robots, the real-time status of the pet robot and / or the living environment in which the pet robot is located can be taken into account. Based on this, food that is suitable for the real-time status of the pet robot and / or the living environment in which the pet robot is located can be recommended to the user as the food to feed the pet robot.

[0032] The real-time status of a pet robot refers to its current state. This status reflects its intrinsic needs or physiological and emotional state, such as hunger, mood, and health. The real-time status can be maintained and updated in real time using the robot's built-in state perception module or state simulation algorithm.

[0033] After obtaining the real-time status of the pet robot, the compatibility between various types of food and the pet robot's real-time status can be calculated, denoted here as the first compatibility. For any food, the first compatibility between the food and the pet robot's real-time status measures how suitable the food is for the pet robot in that real-time state. The higher the first compatibility, the more suitable the food is for the pet robot's real-time status, and the more beneficial it is for satisfying or improving the pet robot's status. The first compatibility can be calculated through comparative analysis of food attributes and real-time status, rule matching, or model reasoning, etc., and this embodiment of the invention does not specifically limit this.

[0034] The living scenario of a pet robot refers to the external environment or temporal context information in which the pet robot exists. The living scenario can include information related to the external natural environment, information related to socio-cultural activities, or a combination of both. The information of the living scenario can be automatically acquired by the pet robot through its built-in state perception module, or it can be pre-configured by the system or manually set by the user; this embodiment of the invention does not specifically limit this.

[0035] After obtaining the living environment of the pet robot, the compatibility degree between various foods and the living environment can be calculated, denoted here as the second compatibility degree. For any food, the second compatibility degree between the food and the living environment of the pet robot can measure the suitability of the food in the living environment. The higher the second compatibility degree, the more the food is consistent with the characteristics or cultural events of the living environment of the pet robot. The second compatibility degree can be calculated by comparing food attributes with living environment information, rule matching, or querying a mapping table, etc., and this embodiment of the invention does not specifically limit this.

[0036] For any food item, after obtaining the first degree of fit between the food and the real-time state of the pet robot, and / or the second degree of fit between the food and the pet robot's living environment, a recommendation score can be calculated for that food. The recommendation score here refers to a quantitative rating calculated for that food, reflecting its recommendation priority to the pet robot at the current moment. A higher recommendation score indicates that the food is more suitable to be recommended to the user for feeding the pet robot.

[0037] In some embodiments, the recommendation score may be determined based solely on the first fit or solely on the second fit. In other embodiments, the recommendation score may be determined by weighted summation or product of the first and second fits, or by other comprehensive scoring methods. The embodiments of the present invention do not specifically limit this.

[0038] Understandably, the first degree of fit between the food and the pet robot's real-time status may differ for different foods, and the second degree of fit with the living environment may also differ. Therefore, the recommendation scores calculated for different foods may also be different.

[0039] For example, the system can first calculate the first degree of fit between various foods in the food database and the current real-time state of the pet robot, and the second degree of fit between various foods and the pet robot's current living environment. Then, according to a preset comprehensive scoring rule, the first and second degrees of fit for each type of food are combined to obtain a recommendation score for each type of food. The weights of the first and second degrees of fit in the comprehensive score can be pre-configured according to actual needs, and this embodiment of the invention does not impose specific limitations on this.

[0040] Step 120: Based on the recommendation scores of the various types of food, display the food recommendation information of the pet robot.

[0041] Specifically, after calculating the recommendation scores for various food categories in step 110, food recommendation information can be determined and displayed based on the recommendation scores. Here, food recommendation information refers to the recommended content generated based on the recommendation scores of various food categories and displayed to the user. Food recommendation information may include the identifiers of one or more foods with high recommendation scores, such as the food names, food icons, etc. of one or more foods with high recommendation scores; food recommendation information may also include the identifiers of one or more foods with high recommendation scores, as well as the recommendation score values ​​or recommendation priority rankings of the aforementioned foods, and one or more optional explanations of recommendation reasons. This embodiment of the invention does not specifically limit this.

[0042] For example, various types of food can be sorted from highest to lowest recommendation score, and the food with the highest recommendation score, or several foods with the highest recommendation scores, can be selected as target recommended foods to generate corresponding food recommendation information for display. Alternatively, all foods can be mapped to a food recommendation list, and the order of various types of food in the food recommendation list can be adjusted according to their recommendation scores from highest to lowest, so that foods with higher recommendation scores in the displayed food recommendation list can be given priority for display.

[0043] Food recommendation information can be displayed through the pet robot's screen, interactive interface, or other output methods. For example, the food with the highest recommendation score can be highlighted, pinned to the top, or specially marked in the pet robot's food interactive interface to guide the user to prioritize feeding the pet robot that food; multiple foods can also be arranged and displayed in descending order of recommendation score for the user to refer to and choose from; the pet robot can also proactively express its need for a certain food to the user through animation, voice, or other forms, and this embodiment of the invention does not specifically limit this.

