Recommended amount output system, recommended amount output method, and program

Abstract

The present invention improves the amount of equol intake of each individual user. This recommended amount output system comprises one or more computers that execute: calculating the amount of equol deficiency of a user by using a predicted value of the amount of equol production of the user and a predetermined target amount; calculating a recommended amount of equol or isoflavone to be ingested by the user on the basis of the calculated amount of equol deficiency and an index value indicating the level of equol production ability of the user; and outputting the calculated recommended amount.

Description

Recommended amount output system, recommended amount output method and program

[0001] The present disclosure relates to a recommended amount output system, a recommended amount output method, and a program.

[0002] A system has been proposed in the past that suggests a list of foods that can supplement a user's nutrient deficiency (see, for example, Patent Document 1). In this system, a mobile device transmits the amount of excess or deficiency of a nutrient the user has ingested to a product suggestion server. The product suggestion server, based on the received nutrient deficiency or excess information, refers to a product database that stores the names of products that supplement various nutrients, and returns information about products that the user should ingest to the mobile device.

[0003] Japanese Patent Application Laid-Open No. 2007-26262

[0004] The ability to produce equol from isoflavones depends on the intestinal environment and varies greatly from person to person. However, equol and isoflavone supplements generally list a standard intake amount, making it difficult to suggest an appropriate amount for each individual user. The present disclosure aims to provide a technology for improving the equol intake of each individual user.

[0005] The present disclosure can be realized by the following aspects. (Aspect 1) A recommended amount output system including one or more computers that perform the following steps: calculate a user's equol deficiency using a predicted value of the user's equol production level and a predetermined target amount; calculate a recommended amount of equol and / or isoflavones to be consumed by the user based on the calculated equol deficiency and an index value representing the user's level of equol production ability; and output the calculated recommended amount. (Aspect 2) A recommended amount output system according to Aspect 1, in which the predicted value of equol production is predicted using the user's equol excretion amount. (Aspect 3) A recommended amount output system according to Aspect 1 or 2, in which the predicted value of equol production is predicted using vital data acquired from the user's wearable device. Note that when Aspect 2 is cited, the user's equol production level is used to create a model for predicting the relationship between vital data and equol production level. (Aspect 4) The recommended amount output system of Aspect 3, wherein the vital data includes at least one of the user's electrocardiogram, heart rate, blood oxygen level, an index representing sleep quality, body temperature, sweat rate, and sweat components. (Aspect 5) The recommended amount output system of any one of Aspects 1 to 4, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user abstained from dietary isoflavone intake for a predetermined period of time or more and ingested a predetermined amount of an isoflavone-containing supplement. (Aspect 6) The recommended amount output system of Aspect 3 or 4, wherein the predicted value of equol production amount is predicted using a prediction model that has learned the relationship between the vital data and equol excretion after the user abstained from dietary isoflavone intake for the predetermined period of time, and the vital data and equol excretion after the user abstained from dietary isoflavone intake for the predetermined period of time and ingested a predetermined amount of an isoflavone-containing supplement.(Aspect 7) The recommended amount output system of Aspect 6, wherein if the post-ingestion equol deficiency calculated after receiving information indicating that the user has ingested the recommended amount of isoflavones satisfies a predetermined condition, the prediction model or a correction coefficient for the deficiency or the recommended amount is modified based on the relationship between the target amount and the post-ingestion equol deficiency. (Aspect 8) A recommended amount output method, wherein one or more computers execute the following steps: calculate the user's equol deficiency using the predicted value of the user's equol production amount and a predetermined target amount; calculate a recommended amount of equol and / or isoflavones to be consumed by the user based on the calculated equol deficiency and an index value representing the user's level of equol production ability; and output the calculated recommended amount. (Aspect 9) The recommended amount output method of Aspect 8, wherein the index value is a value corresponding to the amount of equol excretion or blood concentration after a predetermined time has elapsed since the user abstained from dietary isoflavone intake for at least a predetermined period of time and ingested a predetermined amount of an isoflavone-containing supplement. (Aspect 10) A program causing one or more computers to perform the following steps: calculate a user's equol deficiency using a predicted value of the user's equol production amount and a predetermined target amount; calculate a recommended amount of equol and / or isoflavones to be ingested by the user based on the calculated equol deficiency and an index value representing the user's level of equol production ability; and output the calculated recommended amount. (Aspect 11) The program according to Aspect 10, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user abstained from dietary isoflavone intake for a predetermined period of time or longer and ingested a predetermined amount of an isoflavone-containing supplement.

