Recommendation device, recommendation method, recommendation program, and recording medium for recommending diet

Abstract

A recommendation device 6 for recommending a healthy diet to a user is provided. A recommendation device (6) is provided with: a biological information acquisition unit (61) that acquires biological information of a user; a health index acquisition unit 62 that acquires a health index of the user; a food component evaluation unit 63 that evaluates food components on the basis of the biological information; a diet recommendation means (64) for recommending a healthy diet on the basis of the evaluation results of the food components; and a history generation unit 67 that generates a history of the correspondence relationship between the diet recommended by the diet recommendation unit 64 and the health index.

Description

Technical Field

[0001] The present invention relates to a recommendation device, a recommendation method and a recommendation program for recommending meals beneficial to a user's health. Background Art

[0002] When blood sugar levels rise, the risk of diabetes increases along with the risk of many other diseases. This is because glucose in the blood penetrates the endothelial cells lining the blood vessels, where the reactive oxygen species produced damage the blood vessels. Furthermore, Non-Patent Document 1 discusses the relationship between diet and blood sugar levels, emphasizing that maintaining a normal postprandial blood sugar response is crucial for preventing various metabolic disorders, including diabetes. Non-Patent Document 1 also mentions that individual differences, such as differences in the gut microbiome, can influence postprandial blood sugar response.

[0003] In contrast, the gut intelligence service introduced in Non-Patent Document 1 recommends diets that are beneficial or harmful to health by examining the user's intestinal microbiota. It also recommends meals that have an adverse effect on blood sugar levels or those that do not have such a negative effect.

[0004] Related technical literature

[0005] Non-patent literature

[0006] Non-patent document 1: “Factors Affecting Postprandial Blood Glucose Changes Include Gut Microbiota”, Mykinso Laboratory, Internet<URL:https: / / lab.mykinso.com / kenkyu / 190409>

[0007] Non-Patent Document 2: “Gut Intelligent Service”, Viome Inc., United States, Internet<URL:https: / / www.viome.com> Summary of the Invention

[0008] Technical issues

[0009] However, the service described in Non-Patent Document 2 recommends meals based on academic knowledge, and therefore does not necessarily continuously provide meal recommendations suitable for individual users.

[0010] The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to continuously provide a meal suitable for the user.

[0011] Solution to the problem

[0012] In order to solve the above-mentioned problems, the present invention includes the following embodiments.

[0013] A recommendation device for recommending meals that are beneficial to a user's health, comprising: a biological information acquisition unit configured to acquire biological information of the user; a health index acquisition unit configured to acquire the user's health index; a food ingredient evaluation unit configured to evaluate food ingredients based on the biological information; a meal recommendation unit configured to recommend meals that are beneficial to health based on the evaluation results of the food ingredients; and a history generation unit configured to generate a history of the correspondence between meals recommended by the meal recommendation unit and the health index.

[0014] The recommendation device according to Item 1, wherein the meal recommendation unit recommends a meal that is beneficial to health based on history.

[0015] The recommendation device according to Item 1 or Item 2, wherein the meal recommendation unit recommends a meal that is beneficial to health by comprehensively evaluating the evaluation results and history of food components.

[0016] The recommendation device according to any one of Articles 1 to 3 further includes a similar history search unit, which searches for similar histories of other users whose biological information is similar to that of the user, and the similar histories are records of the correspondence between the diet recommended by the diet recommendation unit and the health index, wherein the diet recommendation unit further recommends diets that are beneficial to health based on the similar histories.

[0017] The recommendation device according to any one of clauses 1 to 4, wherein the biological information is intestinal microbiome information.

[0018] The recommendation device according to any one of clauses 1 to 5, wherein the health index includes blood sugar level.

[0019] A method for recommending meals that are beneficial to a user's health, comprising: a biological information acquisition step for acquiring the user's biological information; a health index acquisition step for acquiring the user's health index; a food ingredient evaluation step for evaluating food ingredients based on the biological information; a meal recommendation step for recommending meals that are beneficial to health based on the evaluation results of the food ingredients; and a history generation step for generating a history of the correspondence between the meals recommended in the meal recommendation step and the health index.