[0044] For example, assuming the food database contains various foods such as rice, mango, hamburgers, and baguettes, after calculating the recommendation scores for each food, the recommendation scores are as follows: hamburgers have a recommendation score of 3.2, apples have a recommendation score of 2.6, rice has a recommendation score of 1.8, and baguettes have a recommendation score of 0.7. Then, food recommendation information can be generated with hamburgers as the primary recommended food, and hamburgers can be prominently displayed in the pet robot's food interaction interface, thereby guiding users to feed the pet robot hamburgers.

[0045] In the food recommendation method provided in this embodiment of the invention, a recommendation score is calculated for each type of food based on a first degree of fit between various types of food and the real-time state of the pet robot and / or a second degree of fit between various types of food and the living scenario in which the pet robot is located. Based on the recommendation score, food recommendation information is displayed to the user, enabling the pet robot to simulate a real pet actively expressing its preferences and needs for specific foods. This guides the user to prioritize feeding foods that are more suitable for the current state and / or living scenario of the pet robot, greatly improving the feeding interaction experience between the user and the pet robot, and making the feeding interaction of the pet robot more realistic, rich, and interesting.

[0046] Furthermore, since the food recommendations are combined with the pet robot's real-time status and / or living scenarios, the feeding process of the pet robot can display food recommendation information to help child users learn to prioritize the appropriate foods for different statuses and living scenarios, thereby guiding the development of healthy eating habits in child users.

[0047] Based on the above embodiments, the real-time status includes at least one of hunger level, mood value, and health status; the life scenario includes seasonal scenarios and / or cultural event scenarios.

[0048] Specifically, the real-time status of a pet robot can include at least one of hunger level, mood value, and health status.

[0049] Hunger level can characterize the current energy reserves of a pet robot, and the level of hunger directly affects the pet robot's preference for specific types of food. For example, when a pet robot is in a state of high hunger, its physical condition simulates the physiological mechanism of the human body, and its demand for nutrients, especially protein, increases. At this time, the first fit between protein-based foods and the pet robot's hunger level will be significantly increased, thus making it more inclined to recommend such foods that can increase satiety.

[0050] Mood scores characterize a pet robot's current emotional state, such as excitement, calmness, or depression. Different mood scores influence food preferences in different ways. For example, when a pet robot's mood score is low and it is depressed, carbohydrates or fruits can raise blood sugar levels and stimulate insulin secretion, thus improving mood. Therefore, these foods have a higher first fit with the mood score, leading to a preference for recommending them to boost the mood score.

[0051] Health status can characterize the physical health of a pet robot, such as whether it is sick or active. Health status also affects the pet robot's food preferences. For example, when a pet robot is sick, it may prefer bland food and actively avoid greasy food. In this case, the first fit between bland food and a healthy state is higher than that between greasy food and a healthy state.

[0052] In addition, the living scenarios in which pet robots operate can include seasonal scenarios and / or cultural event scenarios.

[0053] Among them, seasonal scenes can represent scene information that adaptively adjusts according to real-world seasonal changes. For example, in a winter scene, hot foods have a higher degree of second adaptability to the seasonal scene; while in a summer scene, cold foods have a higher degree of second adaptability to the seasonal scene.

[0054] Cultural event scenarios can represent scenes associated with specific socio-cultural attributes or background stories. For example, in the Spring Festival scenario, the second fit between dumplings and cultural event scenarios will be significantly improved; in the Lantern Festival scenario, the second fit between glutinous rice balls and other foods will be significantly improved; in the birthday scenario preset by the pet robot, the second fit between birthday cakes, longevity noodles, and other foods will be significantly improved. This embodiment of the invention does not specifically limit these aspects.

[0055] In this embodiment of the invention, by refining the real-time status of the pet robot into hunger level, mood value, and health status, and refining the life scenarios into seasonal scenarios and cultural event scenarios, a realistic pet robot food preference is formed from multiple dimensions such as physiological simulation, emotional feedback, environmental perception, and social culture. This makes the food recommendation mechanism for the pet robot more realistic and reasonable, and further optimizes the feeding experience of the pet robot.

[0056] Based on any of the above embodiments, in step 110, determining the recommendation score for each type of food based on the first degree of fit between various types of food and the real-time status of the pet robot, and / or the second degree of fit between various types of food and the living environment of the pet robot, includes: Based on the first fit between the nutritional properties of each type of food and at least one of the hunger and mood values ​​of the pet robot, the mood coefficient of each type of food is determined. Based on the first degree of fit between the health intervention attributes of the various types of food and the health status of the pet robot, the health coefficient of the various types of food is determined. Based on the second degree of fit between the suitable temperature attribute of the various types of food and the seasonal scene in which the pet robot is located, the seasonal bonus coefficient of the various types of food is determined. Based on the second degree of fit between the various types of food and the cultural event scenario in which the pet robot is located, the event bonus coefficient of the various types of food is determined; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient.