[0006] The content of the means for solving the problem can be provided as a device such as a computer, a system including multiple devices, a method executed by one or more computers, or a program executed by one or more computers. A recording medium storing the program may also be provided.

[0007] The disclosed technology can provide a technique for improving an individual user's equol intake.

[0008] FIG. 1 is a diagram illustrating an example of a system according to this embodiment. FIG. 2 is a diagram illustrating an outline of the first embodiment. FIG. 3 is a diagram illustrating the relationship between isoflavone intake and equol production. FIG. 4 is a process flow diagram illustrating an example of pre-processing according to the first embodiment. FIG. 5 is a process flow diagram illustrating an example of recommended amount output processing according to the first embodiment. FIG. 6 is a diagram illustrating an outline of the first embodiment. FIG. 7 is a process flow diagram illustrating an example of pre-processing according to the second embodiment. FIG. 8 is a process flow diagram illustrating an example of recommended amount output processing according to the second embodiment. FIG. 9 is a process flow diagram illustrating an example of recommended amount output processing according to the second embodiment. FIG. 10 is a diagram illustrating the second embodiment. FIG. 11 is a process flow diagram illustrating learning processing according to the third embodiment. FIG. 12 is a process flow diagram illustrating details of data collection and pre-processing. FIG. 13 is a diagram illustrating an example of data accumulated in a storage device. FIG. 14 is a process flow diagram illustrating an example of recommended amount output processing. FIG. 15 is a diagram illustrating an example of data stored in a storage device in the first modified example. FIG. 16 is a diagram showing an example of data stored in a storage device in the second modified example.

[0009] Hereinafter, an embodiment will be described with reference to the drawings.

[0010] <Embodiment 1> FIG. 1 is a diagram illustrating an example of a system according to this embodiment. System 100 includes server 1 and terminals 2 (2A, 2B, 2C). Server 1 estimates a user's equol deficiency and recommends the intake of at least one of isoflavones and equol based on the deficiency and the user's equol production capacity. Terminal 2 is, for example, a computer owned by a user who is to receive a recommendation for recommended components (i.e., at least one of isoflavones and equol) and recommended amounts to be ingested. Terminal 2 may also include a computer installed at a testing institution to which the user requests measurement of the amount of equol in excrement (hereinafter referred to as "equol excretion"). Server 1 collects the user's equol excretion amount via terminal 2 and outputs recommended amounts of components to be ingested to terminal 2. Server 1 and terminal 2 are communicatively connected via network 3. Network 3 includes, for example, an Internet Protocol (IP) network, and devices connected to network 3 can communicate based on a predetermined communication protocol. Part of the network 3 may be a telephone network (a fixed telephone network or a mobile communication network), an ad hoc network, an intranet, a VPN (Virtual Private Network), a LAN (Local Area Network), a Wireless LAN, a WAN (Wide Area Network), or the Internet.

[0011] In the present disclosure, isoflavones include flavonoids with a basic skeleton of isoflavone (3-phenylchromone) that are converted to equol in the intestines of individuals with equol-producing ability. Examples of such flavonoids include daidzein. Daidzein is converted to equol in the intestines of individuals with equol-producing ability. In addition, isoflavones in the present disclosure may further include genistein, glycitein, etc. In addition, examples of isoflavones in the present disclosure include soybean isoflavones, kudzu isoflavones, etc.

[0012] The server 1 is a computer and includes a processor 11, a storage device 12, and a communication interface (IF) 13. The processor 11 is an arithmetic processing device such as a CPU (Central Processing Unit) that executes programs to perform various processes according to this embodiment. The storage device 12 is a main storage device such as a RAM (Random Access Memory) or a ROM (Read Only Memory), and an auxiliary storage device (secondary storage device) such as a HDD (Hard-Disk Drive), an SSD (Solid State Drive), or a flash memory. The main storage device temporarily stores programs read by the processor 11 and secures a working area for the processor 11. The auxiliary storage device stores programs executed by the processor 11 and other data. The communication IF 13 is a network module for communicating via the network 3 and transmits and receives data based on a predetermined protocol.