[0020] A recommendation program for recommending meals that are beneficial to a user's health, the program causing a computer to operate as follows: a biological information acquisition unit configured to acquire the user's biological information; a health index acquisition unit configured to acquire the user's health index; a food ingredient evaluation unit configured to evaluate food ingredients based on the biological information; a meal recommendation unit configured to recommend meals that are beneficial to health based on the evaluation results of the food ingredients; and a history generation unit configured to generate a history of the correspondence between meals recommended by the meal recommendation unit and the health index.

[0021] A non-transitory computer-readable recording medium stores a recommendation program for recommending meals that are beneficial to a user's health, wherein the recommendation program causes a computer to operate as follows: a biological information acquisition unit configured to acquire the user's biological information; a health index acquisition unit configured to acquire the user's health index; a food ingredient evaluation unit configured to evaluate food ingredients based on the biological information; a meal recommendation unit configured to recommend meals that are beneficial to health based on the evaluation results of the food ingredients; and a history generation unit configured to generate a history of the correspondence between meals recommended by the meal recommendation unit and the health index.

[0022] Effects of the present invention

[0023] According to the present invention, a meal suitable for the user can be continuously provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a block diagram showing a schematic configuration of a recommendation system according to Embodiment 1 of the present invention.

[0025] Figure 2 is a flowchart showing the overall processing procedure performed by the recommendation system.

[0026] Figure 3 is a flowchart detailing the steps performed by the recommendation server.

[0027] Figure 4 is an example of the evaluation results of food ingredients based on their effects on blood sugar levels.

[0028] Figure 5 is an example of data recorded in a meal / recipe database.

[0029] Figure 6 is a graph showing the effect of evaluation results of each food ingredient included in a recipe based on a set of food and drink on the blood sugar level.

[0030] Figure 7 is an example of a meal (a set of food and drink) recommended by the meal recommendation unit.

[0031] Figure 8 This is an example of the history of the correspondence between the diets recommended by the diet recommendation unit and the health index.

[0032] Figure 9 is a block diagram showing a schematic configuration of a recommendation system according to Embodiment 2 of the present invention.

[0033] Figure 10 (a) to (c) are graphs showing the intestinal microbiota scores of user A, another user P, and another user Q, respectively.

[0034] Figure 11 (a) and (b) show examples of histories of user A and user P, respectively.

[0035] Figure 12 (a) shows an example of a set of food and drink extracted by the meal recommendation unit, and (b) shows an example of Figure 12 The differences shown in (a) are converted to positive values ​​for scoring.

[0036] Figure 13 Examples of meals (a collection of food and drink) recommended by the meal recommendation unit are presented.

[0037] Figure 14 A graph showing the recommendation rating and purchase frequency for each food and drink set over a specified past time period.

[0038] Figure 15 An example of a screen is provided in which factors to be considered for meal recommendations may be selected.

[0039] Figure 16 Another example of a meal (a set of food and drink) recommended by the meal recommendation unit is shown.

[0040] Figure 17 This is a graph showing the recommendation levels of various food and drink sets. Specific embodiments

[0041] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. It should be noted that the present invention is not limited to the embodiments described below, and various modifications can be made without departing from the scope and spirit of the present invention.

[0042] Example 1

[0043] Figure 1 1 is a block diagram showing a schematic structure of a recommendation system according to Embodiment 1 of the present invention. The recommendation system 1 is designed to recommend meals that are beneficial to a user's health, and includes an intestinal microbiota examination device 2, an intestinal microbiota analysis server 3, a health index measurement device 4, a user terminal 5, and a recommendation server (recommendation device) 6.

[0044] The intestinal microbiome testing device 2 is used to analyze intestinal microbial flora from human feces. It is managed by a service provider that provides testing services. Users collect feces in a intestinal microbiome testing kit and send it to the service provider for analysis.