[0057] Specifically, after obtaining the hunger and / or mood values ​​of the pet robot in real time, the hunger and / or mood values ​​can be matched with the nutritional attributes of various foods to obtain the first fit. Then, the strength of the improvement and soothing effect of the nutritional components contained in various foods on the current hunger and / or mood values ​​of the pet robot can be calculated, that is, the mood coefficient of various foods can be obtained.

[0058] Nutritional composition attributes refer to the classification of food in terms of its component composition, such as protein, carbohydrate / dessert, and high-energy foods.

[0059] In some embodiments, when the pet robot's hunger level is greater than the hunger threshold, the first fit between protein-based foods and this state is extremely high, and its mood coefficient can be set to have a high multiplier bonus. When the pet robot's mood value is lower than the first mood threshold, i.e., when the pet robot is depressed, carbohydrate-based or dessert-based foods can cause blood sugar to rise and thus produce a feeling of happiness, and their mood coefficient multiplier bonus can be set to a high multiplier bonus. When the mood value is greater than the second mood threshold, i.e., when the pet robot is in an excited state, high-energy-based foods have a higher first fit with the current state, and the mood coefficient of these foods can be set to a multiplier bonus. When the mood value is greater than the third mood threshold and less than the fourth mood threshold, i.e., when the pet robot is in a calm state, the first fit between any type of food and the current state is consistent, and the mood coefficient of all foods is the base value.

[0060] For example, the hunger threshold can be 70%. When the hunger level is greater than 70%, the first suitability of protein-based foods can be set to 1.5, and the mood coefficient is the base score × 1.5. The first mood threshold can be 30%. When the mood value is less than 30%, the first suitability of carbohydrates or desserts can be set to 1.8, and the mood coefficient is the base score × 1.8. The second mood threshold can be 80%. When the mood value is greater than 80%, the first suitability of high-energy foods can be 1.3, and the mood coefficient is the base score × 1.3. The third mood threshold can be 40%, and the fourth mood threshold can be 60%. When the mood value is between 40% and 60%, the first suitability of all types of foods is 1, and the mood coefficient is the base score × 1.

[0061] It should be noted that the base score here is preset. The base score for different foods can be the same or different. For example, the base score for all foods can be 0.7.

[0062] After obtaining the real-time health status of the pet robot, the health status can be matched with the health intervention attributes of various foods to obtain the first fit. Then, the strength of the effect of the health intervention attributes contained in various foods on improving and maintaining the current health status of the pet robot can be calculated, that is, the health coefficient of various foods can be obtained.

[0063] Among them, health intervention attributes refer to the potential impact of food on the body's health or the applicable category classification, such as light food, oily food, and high-energy food. For example, rice and vegetables are considered light foods, while batteries and meat are considered high-energy foods.

[0064] In some embodiments, when the pet robot's health status is sick, the first fit between food with the health intervention attribute of bland food and this status is extremely high, and its health coefficient can be set to have a high multiplier bonus, while the first fit between food with the health intervention attribute of oily food and this status is extremely low, and its health coefficient can be set to a low multiplier reduction; when the pet robot is active, the first fit between food with the health intervention attribute of high-energy food and the current status is even higher, and the health coefficient of this type of food can be set to have a multiplier bonus; when the pet robot's health status is normal, the first fit between food with any health intervention attribute and the current status is consistent, and the health coefficient of all foods is the base value.

[0065] For example, when the pet robot is sick, the first suitability score for bland food can be set to 2.0, and the health coefficient for bland food is 2.0 times the base score. The first suitability score for oily food can be set to 0.3, and the health coefficient for oily food is 0.3 times the base score. When the pet robot is active, the first suitability score for high-energy food can be set to 1.5, and the health coefficient for high-energy food is 1.5 times the base score. When the pet robot is in a normal healthy state, the first suitability score for all types of food is 1, and the health coefficient for all types of food is set to 1 by default.

[0066] After obtaining the seasonal scene of the pet robot's living environment, the seasonal scene can be matched with the suitable temperature attributes of various foods to obtain a second degree of suitability. Then, the suitability of the suitable temperature attributes of various foods for the pet robot's current seasonal environment can be calculated, that is, the seasonal bonus coefficient of various foods can be obtained.

[0067] Among them, the suitable temperature attribute refers to the temperature at which food is generally suitable for consumption, such as hot food, cold food, and general food. For example, hot soup and roasted meat are hot food, while ice cream and fruit salad are cold food.

[0068] In some embodiments, when the pet robot is in a winter scene, the second compatibility between hot food with a suitable temperature attribute and the scene is high, and its seasonal bonus coefficient can be set to a positive random numerical bonus; when the pet robot is in a summer scene, the second compatibility between cold food with a suitable temperature attribute and the scene is high, and its seasonal bonus coefficient can also be set to a positive random numerical bonus.