[0013] The terminal 2 is a computer such as a tablet, a smartphone, a PC (Personal Computer), a wearable terminal, etc. The terminal 2 includes a processor, a storage device, a communication IF, and an input / output interface (IF) for inputting and outputting information to and from the user.

[0014] FIG. 2 is a diagram illustrating an outline of an embodiment. In this embodiment, each user's equol production capacity is measured, and the components and recommended amounts to be ingested are determined based on the production capacity. Specifically, each user refrains from dietary isoflavone intake for a predetermined period of time and then measures equol excretion after ingesting a predetermined amount of an isoflavone-containing supplement (FIG. 2: Measurement 1). Dietary isoflavone intake includes consuming isoflavone-containing foods as part of a meal. Supplements are products containing specific ingredients, including those processed into dosage forms such as tablets, capsules, granules, and powders, as well as health foods with specified ingredient contents. Equol is produced from isoflavones by intestinal bacteria (equol-producing bacteria). Therefore, equol production capacity varies depending on an individual's intestinal environment. Furthermore, the level of equol production capacity can be determined based on equol excretion, which can be measured by urine tests and / or stool tests (hereinafter referred to as "urinalysis, etc."), and / or equol blood concentration, which can be measured by blood tests. Figure 3 is a diagram illustrating the relationship between isoflavone intake and equol production. In the graph of Figure 3, the horizontal axis represents isoflavone intake, and the vertical axis represents equol production. The graph of Figure 3 plots a schematic example of equol production when a predetermined amount of isoflavone-containing supplement is taken, with circles. The higher the user's equol production ability, the greater the equol production, and the lower the user's production ability, the smaller the equol production. Furthermore, equol production increases in proportion to the amount of isoflavone intake. Furthermore, it is understood that equol production is reflected in the amount of equol excreted in the user's urine and feces. Measurement 1 in Figure 2 allows the degree of equol production ability of each individual user to be ascertained.

[0015] Thereafter, the measurement of equol excretion is repeated without dietary restrictions (Figure 2: Measurement 2). Measurement 2 allows the most recent equol production level of each individual user to be ascertained, and the amount of excess or deficiency relative to a predetermined reference value to be calculated. Furthermore, the amount of isoflavones, etc. (i.e., at least one of isoflavones and equol) to be ingested through food or supplements can be calculated according to the level of each individual user's equol production ability. Note that for users whose equol production ability is lower than the predetermined standard, the recommended amount of equol to be ingested instead of isoflavones can be calculated.

[0016] <Pre-Processing> FIG. 4 is a processing flow diagram showing an example of pre-processing. In pre-processing, an index value representing each user's equol production capacity is stored in a storage device. First, the processor 11 of the server 1 acquires, via the network 3, a measurement of the user's equol level, for example, measured at a testing institution (FIG. 4: S1). The measurement value can be a direct measurement of equol production capacity, such as a blood test to measure the blood concentration of equol after the user has abstained from dietary isoflavone intake for a predetermined period and then taken a predetermined amount of an isoflavone-containing supplement, or a urinalysis to measure the amount of equol excreted thereafter. It can also be a value determined by any method that can infer equol production capacity, such as sweat testing (testing sweat volume or sweat components, or a combination thereof) or heart rate changes. Generally, equol appears in the blood concentration approximately 8 hours after ingestion of an isoflavone-containing food, reaches a maximum blood concentration approximately 12 to 24 hours, and is excreted from the body approximately 72 hours later. Therefore, it is preferable that the blood test, urine test, and stool test be performed within a predetermined time period after the user has taken the supplement, etc., and within a longer predetermined time period. In other words, the period for measuring the blood concentration and excretion amount of equol can be appropriately determined based on, for example, the timing when the blood concentration is expected to be at its maximum. Furthermore, the measurement values ​​may be transmitted by the user operating the terminal 2, or may be transmitted from the terminal 2 of the testing institution.