[0045] Based on the detection results from the intestinal microbiome testing device 2, the intestinal microbiome analysis server 3 analyzes the types of intestinal bacteria and the proportions of each type of bacteria. The intestinal microbiome analysis results are transmitted to the recommendation server 6. Furthermore, the intestinal microbiome analysis server 3 can further analyze aspects such as the diversity score of intestinal bacteria and disease risk based on the user's biological information (including age, gender, presence of chronic diseases, health checkup results, and genetic analysis results).

[0046] The health index measuring device 4 is designed to measure the user's health index, such as blood sugar level, abdominal condition, skin condition, nail condition, hair color, and vascular age. When the health index is blood sugar level, the health index measuring device 4 can be in the form of a fixed device or a wearable device. The health index measuring device 4 can also measure blood sugar level based on blood test results or subjective questionnaires (for example, questionnaires about alternative indicators related to fluctuations in blood sugar levels (such as postprandial sleepiness, attention, etc.)). If the health index is abdominal condition, the health index measuring device 4 can include a defecation sensor, or the health index measuring device 4 can measure abdominal condition based on a questionnaire about the condition during defecation and the state of feces. When the health index is skin condition, the health index measuring device 4 can be composed of a skin sensor; alternatively, the health index measuring device 4 can evaluate the skin condition based on a subjective questionnaire about the skin, and similarly, nail condition and hair color can also be used as health indices. In addition, if the health index involves anti-aging, measuring vascular age can be considered. The measurement results of the health index are sent to the recommendation server 6.

[0047] The user terminal 5 is a device used by a user, such as a smartphone or a personal computer, and can communicate with the recommendation server 6 via a communication network such as the Internet.

[0048] The recommendation server 6 is a device that recommends meals that are beneficial to the user's health. The recommendation server 6 can be configured using a general-purpose computer. In terms of hardware configuration, the recommendation server 6 includes a processor such as a CPU or GPU (not shown), a main storage device such as DRAM and SRAM (not shown), and an auxiliary storage device 60 such as an HDD or SSD. The auxiliary storage device 60 stores programs and data required for operating the recommendation server 6, including a recommendation program P1, an information acquisition program P2, a health index database DB1, a meal / recipe database DB2, and a recommendation history database DB3.

[0049] It is also possible to use a plurality of computers to configure the recommendation server 6. In addition, the auxiliary storage device 60 may also be externally attached to the recommendation server 6.

[0050] The recommendation server 6 includes the following functional blocks: a biological information acquisition unit 61, a health index acquisition unit 62, a food composition evaluation unit 63, a meal recommendation unit 64, a delivery instruction unit 65, a consumption record unit 66, and a history generation unit 67. These functional blocks can be implemented in hardware, for example, using logic circuits, or they can be implemented in software by the processor of the recommendation server 6. In the case of software implementation, the processor can implement each of these units by loading the recommendation program P1 and the information acquisition program P2 stored in the auxiliary storage device 60 into the main storage device and executing the recommendation program P1 and the information acquisition program P2. The recommendation program P1 and the information acquisition program P2 can be downloaded to the recommendation server 6 via a communication network such as the Internet, or they can be installed on the recommendation server 6 via a computer-readable non-transitory storage medium such as a CD-ROM, which records the recommendation program P1 and the information acquisition program P2. It should be noted that the recommendation program P1 can also include part or all of the information acquisition program P2.

[0051] The processing process of the entire system

[0052] Figure 2 is a flowchart showing the overall processing procedure performed by the recommendation system 1 .

[0053] In step S1, the intestinal microbiota examination device 2 examines the intestinal microbiota of the user. Based on the examination results, the intestinal microbiota analysis server 3 analyzes intestinal microbiota information such as the types of intestinal bacteria and the proportion of individual bacteria.

[0054] In step S2 , the recommendation server 6 recommends weekly meals that are beneficial to the user's health based on the intestinal microbiome information.

[0055] In step S3, the health index measuring device 4 measures the user's health index. In this embodiment, the health index is the blood sugar level.

[0056] Step S1 is repeated every three months (Yes in step S4), step S2 is repeated once a week (Yes in step S5), and step S3 is repeated once a day (Yes in step S6). It should be noted that the cycles of these repetitions are not particularly limited.