[0069] For example, when the seasonal scenario is winter, the seasonal bonus coefficient for hot food can be set to a randomly generated value between 0.2 and 0.5 added to the base value; when the seasonal scenario is summer, the seasonal bonus coefficient for cold food can be set to a randomly generated value between 0.3 and 0.6 added to the base value; in non-corresponding seasonal scenarios, the seasonal bonus coefficient for food can be the base value.

[0070] After obtaining the cultural event scenarios in the pet robot's living environment, the cultural event scenarios can be matched with various foods to obtain a second fit. Then, the degree of fit of various foods with the pet robot's current specific cultural background or holidays can be calculated, that is, the event bonus coefficient of various foods can be obtained.

[0071] In some embodiments, when the cultural event scenario in which the pet robot is located is a specific festival or anniversary, the second compatibility between a certain type of food and the cultural event scenario is extremely high, and its event bonus coefficient can be set to a positive numerical bonus; while for food that is in the cultural event scenario, or when it is not currently in a specific cultural event scenario, the second compatibility between the food and the scenario is at a basic level, and the event bonus coefficient of all types of food is a basic value.

[0072] For example, when the cultural event is the Spring Festival, the event bonus coefficient for dumplings can be set to increase by 1.2 on the base value; when the cultural event is a birthday, the event bonus coefficient for longevity noodles and birthday cake can be set to increase by 0.8 on the base value; and when the cultural event is the Lantern Festival, the event bonus coefficient for glutinous rice balls and sweet dumplings can be set to increase by 1.0 on the base value.

[0073] After calculating at least one of the following factors—mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient—for each type of food, a recommended score can be determined based on this factor. For example, at least one of these factors can be integrated using addition, multiplication, or other arithmetic operations to obtain the recommended score.

[0074] In this embodiment of the invention, by matching the nutritional components, health intervention attributes, and suitable temperature attributes of food with the pet robot's hunger level, mood value, health status, and the seasonal and cultural event scenarios it is in, a mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient are obtained, and a recommendation score is determined accordingly, which effectively improves the objectivity and rationality of the recommendation score.

[0075] Based on any of the above embodiments, in step 110, determining the recommendation score for each type of food based on at least one of the mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient includes: Based on the number of consecutive feedings of each type of food, the continuous penalty coefficient for each type of food is determined; Randomly generate random perturbation coefficients for various types of food; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient, as well as the continuous penalty coefficient and / or random perturbation coefficient of each type of food.

[0076] Specifically, for any food, the number of consecutive feedings refers to the number of times the pet robot is continuously fed that food within a pre-set time window or feeding cycle. Considering that in real pet feeding, if a pet eats the same food continuously, it often becomes bored and refuses to eat, this embodiment determines a continuous penalty coefficient based on the number of consecutive feedings to simulate this situation. This continuous penalty coefficient can be understood as a negative intervention value calculated based on the number of consecutive feedings.

[0077] For example, the continuous penalty coefficient can be set as: the number of times the food has been repeatedly fed in the most recent N feedings multiplied by a preset penalty step size. Here, N can be a preset value, such as 3; the preset penalty step size can be determined based on the actual situation, such as 0.5.

[0078] In addition, if the same food is fed N times in a row, that food can be eliminated and avoided from being recommended.

[0079] Furthermore, to add randomness to the behavior of the pet robot, so that even in the exact same state and scenario, the food the pet robot wants to eat may vary slightly each time, this embodiment of the invention can use a randomly generated perturbation value as a random perturbation coefficient for various types of food. For example, the random perturbation coefficient can be taken from a normally distributed random value, such as a normally distributed random value within a distribution interval with expected value μ=0 and standard deviation σ=0.3.

[0080] Therefore, the recommended score for each type of food can be determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient, as well as the continuous penalty coefficient and / or random disturbance coefficient. For example, the recommended score can be calculated using the following formula: Recommended Score = Base Mood Score × Mood Coefficient + Base Health Score × Health Coefficient + Seasonal Bonus Coefficient + Event Bonus Coefficient - Continuous Penalty Coefficient + Random Disturbance Coefficient.

[0081] For example, suppose the pet robot's mood is 25%, it is in a normal healthy state, it is in a winter scene, a non-cultural event scene, and it has recently eaten hamburgers twice in a row.

[0082] Taking apples as an example, apples are desserts, suitable for situations where one is in a low mood. Assuming the base score used for calculating the mood coefficient is 0.8, the mood coefficient is 0.8 × 1.8 = 1.44. The health coefficient is 1 by default. There is no seasonal bonus for eating non-hot food in winter, that is, the seasonal bonus coefficient is 0, and the event bonus coefficient is 0. In addition, apples are not fed repeatedly, so the continuous penalty coefficient is 0. The randomly generated random disturbance coefficient is 0.2. Therefore, the recommended score for apples is 1.44 + 1.0 + 0 + 0 - 0 + 0.2 = 2.64.