[0017] After S1, the processor 11 stores an index value representing the level of productivity in association with the user's identification information in the storage device 12 (FIG. 4: S2). The index value may be a value corresponding to the measurement value obtained in S1, or may be the measurement value obtained in S1 itself. The index value may be calculated, for example, as the ratio of equol excretion (mg) to isoflavone intake (mg). Furthermore, after S2, the pre-processing ends. Furthermore, in this embodiment, the measurement of equol excretion (above S1 and S2) is repeated without any dietary restrictions.

[0018] <Recommended Amount Output Process> Figure 5 is a process flow diagram showing an example of the recommended amount output process. The recommended amount output process is initiated, for example, periodically or when a request is received from terminal 2 via user operation. In this embodiment, it is assumed that the user repeatedly measures the amount of equol excretion (Figure 4: S1 and S2) without restricting their diet, and that the measurement values ​​are stored in storage device 12.

[0019] The processor 11 of the server 1 determines whether the target user has a production capacity equal to or greater than a predetermined reference value ( FIG. 5 : S11). In this step, the processor 11 reads from the storage device 12 an index value stored in association with the target user's identification information and determines whether the index value is equal to or greater than a predetermined threshold. It is assumed that a threshold value for determining whether or not the target user has production capacity is stored in advance in the storage device 12. For example, the index value is defined as the ratio of equol excretion (mg) to isoflavone intake (mg), and a predetermined value close to 0 is stored as the threshold. If the index value is less than the threshold, it is determined that the target user has no production capacity ( S11 : NO), and information recommending an equol-containing supplement equivalent to a predetermined target amount is transmitted to the user's terminal 2 ( FIG. 5 : S12). For users whose equol production capacity is lower than the reference value, the process of S12 is repeated in the recommended amount output process.

[0020] On the other hand, if the index value is equal to or greater than the threshold, it is determined that the user has the ability to produce equol (S11: YES), and the processor 11 obtains the most recent measurement value of the target user's equol excretion amount (FIG. 5: S13). As mentioned above, in this embodiment, it is assumed that the measurement of equol excretion amount (FIG. 4: S1 and S2) is repeatedly carried out without any dietary restrictions.

[0021] After S13, the processor 11 recommends at least one of isoflavone-containing foods, isoflavone-containing supplements, and equol-containing supplements to the user ( FIG. 5 : S14). In this step, a recommended amount of isoflavones to be ingested is determined. The recommended intake pattern may be any of (1) equol-containing supplements only, (2) isoflavone-containing supplements only, (3) isoflavone-containing foods only, (4) equol-containing supplements and isoflavone-containing supplements, (5) equol-containing supplements and isoflavone-containing foods, (6) isoflavone-containing supplements and isoflavone-containing foods, or (7) equol-containing supplements, isoflavone-containing supplements, and isoflavone-containing foods. The recommended amount of the equol-containing supplement is calculated as the equol deficiency, which is the difference between a predetermined equol target value and the user's equol production amount. The recommended amount of the isoflavone-containing supplement is calculated as the isoflavone deficiency by converting the equol deficiency based on the index value of the user's equol production ability. The recommended amount of isoflavone-containing foods is calculated by converting the isoflavone deficiency amount using the standard isoflavone content per unit amount of food determined in advance. Furthermore, for the combinations shown in (4) to (7) above, combinations can be created that include a recommended amount of equol-containing supplement, a recommended amount of isoflavone-containing supplement, and a recommended amount of isoflavone-containing food in any desired composition ratio. Note that a predetermined ideal equol production amount is pre-stored in the storage device 12 as the target value. The equol production amount is a predicted value estimated based on the measured amount of equol excretion. For example, the equol production amount may be predicted by multiplying the excretion amount by a predetermined coefficient, or the equol excretion amount itself may be predicted as the equol production amount. For example, when the index value is the ratio of equol excretion (mg) to isoflavone intake (mg), the recommended amount of isoflavones, etc. that can be expected to compensate for the equol deficiency can be calculated by multiplying the index value by the inverse of the index value.Then, the processor 11 transmits information suggesting a recommended amount of at least one of the above (1) to (7) to the user's terminal 2 via the network IF 13 and the network 3. The information output to the terminal 2 is at least one of the above (1) to (7), and may be a plurality of pieces of information.