[0057] Recommendation server processing

[0058] Next, the functions of the components of the recommendation server 6 will be described. Figure 3 is a flowchart showing the detailed procedure of step S2 performed by the recommendation server 6 .

[0059] In step S21 (biological information acquisition step), the biological information acquisition unit 61 acquires the user's biological information. In this embodiment, the biological information acquisition unit 61 acquires the user's intestinal microbiome information from the intestinal microbiome analysis server 3. Step S21 is performed after the user's intestinal microbiome is examined (step S1).

[0060] In step S22 (health index acquisition step), the health index acquisition unit 62 acquires the measurement result of the user's health index from the health index measurement device 4. Step S22 is performed every day, and the health index acquisition unit 62 records the acquired health index in the health index database DB1.

[0061] In step S23 (food composition evaluation step), the food composition evaluation unit 63 evaluates the food based on the intestinal microbiome information. Food based on the intestinal microbiome information refers to food that affects the user's blood sugar level and / or intestinal environment. In this embodiment, "food" mainly refers to items such as vegetables, meat, fish, and fruit, but may also include beverages, condiments, and supplements.

[0062] Figure 4 An example of a food ingredient evaluation based on its effect on blood sugar levels is shown. In this example, food ingredients with positive blood sugar impact scores are expected to improve blood sugar levels, while food ingredients with negative blood sugar impact scores indicate a potential adverse effect on blood sugar levels. In this example, the values ​​indicating the blood sugar impact scores are represented by two types: +1 and -1. However, other values, such as +2 and -3, may also be used.

[0063] As an example of evaluating food ingredients based on health indices other than blood sugar levels, the effect on the gut microbiome can be considered. For example, a food ingredient that has a beneficial effect on the gut microbiome can be rated as +2 or +1, while a food ingredient that has an adverse effect can be rated as -2 or -1. Another example involves obtaining an assessment of a food ingredient based on predefined rules by reference to other information (such as the gut microbiome score or the presence of specific bacteria). For example, a rule can be pre-set so that if a certain gut microbiome score exceeds a predetermined threshold, a specific food ingredient is selected.

[0064] In step S24 (meal recommendation step), the meal recommendation unit 64 refers to the meal / recipe database 2 and recommends a meal that promotes the user's health based on the evaluation results of the food components evaluated by the food component evaluation unit 63 .

[0065] Figure 5 is an example of data recorded in the meal / recipe database 2. A recipe refers to information such as food ingredients required for a dish, their quantities, and cooking methods. Figure 5 , recipes 1 to 3 of a single food and drink set are shown. The term "food and drink set" refers to a set of food items consumed in one meal and is synonymous with a meal.

[0066] Figure 6 The figure shows the influence on the blood sugar level based on the food component evaluation result for each food component included in the meal set recipe obtained from the food component evaluation unit 63. The meal recommendation unit 64 calculates the total influence of each food component on the blood sugar level and determines the meal set score as (-1+1+0+0+0+0+0+0+0+0=0).

[0067] It should be noted that the food components that have not been evaluated by the food component evaluation unit 63 may also be treated as zero. Figure 6 The effect on blood sugar levels depicted in does not take into account the amount of food ingredients. However, if the food ingredient evaluation result includes information about the standard amount, the amount of each food ingredient can be taken into account when calculating the meal set score. In addition, the food ingredient evaluation unit 63 can calculate the score at a finer level (for example, in terms of individual nutrients). In this case, an artificial intelligence model of machine learning can be used for the score calculation. For example, if the health index is not blood sugar level, the effect on the intestinal microbiome can be used for food ingredient evaluation, and the total impact value can be calculated accordingly.

[0068] The meal recommendation unit 64 recommends a set of food and drink with a high set score from the food and drink sets recorded in the food / recipe database 2 as a meal that is beneficial to the user's health. Furthermore, the meal recommendation unit 64 records the meal recommended to the user in the recommendation history database 3 (step S25). The recommendation content from the meal recommendation unit 64 is transmitted to the user terminal 5.

