Method and system for recommending recipes

Abstract

The application provides a recipe recommendation method and system, and relates to the technical field of smart home, the method comprises the following steps: obtaining sleep data, determining the sleep state of a target user based on the sleep data; inputting the sleep state into a pre-established association model, outputting recommended nutritional elements and recipe information containing the nutritional elements through the association model; and generating recipe recommendation information containing the nutritional elements and the recipe information. The recipe recommendation method and system provided by the application can generate recipe recommendation information according to the sleep state, and the sleep data is obtained by monitoring the sleep of the target user through a smart sleep monitoring device, so that the generated recipe recommendation information can be targeted for the current health state of the user, can be linked with the health state of the user, can realize personalized diet recommendation, and can further improve the user experience.

Description

Technical Field

[0001] This invention relates to the technical field of smart homes, and in particular to a method and system for recommending recipes. Background Technology

[0002] Existing food recommendation systems mostly rely on static user input, such as users' physiological characteristics like height, weight, and gender, as well as taste preferences like sour, sweet, and spicy.

[0003] Therefore, the current method of making dietary recommendations based on users' static input of physiological characteristics and taste preferences is difficult to form dietary recommendations based on sleep quality, and the linkage with users' health data is relatively weak, making it difficult to improve the user experience. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a recipe recommendation method and system that uses sleep monitoring as an entry point and combines health status to drive recipe recommendations, thereby helping to improve users' quality of life and health management level, and thus alleviating the above-mentioned technical problems.

[0005] In a first aspect, embodiments of the present invention provide a recipe recommendation method, the method comprising: acquiring sleep data, wherein the sleep data is obtained by a smart sleep monitoring device monitoring the sleep of a target user; determining the sleep state of the target user based on the sleep data; inputting the sleep state into a pre-established association model, outputting recommended nutritional elements through the association model, and recipe information containing the nutritional elements; wherein the association model is a model that associates sleep state with recipe recommendation information; and generating recipe recommendation information containing the nutritional elements and the recipe information.

[0006] In conjunction with the first aspect, the present invention provides a first possible implementation of the first aspect, wherein the step of determining the sleep state of the target user based on the sleep data includes: extracting sleep feature data from the sleep data; and determining the sleep state of the target user based on the sleep feature data.

[0007] In conjunction with the first possible implementation of the first aspect, this embodiment of the invention provides a second possible implementation of the first aspect, wherein the step of determining the sleep state of the target user based on the sleep feature data includes: selecting at least one dimension of sleep indicators from the sleep feature data; calculating a sleep score based on at least one of the sleep indicators; obtaining a sleep level based on the score range to which the sleep score belongs; and generating a sleep state including the sleep level.

[0008] In conjunction with the second possible implementation of the first aspect, this embodiment of the invention provides a third possible implementation of the first aspect, wherein the step of calculating a sleep score based on at least one of the sleep indicators includes: obtaining a pre-configured weight parameter for each of the sleep indicators; and performing a weighted calculation on each of the sleep indicators based on the weight parameter to obtain the sleep score.

[0009] In conjunction with the second possible implementation of the first aspect, this embodiment of the invention provides a fourth possible implementation of the first aspect, wherein the step of determining the sleep state of the target user based on the sleep feature data further includes: obtaining pre-configured judgment conditions for each sleep state; and determining the sleep state of the target user based on the judgment conditions satisfied by the sleep feature data.

[0010] In conjunction with the fourth possible implementation of the first aspect, this embodiment of the invention provides a fifth possible implementation of the first aspect, wherein the dimensions of the sleep characteristic data required for the above-mentioned determination conditions include at least deep sleep percentage data, total sleep duration data, number of awakenings, sleep efficiency, and snoring index; the step of determining the sleep state of the target user based on the determination conditions satisfied by the sleep characteristic data includes: if the deep sleep percentage data is less than a preset percentage threshold, and the total sleep duration data is a preset sleep duration threshold, then the sleep state of the target user is determined to be deep sleep-deficient; if the number of awakenings is greater than a preset number of awakenings threshold, and the sleep efficiency is less than a preset efficiency threshold, then the sleep state of the target user is determined to be frequent awakening; if the snoring index is greater than a preset index threshold, then the sleep state of the target user is determined to be suspected of respiratory abnormality; if none of the sleep characteristic data satisfies any of the determination conditions, then the sleep state of the target user is determined to be well-recovered.

[0011] In conjunction with the fifth possible implementation of the first aspect, this embodiment of the invention provides a sixth possible implementation of the first aspect, wherein the aforementioned association model is a large language model; the steps of inputting the sleep state into a pre-established association model, outputting recommended nutritional elements through the association model, and providing recipe information containing the nutritional elements include: generating natural language prompt information for the association model based on the sleep state, the sleep feature data corresponding to the sleep state, and the sleep score; inputting the natural language prompt information into the association model, causing the association model to generate recipe information containing the recommended nutritional elements and recipe information containing the nutritional elements based on the natural language prompt information; wherein the recipe information includes at least a recipe name, ingredients containing the nutritional elements, and the nutritional elements.