[0022] <Effects> According to the above embodiment, it is possible to estimate the equol deficiency of a user and make suggestions based on the individual user's equol deficiency. It is also possible to suggest an appropriate amount based on the user's production capacity. This makes it possible to improve the equol intake of individual users.

[0023] <Embodiment 2> Next, a second embodiment will be described. In this embodiment, a user carries a wearable device, such as a smartwatch or activity monitor, called terminal 2, and the user's equol production and deficiency are estimated based on data obtained from the wearable device. The wearable device is equipped with a biosensor and outputs vital data including at least one of the user's electrocardiogram, heart rate, blood oxygen concentration, information on sleep quality, body temperature (skin temperature), sweat rate, etc. Preferably, the vital data is data that reflects the effects of menopausal symptoms that can be expected to be improved by equol. It is also desirable to standardize the conditions under which the user measures the vital data as much as possible, such as after a predetermined period of rest.

[0024] FIG. 6 is a diagram illustrating an outline of an embodiment. In this embodiment, an individual user refrains from dietary isoflavone intake for a predetermined period of time, and then has their equol excretion measured by a laboratory or the like, while vital data is measured using a wearable device or the like ( FIG. 6 : Measurement 0). The predetermined period is sufficient time for equol to be excreted from the user's body, for example, approximately 72 hours after consuming an isoflavone-containing food. Furthermore, after refraining from dietary isoflavone intake and consuming a predetermined amount of isoflavone-containing supplements, their equol excretion is measured by a laboratory or the like, while vital data is measured using a wearable device or the like ( FIG. 6 : Measurement 1). In this embodiment, the characteristics of vital data from a state in which no isoflavones are ingested (Measurement 0) and a state in which a predetermined amount of isoflavones is ingested (Measurement 1) are learned, allowing isoflavone intake to be predicted from the vital data. Measurements 0 and 1 may be repeated to increase the training data and improve the prediction accuracy of the prediction model. Thereafter, vital data is measured without dietary restrictions (Measurement 2), and the user's equol production amount is predicted based on the vital data. At this time, no measurement is performed by a testing institution, etc. Furthermore, if Measurement 2 predicts that the target amount of equol has been produced, Measurement 2 is repeated without dietary restrictions. On the other hand, if Measurement 2 predicts that the target amount of equol has not been produced, the user takes a recommended amount of supplements in addition to their diet. Note that an isoflavone supplement is recommended for users with equol-producing ability, and an equol supplement is recommended for users without equol-producing ability. Furthermore, vital data is measured, for example, after a predetermined period of time has elapsed since the supplement's ingestion (Measurement 2'). Here, Measurement 2' is performed after a predetermined period of time sufficient for the effects of the supplement to become apparent. For example, if an isoflavone-containing supplement is ingested, Measurement 2' is performed approximately 48 hours after the ingestion of the isoflavone-containing supplement, which is the time it takes for isoflavones to be converted to equol. If Measurement 2' predicts that the target amount of equol has been produced, Measurement 2 is returned to, for example, without dietary restrictions.On the other hand, if Measurement 2' predicts that the target amount of equol is not being produced despite the intake of the recommended amount of supplements, the amount of equol production predicted from the vital data may be modified, for example, by modifying the prediction model created by machine learning itself or by setting a correction value to correct the predicted value output by the prediction model.

[0025] <Pre-Processing> FIG. 7 is a processing flow diagram illustrating an example of pre-processing according to the second embodiment. In the pre-processing, the relationship between the index value representing each user's equol production capacity, vital data, and equol amount is stored in a storage device. First, the processor 11 of the server 1 acquires, via the network 3, equol amount measurements taken by the user, for example, at a testing institution, and vital data measured by a wearable device (FIG. 7: S21). In this step, information from Measurements 0 and 1 in FIG. 6 is acquired. That is, measurements are acquired of the user's equol excretion amount measured by a testing institution after abstaining from dietary isoflavone intake for a predetermined period, and measurements are acquired of the equol excretion amount measured after abstaining from dietary isoflavone intake and ingesting a predetermined amount of an isoflavone-containing supplement. The measurements may be transmitted to the server 1 by the user operating the terminal 2, or may be transmitted from the testing institution's terminal 2 to the server 1. The vital data includes at least one of an electrocardiogram, heart rate, blood oxygen concentration, information on sleep quality, body temperature (skin temperature), sweat rate, sweat components, etc. The information on sleep quality may be sleep duration, sleep duration associated with sleep depth, snoring duration, etc. Furthermore, the heart rate may be a heart rate variability index that indicates the regularity of the heart rate in addition to or instead of the heart rate for a predetermined period of time.