[0069] Figure 7 An example of a meal (a set of food and drink) recommended by the meal recommendation unit 64 is shown. Figure 7 In the example, three meals are shown in order of highest score, but the number and order of recommended meals are not particularly limited. The user can specify the number of desired meals and press the order confirmation button B1 (Yes in step S28), whereby the delivery instruction unit 65 instructs the delivery of the meals, and the consumption recording unit 66 records the meals consumed by the user along with the date and time (step S29).

[0070] exist Figure 7 In the example of FIG. 5 , when the history display button B2 is pressed (Yes in step S26), the history generation unit 67 generates a history of the correspondence between the diets recommended by the diet recommendation unit 64 and the health index (step S27, history generation step). The generated history is recorded in the recommendation history DB3 of the recommendation server 6 and is also transmitted to the user terminal 5.

[0071] Figure 8 An example of the history of the correspondence between the meals (a set of food and drink) recommended by the meal recommendation unit 64 and the health index is shown. In this example, the health index is the blood sugar level, and the correspondence between the recommended meals and the blood sugar levels on the days when these meals were consumed is displayed as a one-week history. The user can confirm that the blood sugar level did not increase after consuming food and drink sets 1 and 3.

[0072] summary

[0073] As described above, in this embodiment, the historical correspondence between recommended meals and health indices is visualized, allowing users to verify whether the recommended meals are indeed suitable for them. This enables users to select meals while taking their dietary history into consideration. Therefore, meals suitable for users can be continuously provided.

[0074] Variations of the embodiment

[0075] In this embodiment, the health index is the blood sugar level, but the present invention is not limited to this. The health index may also include conditions such as gastric health or appearance factors (skin condition), nail condition, hair color, blood vessel age, etc. In addition, in this embodiment, the biological information acquisition unit 61 acquires the user's intestinal microbiome information as biological information. However, the present invention is not limited to this. The biological information acquisition unit 61 can acquire information such as the user's age, gender, presence or absence of chronic diseases, health check results, gene analysis results, etc. as biological information, and the food ingredient evaluation unit 63 can evaluate food ingredients based on the biological information.

[0076] In addition, when there are multiple health indices, in step S23, the food composition evaluation unit 63 evaluates the meal set based on each health index. In step S24, the meal recommendation unit 64 provides a meal recommendation by comprehensively evaluating the meal set that has been evaluated for each health index. For example, if the health index includes blood sugar level and intestinal condition (intestinal microbiome), the food composition evaluation unit 63 evaluates the meal set from the perspective of blood sugar level to calculate score 1 for each meal set, and evaluates the meal set from the perspective of intestinal microbiome to calculate score 2 for each meal set. The meal recommendation unit 64 normalizes score 1 and score 2 to a level between 0 and 1, and then sums these normalized values ​​for each meal set to perform a comprehensive evaluation of the meal set.

[0077] In addition, nutritional information of each food component is stored in the meal / recipe database 2. When the meal recommendation unit 64 recommends a meal, it may consider nutritional information of each set of food and drink calculated from the meal / recipe database 2, and the difference from the required standard amount.

[0078] Example 2

[0079] Figure 9 : is a block diagram showing a schematic configuration of a recommendation system 1' according to a second embodiment of the present invention. The recommendation system 1' includes an intestinal microbiome inspection device 2, an intestinal microbiome analysis server 3, a health index measurement device 4, a user terminal 5, and a recommendation server (recommendation device) 6'. In other words, by Figure 1 The recommendation system 1 shown is configured by replacing the recommendation server 6 with a recommendation server 6'. In this embodiment, components having the same functions as those in the first embodiment are denoted by the same reference numerals, and their descriptions are omitted.