[0012] In conjunction with the first aspect, the present invention provides a seventh possible implementation of the first aspect, wherein the above method further includes: responding to a confirmation operation on the recipe information, generating cooking parameters of the recipe recorded in the recipe information; sending the cooking parameters to a cooking device, and controlling the cooking device to cook according to the cooking parameters.

[0013] In conjunction with the first aspect, this embodiment of the invention provides an eighth possible implementation of the first aspect, wherein the above method further includes: acquiring sleep tracking data corresponding to the sleep data; comparing the sleep data and the sleep tracking data, and generating feedback information corresponding to the recipe recommendation information based on the comparison result.

[0014] Secondly, embodiments of the present invention also provide a recipe recommendation system, including a backend server, and a cooking device and an intelligent sleep monitoring device communicating with the backend server; wherein the backend server is configured with a recipe recommendation device; the recipe recommendation device includes: an acquisition module for acquiring sleep data, wherein the sleep data is obtained by the intelligent sleep monitoring device monitoring the sleep of a target user; a determination module for determining the sleep state of the target user based on the sleep data; a recommendation module for inputting the sleep state into a pre-established association model, outputting recommended nutritional elements and recipe information containing the nutritional elements through the association model; wherein the association model is a model that associates sleep state with recipe recommendation information; and a generation module for generating recipe recommendation information containing the nutritional elements and the recipe information.

[0015] The embodiments of the present invention bring the following beneficial effects: This invention provides a recipe recommendation method and system that can acquire sleep data; determine the sleep state of a target user based on the sleep data; extract sleep feature data; determine the sleep state based on the sleep feature data; input the sleep state into a pre-established association model; output recommended nutritional elements and recipe information containing the nutritional elements through the association model; and generate recipe recommendation information containing nutritional elements and recipe information. Furthermore, the aforementioned sleep data is obtained by an intelligent sleep monitoring device monitoring the sleep of the target user. Therefore, the generated recipe recommendation information can be targeted to the user's current health status, can be linked with the user's health status to achieve personalized dietary recommendations, and thus improve the user experience.

[0016] 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 are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A structural block diagram of a recipe recommendation system provided in an embodiment of the present invention; Figure 2 A flowchart illustrating a recipe recommendation method provided in an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a recipe recommendation device provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Currently, smart homes are gradually developing and becoming integrated with users' health management systems, such as providing dietary recommendations based on user status. However, most of the dietary recommendations currently implemented are in the "passive monitoring" stage, making it difficult to make dynamic interventions and achieve a true closed loop of health feedback and dietary regulation.

[0022] In addition, although the intelligence level of smart home devices such as mattresses, pillows, and wearable devices is improving and they have the ability to connect to the network and customize parameters, most of these smart home devices are still mainly based on "active manual control" and have relatively weak linkage with users' health data, making it difficult to achieve the ability to recommend diets and execute them automatically based on health status.

[0023] Based on this, the recipe recommendation method and system provided in this embodiment of the invention can build an automated diet recommendation function with sleep data as the entry point, and can link with smart devices such as kitchen appliances to complete personalized cooking operations.

[0024] To facilitate understanding of this embodiment, a method for recommending recipes disclosed in this embodiment of the invention will first be described in detail.

[0025] In one possible implementation, this invention provides a recipe recommendation method. This method can be applied to a backend server of a recipe recommendation system. For example, the recipe recommendation system may include a backend server, as well as cooking equipment and a smart sleep monitoring device that communicate with the backend server. Specifically, the cooking equipment in this embodiment may be a steam oven, a smart oven, a smart rice cooker, an integrated stove, or other smart devices, and has the ability to be networked and customized with parameters. The smart sleep monitoring device may be a smart pillow, a smart bracelet, or other smart wearable device. In addition, the recipe recommendation system may also include smart terminals, such as smartphones, tablets, desktop computers, PDAs, etc., used by users, and can establish communication with the backend server, cooking equipment, and smart sleep monitoring device to achieve data exchange.