[0026] After S1, the processor 11 performs machine learning to identify the relationship between the vital data characteristics and equol production (or excretion) amount when no isoflavones are ingested (FIG. 6: Measurement 0) and the vital data characteristics and equol production (or excretion) amount when a predetermined amount of isoflavone-containing supplements is ingested (FIG. 6: Measurement 1) (FIG. 7: S22). In this step, a prediction model is created that predicts equol production amount from the input vital data using techniques such as regression learning and deep learning. Note that the vital data and equol production amount are normalized or standardized as appropriate. Furthermore, if the equol excretion amount in Measurement 1 does not meet a predetermined threshold, it is not necessary to create a prediction model.

[0027] After S2, the processor 11 stores the index value representing the level of production ability and the prediction model in the storage device 12 in association with the user's identification information (FIG. 7: S23). The index value is the same as in the first embodiment. Furthermore, after S3, the pre-processing ends. Because the user's intestinal bacteria may change in the future, the prediction model may be reconstructed as appropriate. Furthermore, by repeatedly performing Measurements 0 and 1 of FIG. 6, training data may be increased to improve the prediction accuracy of the prediction model. Measurement 0 provides the equol production amount and vital data when the isoflavone intake is zero in the graph shown in FIG. 3. Measurement 1 provides the equol production amount and vital data when the isoflavone intake is the standard amount taken through supplements in the graph shown in FIG. 3. Furthermore, when isoflavone intake is increased, equol production does not continue to increase in proportion to isoflavone intake as shown in FIG. 3, but rather reaches an upper limit (threshold) of equol production for each individual. By increasing the number of measurements used as learning data, particularly Measurement 1, the accuracy of the slope of the graph shown in Figure 3 improves, and the accuracy of the prediction model that predicts the relationship between vital data and equol production levels also improves.

[0028] <Recommended Amount Output Process> Figures 8 and 9 are process flow diagrams showing an example of the recommended amount output process. In this embodiment, the recommended amount output process is also started, for example, periodically or when a request is received from terminal 2 by user operation. In this embodiment, terminal 2, which is a wearable terminal, continuously measures vital data (Figure 6: Measurement 2, Measurement 2') and transmits it to server 1 at a predetermined timing. It is assumed that the vital data received from terminal 2 is accumulated in storage device 12 of server 1. It is also assumed that the prediction model created in the pre-processing of Figure 7 is stored in storage device 12.

[0029] The processor 11 of the server 1 determines whether the target user has a production capacity equal to or greater than a predetermined standard value (FIG. 8: S31). This step is the same as S11 in FIG. 5. If the index value is less than the threshold, it is determined that the user has no production capacity (S31: NO), and information recommending an equol supplement equivalent to a predetermined target amount is sent to the user's terminal 2 (FIG. 8: S32), terminating the processing of FIG. 8. For users whose equol production capacity is lower than the standard, it is recommended that they take an equol-containing supplement equivalent to the predetermined target amount. In other words, S32 is repeated.

[0030] On the other hand, if the index value is equal to or greater than the threshold, it is determined that the subject has the ability to produce equol (S31: YES), and the processor 11 predicts the subject's equol production amount from the most recent vital data (FIG. 8: S33). In this step, the vital data measured in Measurement 2 is input into the prediction model, and a predicted value for the equol production amount is calculated.

[0031] After S33, the processor 11 recommends at least one of isoflavone-containing foods, isoflavone supplements, and equol-containing supplements (hereinafter referred to as "supplements, etc.") to the user ( FIG. 8 : S34). This step is similar to S14 in FIG. 5. FIG. 10 is a diagram for explaining an embodiment. In the graph of FIG. 10, the horizontal axis represents time and the vertical axis represents the predicted value of equol production. For example, assume that at time t, the predicted value in S33 is the value e1 plotted with a circle. At this time, a recommended amount of isoflavone that is expected to make up for the deficiency equivalent to the difference d1 between the target value and the predicted value e1 is recommended to the user in S34.