[0083] Based on any of the above embodiments, the food recommendation method further includes: In response to a food selection operation, the actual food to be fed to the pet robot is determined; Based on the actual food fed to the pet robot, determine the feedback behavior information of the pet robot in response to the actual food fed to the pet robot; Display the feedback behavior information.

[0084] Specifically, after displaying the pet robot's food recommendation information, users can refer to the information or, according to their own preferences, perform food selection operations by clicking food icons on the screen, dragging food models to the pet robot's mouth, issuing specific voice commands, or scanning physical cards, thereby achieving feeding interaction with the pet robot. In this embodiment of the invention, the aforementioned interactive actions such as clicking icons, dragging food, issuing voice commands, and scanning physical cards can all be used as food selection operations.

[0085] When the pet robot is in the feeding interaction interface, it can monitor the user's input in real time. Once a food selection operation is detected, it can be parsed, and the corresponding food can be matched from a preset food database and identified as the actual food fed to the pet robot in this interaction. It should be noted that the actual food fed can be the food with the highest recommendation score shown in the food recommendation information, or it can be a non-recommended food that the user chooses on their own, ignoring the food recommendation information. This embodiment of the invention does not make specific limitations in this regard.

[0086] Once the actual food is determined, the pet robot's feedback behavior in response to that food can be identified. This feedback behavior information can be the pet robot's external behaviors after consuming the food, including changes in appearance, facial expressions, and sound effects, based on the type or properties of the food. This feedback behavior information can simulate the physiological or emotional reactions of a pet after eating different foods.

[0087] In some embodiments, corresponding feedback behavior information can be pre-set for each type of food, so that after determining the actual food being fed, the feedback behavior information corresponding to that actual food can be directly invoked. For example, the feedback behavior information set for rice could be lying down to rest, showing satisfaction; the feedback behavior information set for a hamburger could be running around, showing energy; the feedback behavior information set for a hydrogen-powered hamburger could be full of energy, running faster or full of vitality; and the feedback behavior information set for lemon could be showing discomfort.

[0088] After obtaining feedback behavior information from the pet robot in response to actual food feeding, the feedback behavior information can be presented to the user through the pet robot's hardware output module. For example, animations of running around or lying down to rest can be played on the screen, along with happy or uncomfortable sounds. Alternatively, the user can control the pet robot's body or head to make corresponding body movements or facial expressions.

[0089] In this embodiment of the invention, the actual food is determined by responding to the user's food selection operation, and differentiated behavioral feedback information is displayed based on the type of actual food, thereby enhancing the realism and fun of human-computer interaction.

[0090] Based on any of the above embodiments, the food recommendation method further includes: The real-time status of the pet robot is updated based on the actual food fed to it.

[0091] Specifically, food attributes that affect the pet robot's state can be pre-configured for various types of food. Therefore, after determining the actual food to be fed, the corresponding food attributes can be extracted, and the pet robot's current real-time state can be updated based on these attributes. For example, the food attributes and real-time state can be input into a pre-trained real-time state update model, which will update the input real-time state and output a new real-time state.

[0092] For example, suppose a pet robot is currently in a state of high hunger and urgently needs to be fed. If the user feeds it a high-energy food such as a hamburger or meat, the pet robot's hunger level will decrease significantly. On the other hand, if the pet robot is currently sick and the user continues to feed it greasy barbecue, its health may deteriorate further, and its recovery period may become longer.

[0093] In this embodiment of the invention, the real-time status of the pet robot is updated by actually feeding it food, thereby realizing the correlation between the real-time status changes of the pet robot and the user's feeding behavior, and enhancing the realism of human-computer interaction.

[0094] Based on any of the above embodiments, updating the real-time status of the pet robot based on the actual food fed includes: Based on the differences in food attributes between the actual feeding food and the target feeding food, the recommendation deviation of the actual feeding food is determined; the target feeding food is the food with the highest recommendation score among the various food categories. The real-time status of the pet robot is updated based on the recommended deviation.

[0095] Specifically, after obtaining recommendation scores for various types of food, the food with the highest recommendation score can be identified, which is denoted as the target feeding food in this embodiment of the invention. The target feeding food is the food that best matches the pet robot's current real-time state and living environment.

[0096] After obtaining the target food and the actual food fed, the differences in food attributes between the two can be calculated. Here, the difference in food attributes refers to the numerical or characteristic gap between the various attribute indicators of the actual food fed by the user and the corresponding attribute indicators of the target food. For example, this may include differences in energy values, differences in mood impact values, etc.