[0026] For ease of understanding, Figure 1 A structural block diagram of a recipe recommendation system is shown, including a backend server 100, a cooking device 101, a smart sleep monitoring device 102, and a smart terminal 103 that communicate with the backend server 100. The backend server, also known as a cloud server, analyzes and provides feedback on the data monitored by the smart sleep monitoring device, and simultaneously makes personalized recipe recommendations. The smart terminal typically has an application (APP) installed corresponding to the recipe recommendation system to facilitate communication with the backend server. The smart sleep monitoring device is usually designed to have the ability to collect multimodal physiological signals. Taking a smart pillow as an example, it can integrate multiple high-precision sensors and communication modules to achieve non-intrusive monitoring and data reporting. For example, a smart pillow may include, but is not limited to, the following components: (1) Physiological signal acquisition module: such as pressure sensor array: placed in the core position inside the pillow, used to monitor the contact pressure distribution and dynamic changes of the user's head and neck in real time, and can calculate body movement frequency, turning behavior and sleeping posture changes. Miniature microphone module: embedded in the area close to the user's breathing area, used to continuously collect sound wave information such as snoring and breathing sounds at night, and assist in judging potential risks such as sleep apnea and airway obstruction. Non-contact physiological sensors, such as micro-motion radar, physiological micro-vibration sheet, etc., use microwave radar or piezoelectric materials to sense the user's small chest and abdominal movements, extract key indicators such as respiratory rate and heart rate changes, and realize low power consumption and non-invasive monitoring. Temperature and humidity sensor: monitors the temperature and humidity changes of the sleep environment, and provides environmental parameter reference for sleep quality analysis.

[0027] (2) Data transmission and local buffering module: For example, the pillow has an embedded Bluetooth / Wi-Fi communication module responsible for establishing a stable connection with local smart gateway devices, such as smart home speakers, smart routers, or edge boxes. To reduce power consumption, data can be temporarily stored locally using event-triggered or periodic collection strategies and uploaded to the backend server periodically. Data upload supports protocols such as MQTT or HTTPS, and has end-to-end encryption capabilities to ensure user privacy and security.

[0028] Furthermore, the backend server can analyze the sleep data, such as using large models or rule models, to ultimately realize the recipe recommendation function provided in this embodiment of the invention.

[0029] Specifically, Figure 2 A flowchart of a recipe recommendation method is shown, which includes the following steps: Step S202: Obtain sleep data; The sleep data is obtained by intelligent sleep monitoring devices monitoring the sleep of target users; for example, the data obtained by the aforementioned smart pillow monitoring the sleep of target users.

[0030] Step S204: Determine the sleep state of the target user based on sleep data; In practice, the smart sleep monitoring device monitors the sleep of the target user to obtain sleep data, including direct data from various sensors of the smart sleep monitoring device, as well as data obtained by the backend server through analysis and statistics based on the sensor data. For example, when the smart sleep monitoring device is a smart pillow, in the above step S202, the sleep data obtained is the sensor data collected by the smart pillow, as well as the analysis data generated based on the sensor data collected by the smart pillow; sleep data is generated based on the sensor data and the analysis data.

[0031] Sensor data refers to key physiological data collected directly and non-contactly by sensors, such as body movement, snoring, respiratory rhythm, and heart rate changes. Analytical data, on the other hand, refers to data obtained by the backend server through analysis and modeling. For example, the backend server can perform the signal processing capabilities required by the design, such as outlier removal and denoising, sliding window slicing and standardization, and time alignment of data from different channels. Furthermore, the backend server can access multi-dimensional feature signals uploaded by smart sleep monitoring devices through a unified data interface standard, extracting indicators with analytical value, including but not limited to: respiratory cycle and rhythm, body movement frequency and amplitude, heart rate change trends, and snoring event characteristics. Further analysis can yield data such as user sleep duration, sleep efficiency, the ratio of deep to light sleep, and the number of awakenings during the night, forming multi-dimensional sleep data.

[0032] Therefore, the sleep data in the embodiments of the present invention typically includes the data directly monitored by the intelligent sleep monitoring device and the data after certain preprocessing by the back-end server. The final data can realize multi-dimensional data analysis and thus obtain the sleep state.

[0033] Step S206: Input the sleep state into the pre-established association model, and output recommended nutritional elements and recipe information containing the nutritional elements through the association model; In this embodiment of the invention, the association model is a model that associates sleep state with recipe recommendation information; Step S208: Generate recipe recommendation information containing nutritional elements and recipe information.

[0034] This invention provides a recipe recommendation method that can acquire sleep data; determine the sleep state of a target user based on the sleep data; extract sleep feature data; determine the sleep state based on the sleep feature data; input the sleep state into a pre-established association model; output recommended nutritional elements and recipe information containing the nutritional elements through the association model; and generate recipe recommendation information containing nutritional elements and recipe information. Since the aforementioned sleep data is obtained by an intelligent sleep monitoring device monitoring the sleep of the target user, the generated recipe recommendation information can be targeted to the user's current health status, can be linked with the user's health status, realize personalized dietary recommendations, and thus improve the user experience.