[0032] After S34, input is accepted from the user via terminal 2 as to whether the user has taken the recommended amount of supplements, etc. ( FIG. 9 : S35). In this embodiment, after pre-processing, the amount of equol excretion is not generally measured, and the amount of equol is predicted from vital data while correcting the prediction model. In this step, input is accepted as to whether the user has taken the recommended amount of supplements, etc., for later use in determining whether or not corrections to the learning model are necessary. Alternatively, the server 1 may make an inquiry to the user's terminal 2 and obtain a response in S34, or application software installed on terminal 2 may prompt the user for input at a predetermined timing, and the server 1 may obtain a response in S34.

[0033] After S35, after a predetermined time has elapsed, the processor 11 predicts the amount of equol produced for the target user based on the most recent vital data ( S36 in FIG. 9 ). This step is similar to S33 in FIG. 8 , but the prediction is made using the measurements from Measurement 2' in FIG. 6 . As described above, equol generally appears in the blood approximately 8 hours after ingesting an isoflavone-containing food, reaches a maximum blood concentration approximately 12 to 24 hours, and is excreted from the body approximately 72 hours later. Therefore, if the user ingests a supplement or the like after S34 in FIG. 8 , the equol amount is predicted in S36 after a predetermined time interval until equol is produced. In other words, the period for measuring vital data can be appropriately determined based on the timing when blood concentrations are expected to peak, such as 48 hours after ingesting an isoflavone-containing supplement.

[0034] After S36, the processor 11 also suggests supplements and the like to the user (FIG. 9: S37). This step is similar to S14 in FIG. 5 and S34 in FIG.

[0035] After S37, the processor 11 determines whether to revise the prediction model, etc. ( FIG. 9 : S38). Specifically, the processor 11 determines whether the user has taken the recommended amount of supplements, etc., and the predicted value of equol production is less than a predetermined threshold. Whether the user has taken the recommended amount of supplements, etc. is determined based on the response in S35. The predetermined threshold may be a target value for equol production, or a value that takes a predetermined tolerance range into account based on the target value. In other words, if the user has taken the recommended amount of supplements, etc. to make up for an equol deficiency, but the equol production subsequently falls short of the target value, the processor 11 determines that the prediction model should be revised.

[0036] For example, if the user takes a recommended amount of supplements, etc., and the predicted value of equol production reaches the target value as expected, as shown at time t+1 in FIG. 10 , the prediction model functions as expected and does not need to be corrected. On the other hand, if the user takes the recommended amount of supplements, etc., but the predicted value of equol production falls short of the target value, as shown at time t+1' in FIG. 10 , the prediction model is corrected based on the difference d2 between the target value and the predicted value e2. It may also be determined that the prediction model needs to be corrected if the user takes the recommended amount of supplements, etc., and the predicted value of equol production is determined to be below a predetermined threshold multiple times. It may also be determined that the prediction model needs to be corrected if the user takes the recommended amount of supplements, etc., and the predicted value of equol production is determined to significantly deviate from the target value. The predicted value of equol production may also be determined using a moving average over the most recent predetermined period. The prediction model may also be corrected if the predicted value of equol production deviates beyond the acceptable range in the direction toward the target value.

[0037] If it is determined that the prediction model should be revised (S38: YES), the processor 11 revises the prediction model, etc., based on the difference between the target value and the predicted value ( FIG. 9 : S39). It is also possible to adjust coefficients for correcting the prediction results of the prediction model, the equol deficiency, or the recommended amount of supplements to be taken. In this step, if the user ingests the recommended amount of isoflavones, the prediction model (or correction coefficient) is revised so that the predicted value of equol production approaches the target value. For example, if the recommended amount is determined based on the difference d1 between the target amount and equol production at time t in FIG. 10 , and the user subsequently ingests the recommended amount of supplements, etc., but a difference d2 occurs between the target amount and equol production at time t+1′, it can be seen that the recommended amount at time t would only increase equol production by d1-d2. In other words, it can be interpreted that the user should have ingested isoflavones at time t in an amount (d1 / d1-d2) times the calculated recommended amount, and therefore the correction coefficient is set to, for example, (d1 / d1-d2). The correction coefficient may be adjusted based on a predetermined learning rate so as to gradually approach (d1 / d1-d2).The prediction model itself may also be modified so as to predict a lower amount of equol production.