[0097] After obtaining the differences in food attributes between the actual food fed and the target food, the recommended deviation of the actual food from the target food can be calculated. Here, the recommended deviation quantifies the degree to which the actual food deviates from the target food. The smaller the recommended deviation, the more the actual food fed by the user meets the needs of the pet robot; the larger the recommended deviation, the greater the gap between the actual food fed by the user and the pet robot's expectations.

[0098] Considering that a pet robot might experience disappointment and a decrease in mood if it doesn't receive the food it expects, its real-time status can be updated based on the recommendation deviation.

[0099] Furthermore, the pet robot's mood value can be updated based on the recommendation deviation. For example, assuming the recommendation deviation is... If the user directly feeds the target food, that is If the pet robot is extremely satisfied, its mood value will increase by an additional 2; if the deviation from the recommended food for actual feeding is minimal, that is... If the pet robot's mood value increases slightly, it gains an additional +1; if the deviation from the recommended food is moderate, that is... If the pet robot's mood value remains constant, it will receive an additional +0; if the actual food fed deviates significantly from the recommended amount, i.e. If the pet robot is slightly disappointed, its mood value will decrease by 1; if the actual food fed deviates significantly from the recommended food, that is... If this happens, the pet robot will be extremely dissatisfied, its mood value will drop significantly, and it will receive an additional -2.

[0100] In this embodiment of the invention, the real-time status of the pet robot is updated based on the deviation of the target feeding food from the recommended actual feeding food, so that the pet robot can provide status feedback when the actual feeding differs from the expectation, thereby enhancing the realism of human-computer interaction.

[0101] Based on any of the above embodiments, the food attributes include energy intake value and / or mood optimization value.

[0102] Specifically, energy intake value refers to a quantitative indicator that reflects how much physiological replenishment food can provide to a pet robot. The level of energy intake value is directly related to the pet robot's ability to recover from real-time physiological states such as hunger, vitality, and health. For example, hydrogen batteries typically have a high energy intake value, quickly alleviating hunger; while lemons have a low energy intake value and are not very satiating.

[0103] Mood optimization value is a quantitative indicator that reflects the level of emotional pleasure or psychological satisfaction that food can bring to a pet robot. The sign and absolute value of the mood optimization value are directly related to the degree of intervention in the real-time emotional state of the pet robot, such as its mood score. For example, desserts usually have a high positive mood optimization value, while foods with unusual tastes or that the pet robot dislikes may have a very low mood optimization value, or even a negative value.

[0104] When calculating the difference in food attributes between the actual food fed and the target food fed, you can calculate only the difference in energy intake values, or only the difference in mood optimization values, or calculate the differences in energy intake values ​​and mood optimization values ​​separately.

[0105] For example, suppose the target food is a hydrogen fuel cell, with an energy intake value of 100 and a mood optimization value of 80; the actual food fed is a lemon, with an energy intake value of 10 and a mood optimization value of -10. When calculating the difference in food attributes between these two foods, the difference in energy intake value can be calculated as follows: And the differences in the dimensions of calculating the mood optimization value. .

[0106] Based on this, the deviation from the recommended food for actual feeding can be calculated using the following formula: Substituting the energy and mood differences between the hydrogen battery and the lemon into the above formula for calculating the recommendation deviation, we get the lemon's recommendation deviation compared to the hydrogen battery: (90 + 90) / 200 = 0.9. A very large recommendation deviation will trigger negative emotional feedback from the pet robot.

[0107] In this embodiment of the invention, the impact of different foods on the pet robot is measured from two dimensions: energy intake value and mood optimization value. This ensures the rationality and realism of the recommendation deviation calculation, thereby providing a basis for the pet robot to exhibit emotional changes that conform to real biological logic.

[0108] Based on any of the above embodiments, this invention provides a food recommendation method, which may include the following steps: First, attribute labels are created for each type of food in the food database. For example, for food-based hydrogen fuel cells, the following attributes can be labeled: Name: Battery; Type: High energy, special; Base score: 0.7; Season type: General; Cultural tags: None.

[0109] In addition, the system can monitor the pet robot's real-time status and living environment, as well as update the number of consecutive feedings for various foods. For example, if the real-time status is "low mood," the living environment includes winter scenes and culturally inactive events, and the consecutive feeding frequency for hamburgers is 2 times.

[0110] Based on this, we can calculate the mood coefficient, health coefficient, seasonal bonus coefficient, event bonus coefficient, continuous penalty coefficient, and random disturbance coefficient of various foods under their current real-time state and living scenario, thereby obtaining the recommendation score for each type of food. For example, the recommendation score for rice is 0.2, the recommendation score for mango is 1.1, and the recommendation score for baguette is 0.7, etc.

[0111] After obtaining recommendation scores for various foods, the food with the highest recommendation score can be identified as the target feeding food. During the feeding interaction between the user and the pet robot, the target feeding food is displayed as food recommendation information, thereby showing the user the pet robot's need for the target feeding food, so as to guide the user to prioritize feeding the pet robot with the target feeding food.