[0035] In practical use, in the embodiments of the present invention, when determining the sleep state in step S204 above, a multi-dimensional determination method can be implemented. Furthermore, based on the relationship between sleep data and sleep state, the numerical range of sleep feature data corresponding to different sleep states can be pre-configured, which helps to accurately determine the sleep state.

[0036] Specifically, in this embodiment of the invention, when determining sleep state, sleep feature data can be extracted from sleep data; then, the sleep state of the target user can be determined based on the sleep feature data. Furthermore, in this embodiment of the invention, overall sleep quality can be assessed from the dimension of sleep score. Therefore, when determining sleep state, at least one dimension of sleep indicators can be selected from the sleep feature data, and a sleep score can be calculated based on the at least one dimension of sleep indicators; a sleep level can be obtained based on the score range to which the sleep score belongs; and a sleep state including the sleep level can be generated. Moreover, when actually calculating the sleep score, it is necessary to pre-configure the weight parameters of each sleep indicator. Then, when calculating the sleep score, the pre-configured weight parameters of each sleep indicator can be obtained; and a weighted calculation is performed on each sleep indicator based on the weight parameters to obtain the sleep score.

[0037] For example, in daily life, people are often concerned about sleep quality and breathing during sleep. Therefore, multiple dimensions of sleep indicators can be determined based on sleep quality and breathing, such as restorative dimension, maintenance dimension, normal breathing dimension, emotional and brain recovery dimension, and physiological state dimension, etc. For ease of understanding, Table 1 below shows the settings of different dimensions of sleep indicators. Table 1 also gives the weight parameters corresponding to each sleep indicator, as well as the meaning of each dimension, as shown in Table 1 below: Table 1:

[0038] In Table 1 above, DeepSleepPct represents the proportion of deep sleep, or the percentage of deep sleep with a weight parameter of 0.35, indicating the percentage of deep sleep in total sleep (unit: percentage %). Deep sleep is a crucial stage for restoring physical strength, enhancing immunity, and regulating hormones. Therefore, in this embodiment of the invention, restorative indicators related to deep sleep are set. In calculation, the percentage of deep sleep in total sleep can be used as a reference. For example, if deep sleep accounts for 25%, the score = 0.35 × 25 = 8.75. This is the most crucial scoring factor, directly reflecting the quality of restorative sleep.

[0039] In Table 1 above, SleepEfficiency represents sleep efficiency, with a weighting parameter of 0.20. Typically, sleep efficiency is calculated as: Total sleep time ÷ Time spent in bed × 100%. For a normal person, it should be >85%, with higher values ​​indicating faster sleep onset and fewer awakenings. For example, if someone lies in bed for 8 hours and sleeps for 7 hours, their sleep efficiency is: Sleep Efficiency = 87.5, and their score = 0.20 × 87.5 = 17.5, reflecting their ability to fall asleep and maintain sleep.

[0040] In Table 1 above, SnoreIndex represents the snoring index with a weighting parameter of 0.15. Assuming the SnoreIndex value ranges from 0 to 10, a higher value indicates severe snoring, implying potential breathing difficulties. Therefore, in this embodiment of the invention, the SnoreIndex is used as a dimension of slightly unobstructed breathing, and a "deduction-based" mapping is performed using the formula 100 - SnoreIndex × 10. That is, the higher the score, the more points are deducted. For example, if SnoreIndex = 4, the corresponding sleep index is calculated as 100 - 4 × 10 = 60; this is divided equally into 0.15 × 60 = 9 points for negative penalty, reflecting breathing patency and sleep stability.

[0041] In Table 1 above, REMCyclesNormalized indicates the normalization of REM cycles. Its weighting parameter is typically set to 0.15, referring to the number of REM (Rapid Eye Movement) sleep cycles occurring in one night, generally 4-6 times. Normal values ​​are mapped to 0-100; for example, 4 times corresponds to 80, and 5 times corresponds to 100. The REM index refers to the dreaming period, which is related to memory consolidation and emotion regulation, supplementing the cognitive and mental recovery dimensions.

[0042] In Table 1 above, HRVNormalized represents heart rate variability standardization, with a weighting parameter of 0.15. HRV (Heart Rate Variability) reflects the activity level of the autonomic nervous system; a higher HRV indicates better rest and stronger stress resistance. This indicator can also be mapped from 0 to 100 (e.g., SDNN=50ms corresponds to a score of 60). This indicator serves as a supplement to overall health and sleep quality indicators.