[0080] Recommendation server 6' has Figure 1The hardware configuration of the recommendation server 6 shown in FIG. 1 is similar to the hardware configuration of the recommendation server 6 shown in FIG. As functional blocks, the recommendation server 6' includes a biological information acquisition unit 61, a health index acquisition unit 62, a food ingredient evaluation unit 63, a meal recommendation unit 64, a delivery instruction unit 65, a consumption record unit 66, a history generation unit 67, and a similar history search unit 68. In other words, the recommendation server 6' is configured as a recommendation server 6 to which the similar history search unit 68 is added. In addition, the health index database DB1 and the recommendation history database DB3 stored in the auxiliary storage device 60 of the recommendation server 6' include not only the health index and recommendation history of the user of the user terminal 5 (hereinafter referred to as user A), but also the health index and recommendation history of users who utilize the service other than user A (hereinafter referred to as other users).

[0081] The similarity history search unit 68 searches for a historical correspondence (similar history) between the meals recommended by the meal recommendation unit 64 and the health index of other users having the same intestinal microbiota as that of user A. Specifically, the similarity history search unit 68 first searches for other users having intestinal microbiota similar to that of user A by referring to the health index database DB1. The similarity of the intestinal microbiota is determined by using the intestinal microbiota score.

[0082] Figure 10 (a) to Figure 10 (c) shows the gut microbiome scores of user A, another user P, and another user Q, respectively. To compare the gut microbiome scores, the scores are considered as vectors and the distances between these vectors are calculated. The distance between the gut microbiome scores of user A and user P is calculated as follows: (0.9-0.8) 2 +(0.02-0.03) 2 +(0.15-0.15) 2 =0.0101. Meanwhile, the distance between the gut microbiome score of user A and the gut microbiome score of another user Q is calculated as follows: (0.9-0.4) 2 +(0.02-0.05) 2 +(0.15-0.3) 2 =0.2734. These distances are then compared to a predetermined threshold (e.g., 0.05). If the distance is less than or equal to the threshold, the gut microbiota of the two users are determined to be similar. In this example, the gut microbiota of user P is determined to be similar to the gut microbiota of user A. Although only one user with a similar gut microbiota is selected in this example, the number of such users can be any number, including zero or more.

[0083] The similarity of gut microbiota can be determined not only by the distance between the vectors of gut microbiota scores as described above, but also by using other metrics such as Mahalanobis distance or correlation coefficient.

[0084] Subsequently, the similarity history search unit 68 refers to the recommendation history database 3 to search for a history of correspondence between the recommended meals of the user P and the health index, that is, the similarity history search unit 68 searches for similarity histories.

[0085] Figure 11 (a) shows an example of user A's history. User A has been recommended meal sets 1 to 3, and their average blood glucose level during these periods is 130 mg / dL. Blood glucose levels are preferably measured a certain time after a meal (e.g., 60 minutes after a meal). Alternatively, HbA1c (hemoglobin A1c) or a sleepiness score can be used instead of blood glucose levels.

[0086] Figure 11 (b) shows an example of the history of user P. Meal sets 4-6 have been recommended to user P, and the average blood sugar level during this period is 135 mg / dL. The history of user P is similar to that of user A.

[0087] The meal recommendation unit 64 recommends a meal that is beneficial to the health of user A by considering similar historical data such as the history of user P in addition to the history of user A himself. Figure 11 As shown in (a) and (b) of FIG. 1 , the meal recommendation unit 64 extracts meals whose difference from the average value is less than or equal to zero from sets 1 to 6. The difference is then converted to a positive value and scored accordingly.

[0088] Figure 12 (a) shows an example of food and drink sets extracted by the meal recommendation unit 64. The difference values ​​in food and drink sets 1 and 2 represent the difference from the average blood glucose level (130 mg / dL) of user A, while the difference value in food and drink set 4 corresponds to the difference from the average blood glucose level (135 mg / dL) of user P. It should be noted that for food and drink sets 1 to 6, the difference values ​​from a fixed value (e.g., 140 mg / dL) can also be calculated.

[0089] Figure 12 (b) is to Figure 12 The value shown in (a) is converted into a positive value and scored. The meal recommendation unit 64 is from Figure 12 Among the meal sets shown in (b), meal sets having a score greater than a predetermined value (eg, 1) are selected as meals recommended to user A. In this example, meal sets 2 and 4 are selected as meals recommended to user A.