[0112] During the feeding interaction, the system can detect the user's food selection actions to determine the actual food to be fed. Furthermore, based on the feedback behavior information corresponding to the actual food fed, the pet robot displays this feedback behavior information, allowing the pet robot to exhibit different behaviors when the user feeds it different foods.

[0113] Furthermore, based on the differences in food attributes between the actual food fed and the target food, the recommended deviation of the actual food compared to the target food can be determined, and feedback behavior information and the pet robot's mood value can be determined based on the recommended deviation. For example, if the target food is a hydrogen battery and the actual food is a lemon, assuming the recommended deviation of the lemon is 0.9, the pet robot will play a rejection animation, and its mood value will decrease further.

[0114] In this embodiment of the invention, a recommendation score is calculated for each type of food based on a first degree of fit between various foods and the real-time status of the pet robot and / or a second degree of fit between various foods and the living environment of the pet robot. Based on the recommendation score, food recommendation information is displayed to the user, enabling the pet robot to simulate a real pet actively expressing its preferences and needs for specific foods. This guides the user to prioritize feeding foods that are more suitable for the current status and / or living environment of the pet robot, greatly improving the feeding interaction experience between the user and the pet robot, and making the feeding interaction of the pet robot more realistic, rich, and interesting.

[0115] Furthermore, since the food recommendations are combined with the pet robot's real-time status and / or living scenarios, the feeding process of the pet robot can display food recommendation information to help child users learn to prioritize the appropriate foods for different statuses and living scenarios, thereby guiding the development of healthy eating habits in child users.

[0116] Furthermore, by displaying differentiated behavioral feedback information through actual feeding and updating the real-time status of the pet robot, the system establishes a correlation between the real-time status changes of the pet robot and the user's feeding behavior, thereby enhancing the realism of human-computer interaction.

[0117] The food recommendation device provided by the present invention will be described below. The food recommendation device described below can be referred to in correspondence with the food recommendation method described above.

[0118] Figure 2 This is a schematic diagram of the food recommendation device provided by the present invention, as shown below. Figure 2 As shown, the device includes: The calculation unit 210 is used to determine the recommendation score of each type of food based on the first degree of fit between the real-time status of each type of food and the pet robot, and / or the second degree of fit between each type of food and the living scenario in which the pet robot is located. The recommendation unit 220 is used to display food recommendation information for the pet robot based on the recommendation scores of the various types of food.

[0119] In the food recommendation device provided in this embodiment of the invention, a recommendation score is calculated for each type of food based on a first degree of fit between various types of food and the real-time state of the pet robot and / or a second degree of fit between various types of food and the living scenario in which the pet robot is located. Based on the recommendation score, food recommendation information is displayed to the user, enabling the pet robot to simulate a real pet actively expressing its preference and need for specific foods. This guides the user to prioritize feeding foods that are more suitable for the current state and / or living scenario of the pet robot, greatly improving the feeding interaction experience between the user and the pet robot, and making the feeding interaction of the pet robot more realistic, rich, and interesting.

[0120] Based on the above embodiments, the real-time status includes at least one of hunger level, mood value, and health status; the life scenario includes seasonal scenarios and / or cultural event scenarios.

[0121] Based on any of the above embodiments, the computing unit is specifically used for: Based on the first fit between the nutritional properties of each type of food and at least one of the hunger and mood values ​​of the pet robot, the mood coefficient of each type of food is determined. Based on the first degree of fit between the health intervention attributes of the various types of food and the health status of the pet robot, the health coefficient of the various types of food is determined. Based on the second degree of fit between the suitable temperature attribute of the various types of food and the seasonal scene in which the pet robot is located, the seasonal bonus coefficient of the various types of food is determined. Based on the second degree of fit between the various types of food and the cultural event scenario in which the pet robot is located, the event bonus coefficient of the various types of food is determined; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient.

[0122] Based on any of the above embodiments, the computing unit is specifically used for: Based on the number of consecutive feedings of each type of food, the continuous penalty coefficient for each type of food is determined; Randomly generate random perturbation coefficients for various types of food; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient, as well as the continuous penalty coefficient and / or random perturbation coefficient of each type of food.

[0123] Based on any of the above embodiments, the device further includes a feedback unit, used for: In response to a food selection operation, the actual food to be fed to the pet robot is determined; Based on the actual food fed to the pet robot, determine the feedback behavior information of the pet robot in response to the actual food fed to the pet robot; Display the feedback behavior information.

[0124] Based on any of the above embodiments, the feedback unit is further configured to: The real-time status of the pet robot is updated based on the actual food fed to it.