[0043] In practical use, the dimensions and sleep indicators shown in Table 1 above, as well as the corresponding weight parameters, can be set according to the actual usage. This embodiment of the invention does not impose any restrictions on this. Furthermore, based on each of the above sleep indicators and the corresponding weight parameters, the sleep score can be calculated according to the following linear model: Score=0.35×DeepSleepPct+0.20×SleepEfficiency+0.15×(100-SnoreIndex×10)+0.15×REMCcyclesNormalized+0.15×HRVNormalized; In the actual calculation process, when determining each sleep index, normalization and other processing procedures are required. For example, the snoring index is mapped to the 0-10 range, and the deep sleep ratio is converted into the z-score standard formula for calculation. The specific processing procedures can be set according to the actual use situation, and the embodiments of the present invention do not limit them.

[0044] Furthermore, the aforementioned sleep scores need to be graded. Therefore, different score ranges need to be pre-configured to determine the final sleep level based on the specific score range. For example, 90-100 corresponds to an excellent sleep level; 70-89 corresponds to a good sleep level; 50-69 corresponds to a moderate sleep level (requiring improvement); and below 50 corresponds to a poor sleep level (high risk). Different labels can be used to correspond to the corresponding sleep level, such as A for excellent; B for good; C for moderate; and D for poor. This simplifies the representation of sleep levels. For example, when the sleep state is determined, the sleep score and corresponding sleep level can be directly output, such as sleep_score=63 and sleep_grade=C. Users can easily obtain their sleep quality and sleep state through smart terminals.

[0045] Furthermore, in addition to the aforementioned sleep scores, this embodiment of the invention can also set judgment conditions for each sleep state. For example, the data directly monitored by the intelligent sleep monitoring device and the data after certain preprocessing by the backend server can be aggregated to obtain multiple sleep data. Based on big data health assessment, corresponding sleep characteristic data can be extracted from these data, and the sleep state can be obtained according to the judgment conditions corresponding to each sleep state. Specifically, the sleep characteristic data can include the sleep indicators in Table 1 above, as well as more sleep characteristic data obtained from backend server analysis. For ease of understanding, Table 2 below shows a data table of sleep characteristic data. The first column indicates the specific data included in the sleep characteristic data, the second column shows the data type, the third column describes the sleep characteristic data, and the fourth column provides example values. As shown below: Table 2:

[0046] Based on the sleep characteristic data shown in Table 2 above, a portion of the data can typically be extracted. Different users or groups can extract different sleep characteristic data from it, such as teenagers, middle-aged and elderly people, or other groups classified in other ways. The corresponding sleep characteristic data can be selected for analysis according to actual needs.

[0047] Furthermore, when determining sleep state, pre-configured judgment conditions corresponding to each sleep state can be obtained, and then the sleep state of the target user can be determined based on the judgment conditions satisfied by the sleep feature data.

[0048] The dimensions of sleep characteristic data required for the determination criteria typically include at least the percentage of deep sleep, total sleep duration, number of awakenings, sleep efficiency, and snoring index. When determining the sleep state of a target user based on the criteria met by the sleep characteristic data, if the percentage of deep sleep is less than a preset percentage threshold, and the total sleep duration is within a preset sleep duration threshold, then the target user's sleep state is determined to be deep sleep-deficient. If the number of awakenings is greater than a preset number threshold, and the sleep efficiency is less than a preset efficiency threshold, then the target user's sleep state is determined to be frequent awakening. If the snoring index is greater than a preset index threshold, then the target user's sleep state is determined to be suspected of having breathing abnormalities. If none of the sleep characteristic data meets any of the determination criteria, then the target user's sleep state is determined to be well-recovered.

[0049] For example, based on the total sleep duration and the percentage of deep sleep, we can set the criteria for determining a deep sleep deficiency type. For instance, if the percentage of deep sleep is less than 12% and the total sleep duration is greater than 360, it can be identified as a deep sleep deficiency type. Based on the number of awakenings and the sleep efficiency in Table 1 above, we can set the criteria for determining a frequent awakening type. For instance, if the number of awakenings is greater than or equal to 5 and the sleep efficiency is less than 80, it can be identified as a frequent awakening type. Based on the snoring intensity index, we can also set the criteria for determining a suspected breathing abnormality type. For instance, if the snoring intensity index is greater than 4, it is considered a suspected breathing abnormality type, and so on.

[0050] Furthermore, when determining sleep state, one or more of the above-mentioned criteria can be used. For example, when there are multiple criteria, sleep feature data can be judged sequentially to obtain the sleep state. For ease of understanding, referring to Tables 1 and 2 above, the following is pseudocode for a process of determining sleep state: IF deep_sleep_pct<12% AND total_sleep_minutes>360 THEN sleep state = "deep sleep deficit"; ELSE IF awake_count ≥ 5 AND sleep_efficiency < 80 THEN sleep state = "Frequent awakenings"; ELSE IF snore_index > 4.0 THEN sleep state = "Suspected abnormal breathing"; ELSE sleep status = "well recovered".