[0090] Figure 13 1 shows an example of meals (a set of food and drink) recommended by the meal recommendation unit 64. In this screen, when the recommendation level display button B3 is pressed, the recommendation level of each set of food and drink and the purchase frequency during a predetermined past time period are displayed in a graph, as shown in FIG. Figure 14 shown.

[0091] Brief Summary

[0092] In this embodiment, as described above, meals that are beneficial to user A's health are recommended based not only on user A's personal history but also on a comparable history from user P, whose intestinal microbiota is similar to that of user A. If meals were recommended based solely on user A's history, only those meals that had been previously recommended to user A would be recommended. However, by also considering similar histories, meals that had not been previously recommended can be recommended to user A.

[0093] exist Figure 12 In (b), the score of the set of food and drink included in user P's history can be adjusted based on the gut microbiota similarity between user A and user P. For example, similarity can be calculated by subtracting the distance between user A's gut microbiota score and user P's gut microbiota score (0.0101) from 1, resulting in a value of 0.9899. This similarity value can then be multiplied by the score of the set of food and drink included in user P's history.

[0094] Variations of the embodiment

[0095] In this embodiment, the meal recommendation unit 64 recommends meals based on history. However, the meal recommendation unit 64 can further recommend meals by comprehensively evaluating the food composition evaluation results and history. For example, when the health index is blood sugar level and abdominal condition (intestinal microbiome), the food composition evaluation unit 63 evaluates the set of food and drink for each health index. Then, the meal recommendation unit 64 can also comprehensively evaluate the food and drink set evaluated for each health index with the food and drink set based on the historical evaluation, thereby recommending a meal that is beneficial to health.

[0096] In addition, the user can select which health index to consider when displaying the meal recommendation screen. Figure 15An example of a screen for selecting factors to be considered in meal recommendations is shown, in which the user can select the factors to be considered from the available options. For example, if the user selects the factors of blood sugar level, intestinal microbiome, and history, and then presses the recommendation display button B4, the meal recommendation unit 64 evaluates the set of foods and drinks that are effective for each of blood sugar level and intestinal microbiome. The meal recommendation unit 64 calculates a score for each set of food and drink, and also calculates a score for the set of food and drink based on the user's history. The meal recommendation unit 64 combines these scores to make an overall evaluation of the set of food and drink, and recommends the meal with the highest score. Therefore, as Figure 16 , a screen showing meal recommendations that take into account the selected factors is shown. In this way, the meal recommendation unit 64 can automatically recommend meals suitable for the user. Furthermore, the meal recommendation unit 64 can also consider the user's dietary preferences (likes and dislikes) and food allergies in the overall evaluation of a set of foods and drinks.

[0097] In addition, Figure 16 In the screen shown, when the detailed recommendation level button B5 is pressed, Figure 17 As shown, a screen showing details of the recommendation level is shown. Figure 17 In the example, the recommendation level of each food and drink set based on each factor is shown in the graph.

[0098] In addition, the meal recommendation unit 64 can use an artificial intelligence (AI) model to recommend meals based on the user's history. In this scenario, as a preparation, an AI model is constructed using machine learning. In the score calculation stage, explanatory variables are input into the AI ​​model, resulting in the output of values ​​related to blood glucose levels, which are used as objective variables. Examples of explanatory variables include intestinal microbiome scores, data on a collection of food and drinks (or data broken down into nutrients), blood glucose level information and other health data. Examples of objective variables include values ​​related to blood glucose levels (such as postprandial blood glucose levels measured 30 minutes after a meal) and scores provided as a result of intestinal microbiome analysis. Machine learning techniques such as neural networks, support vector machines (SVMs) and ensemble learning can be used.

[0099] Additionally, it is contemplated that recommendations are made based on predefined rules. An example of a rule-based recommendation method involves predefined rules where a specific meal is recommended if a certain index value, or an index value calculated from a plurality of values, falls within a specific range.

[0100] In addition, the auxiliary storage device 60 is equipped with a behavior database (DB) that records actions and / or lifestyle habits that are recommended (or should be recommended) to the user. By referring to this behavior DB, the meal recommendation unit 64 can determine the degree of improvement in blood sugar levels associated with a previously recommended set of food and drink and a combination of actions and / or lifestyle habits. Based on this determination, the unit can recommend a meal.