[0125] Based on any of the above embodiments, the feedback unit is specifically used for: Based on the differences in food attributes between the actual feeding food and the target feeding food, the recommendation deviation of the actual feeding food is determined; the target feeding food is the food with the highest recommendation score among the various food categories. The real-time status of the pet robot is updated based on the recommended deviation.

[0126] Based on any of the above embodiments, the food attributes include energy intake value and / or mood optimization value.

[0127] Figure 3 An example is a schematic diagram of the physical structure of a pet robot, such as... Figure 3 As shown, the pet robot may include a pet robot body 300, and a processor 310, a communication interface 320, a memory 330, and a communication bus 340 disposed inside the pet robot body 300. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a food recommendation method, which includes: Based on the first degree of fit between various foods and the real-time status of the pet robot, and / or the second degree of fit between various foods and the living scenario of the pet robot, the recommendation score of each type of food is determined. Based on the recommendation scores for each type of food, the pet robot displays food recommendation information.

[0128] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to related technologies, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0129] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the food recommendation method provided by the above methods, the method comprising: Based on the first degree of fit between various foods and the real-time status of the pet robot, and / or the second degree of fit between various foods and the living scenario of the pet robot, the recommendation score of each type of food is determined. Based on the recommendation scores for each type of food, the pet robot displays food recommendation information.

[0130] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the food recommendation method provided by the methods described above, the method comprising: Based on the first degree of fit between various foods and the real-time status of the pet robot, and / or the second degree of fit between various foods and the living scenario of the pet robot, the recommendation score of each type of food is determined. Based on the recommendation scores for each type of food, the pet robot displays food recommendation information.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0132] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A food recommendation method, characterized in that, include: Based on the first degree of fit between various foods and the real-time status of the pet robot, and / or the second degree of fit between various foods and the living scenario of the pet robot, the recommendation score of each type of food is determined. Based on the recommendation scores for each type of food, the pet robot displays food recommendation information.

2. The food recommendation method according to claim 1, characterized in that, The real-time status includes at least one of hunger level, mood value, and health status; the life scenario includes seasonal scenarios and / or cultural event scenarios.

3. The food recommendation method according to claim 2, characterized in that, The recommendation score for each type of food is determined based on a first degree of fit between various foods and the real-time status of the pet robot, and / or a second degree of fit between various foods and the living environment of the pet robot, including: Based on the first fit between the nutritional properties of each type of food and at least one of the hunger and mood values ​​of the pet robot, the mood coefficient of each type of food is determined. Based on the first degree of fit between the health intervention attributes of the various types of food and the health status of the pet robot, the health coefficient of the various types of food is determined. Based on the second degree of fit between the suitable temperature attribute of the various types of food and the seasonal scene in which the pet robot is located, the seasonal bonus coefficient of the various types of food is determined. Based on the second degree of fit between the various types of food and the cultural event scenario in which the pet robot is located, the event bonus coefficient of the various types of food is determined; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient.

4. The food recommendation method according to claim 3, characterized in that, The recommendation score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient. This includes: Based on the number of consecutive feedings of each type of food, the continuous penalty coefficient for each type of food is determined; Randomly generate random perturbation coefficients for various types of food; The recommended score for each type of food is determined based on at least one of the following: mood coefficient, health coefficient, seasonal bonus coefficient, and event bonus coefficient, as well as the continuous penalty coefficient and / or random perturbation coefficient of each type of food.

5. The food recommendation method according to any one of claims 1 to 4, characterized in that, Also includes: In response to a food selection operation, the actual food to be fed to the pet robot is determined; Based on the actual food fed to the pet robot, determine the feedback behavior information of the pet robot in response to the actual food fed to the pet robot; Display the feedback behavior information.

6. The food recommendation method according to claim 5, characterized in that, Also includes: The real-time status of the pet robot is updated based on the actual food fed to it.

7. The food recommendation method according to claim 6, characterized in that, The step of updating the real-time status of the pet robot based on the actual food fed includes: Based on the differences in food attributes between the actual feeding food and the target feeding food, the recommendation deviation of the actual feeding food is determined; the target feeding food is the food with the highest recommendation score among the various food categories. The real-time status of the pet robot is updated based on the recommended deviation.

8. The food recommendation method according to claim 7, characterized in that, The food attributes include energy intake values ​​and / or mood optimization values.

9. A food recommendation device, characterized in that, include: The calculation unit is used to determine the recommendation score of each type of food based on the first degree of fit between the real-time status of each type of food and the pet robot, and / or the second degree of fit between each type of food and the living scenario in which the pet robot is located. The recommendation unit is used to display food recommendation information for the pet robot based on the recommendation scores of the various types of food.

10. A pet robot, comprising a pet robot body, a memory, a processor, and a computer program stored in the memory and running on the processor, wherein, When the processor executes the computer program, it implements the food recommendation method as described in any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the food recommendation method as described in any one of claims 1 to 8.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the food recommendation method as described in any one of claims 1 to 8.

Images