[0051] In other words, multiple judgment conditions can be set according to actual usage needs to determine the corresponding sleep state. The specific settings can be configured according to actual usage, and this embodiment of the invention does not impose any limitations on this.

[0052] Furthermore, the determined sleep state can be fed back to the user, for example, by being sent to the user's smart terminal for display. In addition to the sleep state mentioned above, a sleep summary can be displayed to the user, including sleep score, sleep level, and sleep data related to the determined sleep state, as well as sleep data of interest to the user, etc. This data can be output in the form of a structure, such as: { "user_id": "user_123456", "sleep_score": 63, "score_grade": "C", "status_tag": "Deep sleep absence type", "sleep_summary": { "total_sleep_minutes": 430, "deep_sleep_pct": 11.2, "awake_count": 5, "snore_index": 4.3, "rem_cycles": 2 }, "recommend_flag": true, "timestamp": "2025-06-11T07:10:00Z" }

[0053] By displaying the sleep score, sleep level, sleep state, and related sleep data to users, they can clearly understand their current sleep status, which helps in making further dietary recommendations to users.

[0054] It should be understood that Tables 1 and 2 above only exemplarily show some sleep data. In this embodiment of the invention, in addition to Tables 1 and 2 above, other sleep data can be obtained or analyzed according to actual needs, and this embodiment of the invention does not impose any restrictions on this. Furthermore, sleep feature data or combinations of sleep feature data can be extracted from the sleep data to set corresponding judgment conditions, thereby determining the corresponding sleep state. The specific determination depends on the actual usage, and this embodiment of the invention does not impose any restrictions on this.

[0055] Furthermore, after obtaining the sleep state through the above method, it is equivalent to completing the user's health assessment. In this embodiment of the invention, the sleep state is further used as input to generate personalized recipe recommendations. In actual use, the sleep state can be represented in the form of tags, such as "deep sleep deficiency," "suspected breathing abnormality," or "good recovery," which can be used as input features. Then, a pre-established association model is obtained, and the input features containing the sleep state are input into the association model. The association model outputs recommended nutritional elements and recipe information containing the nutritional elements, thereby generating recipe recommendation information containing nutritional elements and recipe information.

[0056] Furthermore, the association model in this embodiment of the invention is a large language model; when the sleep state is input into the pre-established association model, and the association model outputs recommended nutritional elements and recipe information containing nutritional elements, natural language prompts for the association model can be generated based on the sleep state, the sleep feature data corresponding to the sleep state, and the sleep score; the natural language prompts are input into the association model, so that the association model generates recipe information containing recommended nutritional elements and recipe information containing nutritional elements based on the natural language prompts; wherein, the recipe information includes at least the recipe name, the ingredients containing nutritional elements, and the nutritional elements.

[0057] Specifically, the aforementioned association model in this embodiment of the invention is a model that associates sleep state with recipe recommendation information. Ultimately, it achieves a closed-loop linkage from sleep monitoring to health assessment and then to dietary intervention, truly transforming the user's nighttime physiological state into intelligent guidance for the next day's dietary behavior. Furthermore, different nutritional keywords and recipe keywords can be formulated for different sleep states, laying the foundation for further recipe recommendations.

[0058] For example, Table 3 shows a correspondence between sleep states and recipe recommendations, as shown below: Table 3:

[0059] Based on Table 3 above, the large language model can be used to output the corresponding recipe information and generate healthy eating suggestions. This is because the large language model can integrate information from multiple dimensions, identify complex nonlinear relationships, and flexibly classify, interpret, and recommend based on prompt information.

[0060] For example, the sleep scores, sleep levels, sleep states, and related sleep data obtained above can be organized into structured data and input as cue word information into a large language model. The structured data can be represented as follows: { "total_sleep_minutes": 410, "deep_sleep_pct": 10.5, "wake_episodes": 6, "snore_index": 5.1, "rem_cycles": 1, "sleep_score": 67, "sleep_state_tag": "Deep sleep absent type" } This structured data, after being converted into natural language prompts, can be input into a large language model. For example, the natural language prompt might be: "The user's total sleep duration last night was 410 minutes, with deep sleep accounting for only 10.5%, 6 awakenings during the night, a snoring index of 5.1, and only 1 REM cycle, indicating a 'deep sleep deficit' state. Please recommend suitable recipes for breakfast and lunch to improve deep sleep and nighttime recovery." The large language model then outputs recipe information based on the natural language prompts, such as suggesting the inclusion of nutrients beneficial for melatonin synthesis, magnesium, calcium, and vitamin B6, generating the following recipe suggestions and nutritional information: { "breakfast": { Title: Banana Walnut Oatmeal Porridge "ingredients": ["oatmeal", "banana", "walnut", "low-fat milk"], "nutrition_focus": Provides complex carbohydrates, magnesium, and B vitamins, which help stabilize the nervous system. }, "lunch": { Title: Salmon, Spinach, and Quinoa Bowl Ingredients: ["Salmon", "Spinach", "Quinoa", "Olive Oil", "Lemon Juice"] "nutrition_focus": "Rich in omega-3 fatty acids, calcium, and iron, supporting melatonin synthesis." }, "general_suggestion": "It is recommended to reduce caffeine and sugary food intake, and avoid drinking tea or energy drinks in the afternoon." } The content following "title" is the recommended recipe name, "ingredients" refers to the ingredients, and "nutrition_focus" indicates the recommended recipe name or the nutritional elements contained in the ingredients.