[0101] Additional Notes

[0102] The present invention is not limited to the above-described embodiments, but can be modified in various ways within the scope of the claims, and forms obtained by appropriately combining the technical means disclosed in the embodiments are also included in the technical scope of the present invention.

[0103] Description of Reference Signs

[0104] 1 Recommendation System

[0105] 1'Recommendation System

[0106] 2Intestinal microbiome testing device

[0107] 3. Gut Microbiome Analysis Server

[0108] 4Health index measurement equipment

[0109] 5 User Terminal

[0110] 6 Recommended Servers (Recommended Devices)

[0111] 6'Recommended Server (Recommended Device)

[0112] 60 auxiliary storage devices

[0113] 61 Biological Information Acquisition Unit

[0114] 62 Health Index Acquisition Unit

[0115] 63 Food Composition Evaluation Unit

[0116] 64 dietary recommendation units

[0117] 65 delivery instruction unit

[0118] 66 Consumption Record Unit

[0119] 67 History Generation Unit

[0120] 68 Similar history search units

[0121] P1 referral program.

Claims

1. A recommendation device for recommending a meal that is beneficial to a user's health, comprising: a biometric information acquisition unit configured to acquire biometric information of a user; a health index obtaining unit, configured to obtain the health index of the user; a food composition evaluation unit configured to evaluate food composition based on the biological information; a meal recommendation unit configured to recommend a meal that is beneficial to health based on the evaluation result of the food components; as well as The history generating unit is configured to generate a history of the correspondence between the meals recommended by the meal recommendation unit and the health index.

2. The recommendation device according to claim 1, wherein: The meal recommendation unit recommends a meal that is beneficial to health based on the history.

3. The recommendation device according to claim 2, wherein: The meal recommendation unit recommends a meal that is beneficial to health by comprehensively evaluating the history and the evaluation results of the food components.

4. The recommendation device according to claim 2, further comprising a similar history search unit, wherein the similar history search unit searches for similar histories of other users whose biological information is similar to the biological information of the user, wherein the similar history is a record of the correspondence between the diet recommended by the diet recommendation unit and the health index. in, The meal recommendation unit further recommends the meal that is beneficial to health based on the similar history.

5. The recommendation device according to any one of claims 1 to 4, wherein: The organism information is intestinal microbial flora information.

6. The recommendation device according to any one of claims 1 to 4, wherein: The health index includes blood sugar level.

7. A method for recommending a meal that is beneficial to a user's health, comprising: A biometric information acquisition step for acquiring the user's biometric information; A health index obtaining step, for obtaining the health index of the user; a food composition evaluation step for evaluating food composition based on the biological information; A meal recommendation step for recommending a meal that is beneficial to health based on the evaluation results of the food components; as well as The history generating step is used to generate a history of the correspondence between the diet recommended in the diet recommending step and the health index.

8. A recommendation program for recommending a meal that is beneficial to a user's health, the program causing a computer to operate as follows: a biometric information acquisition unit configured to acquire biometric information of a user; a health index obtaining unit, configured to obtain the health index of the user; a food composition evaluation unit configured to evaluate food composition based on the biological information; a meal recommendation unit configured to recommend a meal that is beneficial to health based on the evaluation result of the food components; as well as The history generating unit is configured to generate a history of the correspondence between the meals recommended by the meal recommendation unit and the health index.

9. A non-transitory computer-readable recording medium storing a recommendation program for recommending a meal beneficial to a user's health, wherein: The recommended program causes the computer to operate as follows: a biometric information acquisition unit configured to acquire biometric information of a user; a health index obtaining unit, configured to obtain the health index of the user; a food composition evaluation unit configured to evaluate food composition based on the biological information; a meal recommendation unit configured to recommend a meal that is beneficial to health based on the evaluation result of the food components; as well as The history generating unit is configured to generate a history of the correspondence between the meals recommended by the meal recommendation unit and the health index.

Images