[0061] The above data can be used as the final recipe recommendation information for output. Furthermore, the backend server can link this recipe recommendation information with the sleep analysis results and store it in the database. Users can then view their sleep data, analysis results, and corresponding recommended recipe information through their smart devices, allowing them to plan their breakfast, lunch, and dinner accordingly.

[0062] In addition, in this embodiment of the invention, in response to a confirmation operation on the recipe information, the cooking parameters of the recipe recorded in the recipe information can be generated; the cooking parameters can be sent to the cooking device, and the cooking device can be controlled to cook according to the cooking parameters.

[0063] In practice, for the aforementioned recipe information, the backend server can search for relevant cooking parameters in the recipe database, such as temperature, power, time, and steam ratio, and then send them to the smart cooking devices already connected to the network in the user's home, such as steam ovens and smart ovens. The smart cooking devices can execute automatic cooking based on the received cooking parameters. Users only need to prepare ingredients to complete the entire cooking process, realizing a closed-loop control of intelligent diet driven by health analysis. The backend server or smart terminal can communicate with the cooking devices through standardized protocols, such as MQTT (Message Queuing Telemetry Transport) and HTTP (Hypertext Transfer Protocol), or manufacturer-specific protocols to control the cooking devices.

[0064] Furthermore, in this embodiment of the invention, the intervention effect of recipe recommendations can be continuously tracked, such as obtaining sleep tracking data corresponding to sleep data; comparing sleep data and sleep tracking data, and generating feedback information corresponding to recipe recommendation information based on the comparison results.

[0065] For example, after generating recipe recommendations based on sleep status, the system can continuously track the user's sleep data and determine their sleep state over a subsequent period. This data can be compared and analyzed with previous sleep data and states, such as recording changes in sleep efficiency, the number of awakenings during the night, and improvements in sleep scores or sleep levels. The recipe recommendations can also be adjusted based on the comparison results, or adjusted according to the user's dietary preferences. For instance, a user profile can be built based on historical sleep analysis and recommended recipes, including gender, age, basal metabolic rate, etc. The recommendations can be dynamically adjusted based on current sleep status tags, such as "deep sleep deficit," recent dietary history and preference distribution, and current recommendation results, such as breakfast / lunch recipes and ingredients. For example, specific nutritional combinations or recipe paths (such as high magnesium + high protein) can be recommended, ingredient combinations can be replaced, and cooking methods can be adjusted, such as changing from boiling to steaming, to achieve personalized dietary recommendations and thus improve the user experience.

[0066] Furthermore, based on the above embodiments, this invention also provides a recipe recommendation system, including a backend server, and cooking equipment and intelligent sleep monitoring equipment communicating with the backend server; wherein, the backend server is configured with a recipe recommendation device; specifically, as shown... Figure 3 The diagram shows a structural schematic of a recipe recommendation device, which includes: The acquisition module 30 is used to acquire sleep data, wherein the sleep data is obtained by the intelligent sleep monitoring device monitoring the sleep of the target user; The determining module 32 is used to determine the sleep state of the target user based on the sleep data; Recommendation module 34 is used to input the sleep state into a pre-established association model, and output recommended nutritional elements and recipe information containing the nutritional elements through the association model; wherein, the association model is a model that associates the sleep state with the recipe recommendation information; The generation module 36 is used to generate recipe recommendation information that includes the nutritional elements and the recipe information.

[0067] The apparatus and system provided in this embodiment of the invention have the same technical features as the recipe recommendation method provided in the above embodiments, so they can also solve the same technical problems and achieve the same technical effects.

[0068] Furthermore, embodiments of the present invention also provide 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 computer program to implement the steps of the above method.

[0069] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the above-described method.

[0070] Furthermore, embodiments of the present invention also provide a schematic diagram of the structure of an electronic device, such as... Figure 4 The diagram shows the structure of the electronic device, which includes a processor 41 and a memory 40. The memory 40 stores computer-executable instructions that can be executed by the processor 41, and the processor 41 executes the computer-executable instructions to implement the above-described method.

[0071] exist Figure 4 In the illustrated embodiment, the electronic device further includes a bus 42 and a communication interface 43, wherein the processor 41, the communication interface 43, and the memory 40 are connected via the bus 42.

[0072] The memory 40 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 43 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 42 may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 42 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0073] Processor 41 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 41 or by software instructions. Processor 41 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in the memory, and the processor 41 reads the information in the memory and uses its hardware to complete the aforementioned method.

[0074] The computer program product of the recipe recommendation method and system provided in this embodiment of the invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0075] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0076] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0077] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, 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 this 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.

[0078] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0079] Finally, it should be noted that the above embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, 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, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for recommending recipes, characterized in that, The method includes: Acquire sleep data, wherein the sleep data is obtained by a smart sleep monitoring device monitoring the sleep of a target user; The sleep status of the target user is determined based on the sleep data; The sleep state is input into a pre-established association model, which outputs recommended nutritional elements and recipe information containing the nutritional elements; wherein, the association model is a model that associates the sleep state with the recipe recommendation information. Generate recipe recommendation information that includes the nutritional elements and the recipe information.

2. The method according to claim 1, characterized in that, The step of determining the sleep state of the target user based on the sleep data includes: Extract sleep feature data from the sleep data; The sleep state of the target user is determined based on the sleep characteristic data.

3. The method according to claim 2, characterized in that, The step of determining the sleep state of the target user based on the sleep feature data includes: Select at least one dimension of sleep index from the sleep feature data, and calculate a sleep score based on at least one of the sleep indexes; The sleep level is determined based on the score range to which the sleep score belongs; Generate a sleep state that includes the sleep level.

4. The method according to claim 3, characterized in that, The step of calculating a sleep score based on at least one of the said sleep indicators includes: Obtain the pre-configured weight parameters for each of the sleep metrics; The sleep score is obtained by weighting each sleep index based on the weight parameters.

5. The method according to claim 3, characterized in that, The step of determining the sleep state of the target user based on the sleep feature data further includes: Obtain the pre-configured criteria for each sleep state; The sleep state of the target user is determined based on the judgment conditions satisfied by the sleep feature data.

6. The method according to claim 5, characterized in that, The dimensions of the sleep characteristic data required for the determination criteria include at least the percentage of deep sleep, the total sleep duration of the current sleep, the number of awakenings, sleep efficiency, and snoring index. The step of determining the sleep state of the target user based on the determination criteria satisfied by the sleep feature data includes: If the percentage of deep sleep data is less than a preset percentage threshold, and the total sleep duration data is at a preset sleep duration threshold, then the sleep state of the target user is determined to be deep sleep-deficient. If the number of awakenings is greater than a preset threshold, and the sleep efficiency is less than a preset efficiency threshold, then the sleep state of the target user is determined to be frequent awakening type. If the snoring index is greater than a preset index threshold, the sleep state of the target user is determined to be suspicious of abnormal breathing. If none of the sleep feature data meets any of the determination conditions, then the sleep state of the target user is determined to be well-recovered.

7. The method according to claim 6, characterized in that, The association model is a large language model; The steps of inputting the sleep state into a pre-established association model, outputting recommended nutritional elements through the association model, and providing recipe information containing the nutritional elements include: The association model generates natural language prompts based on the sleep state, the sleep feature data corresponding to the sleep state, and the sleep score. The natural language prompt information is input into the association model, so that the association model generates, based on the natural language prompt information, the recommended nutritional elements and the recipe information containing the nutritional elements; The recipe information includes at least the recipe name, the ingredients containing the nutrients, and the nutrients themselves.

8. The method according to claim 1, characterized in that, The method further includes: In response to a confirmation operation on the recipe information, the cooking parameters of the recipe recorded in the recipe information are generated; The cooking parameters are sent to the cooking device, and the cooking device is controlled to cook according to the cooking parameters.

9. The method according to claim 1, characterized in that, The method further includes: Obtain the sleep tracking data corresponding to the sleep data; By comparing the sleep data and the sleep tracking data, feedback information corresponding to the recipe recommendation information is generated based on the comparison results.

10. A recipe recommendation system, characterized in that, It includes a back-end server, as well as cooking equipment and smart sleep monitoring equipment that communicate with the back-end server; The backend server is configured with a recipe recommendation device; The recipe recommendation device includes: The acquisition module is used to acquire sleep data, wherein the sleep data is obtained by the intelligent sleep monitoring device monitoring the sleep of the target user; The determination module is used to determine the sleep state of the target user based on the sleep data; The recommendation module is used to input the sleep state into a pre-established association model, and output recommended nutritional elements and recipe information containing the nutritional elements through the association model; wherein, the association model is a model that associates the sleep state with the recipe recommendation information; The generation module is used to generate recipe recommendation information that includes the nutritional elements and the recipe information.

Images