[0038] If it is determined in S37 that the prediction model should not be revised (S38: NO) or after S39, the processor 11 returns to S35 and repeats the process. Note that in Measurement 2 of FIG. 6, if the user has had the amount of equol excreted measured at a testing institution, for example, the actual measured value may be used to revise the prediction model.

[0039] <Effects> In this embodiment, after pre-processing, the equol production amount can be predicted from vital data while modifying the prediction model, without the need to measure equol excretion at a testing institution, etc. Furthermore, an appropriate amount can be suggested based on the deficiency and production capacity of each individual user. This makes it possible to improve the equol intake of each individual user.

[0040] <Embodiment 3> Next, a third embodiment will be described. In this embodiment, server 1 learns information about the user's equol production capacity, information about the user's condition, such as subjective symptoms related to menopausal disorders, and the relationship between the intake amounts of at least one of isoflavones and equol, and performs a process of recommending ingredients to the user (i.e., at least one of isoflavones and equol) based on the learning results. Note that the learning process and the recommendation process may be performed by different devices. Terminal 2 is, for example, a computer owned by the user who is to receive recommendations about ingredients to be ingested. Server 1 collects information about the user's condition, such as the amount of equol in the user's excrement (hereinafter referred to as "equol excretion amount"), via terminal 2, and outputs recommended amounts of ingredients to be ingested to terminal 2. Note that terminal 2 may also include a computer, such as a testing institution's computer, that measures the user's equol excretion amount. Server 1 and terminal 2 are communicatively connected via network 3.

[0041] In this embodiment, the equol production capacity of each user is measured, and the components and recommended amounts to be ingested are determined based on the production capacity. Specifically, each user refrains from dietary isoflavone intake for a predetermined period of time and then measures the amount of equol excretion after ingesting a predetermined amount of an isoflavone-containing supplement. By measuring the amount of equol excretion in advance, the level of equol production capacity of each user can be determined. Generally, after ingesting an isoflavone-containing food, equol appears in the blood concentration approximately 8 hours, reaches its maximum within 12 to 24 hours, and is excreted from the body within approximately 72 hours. Therefore, it is preferable to perform blood tests, urine tests, and stool tests within a predetermined time period after the user ingests the supplement, etc., but within a longer predetermined time period. In other words, the period for measuring equol blood concentration and excretion can be appropriately determined based on, for example, the timing when blood concentration is expected to reach its maximum.

[0042] In this embodiment, information about the user's condition, such as the level of symptoms that may be caused by menopausal disorders, and information about the ingredients (i.e., at least one of isoflavones and equol) and their intake amounts are collected from multiple users, and recommendations are made based on the ingredients and intake amounts of users with similar conditions. The aforementioned ingredients can generally be ingested through at least one of isoflavone-containing foods, isoflavone-containing supplements, and equol-containing supplements. Information about the user's condition includes, for example, information representing the severity of at least one of symptoms that may be caused by menopausal disorders, such as stiff shoulders, fatigue, headaches, hot flashes, lower back pain, sweating, insomnia, irritability, skin itching, palpitations, depression, dizziness, stomach upset, and vaginal dryness, expressed as a graded score. Information about the user's condition, the ingredients ingested, and their intake amounts may also be collected from users, for example, as responses to a questionnaire. The recommendation process may also involve, for example, user-based collaborative filtering. That is, the components and intake amounts of users with similar symptom trends to the target user are identified, and the components and recommended amounts to be recommended to the target user are determined based on the identified components and intake amounts. The recommendation process is not particularly limited, and model-based collaborative filtering may be performed by pre-modeling the relationship between symptom trends and ingested components and intake amounts. Alternatively, the relationship between symptom trends as explanatory variables and ingested components and intake amounts as objective variables may be modeled using deep learning, and recommendation processing based on a model that determines recommended amounts based on the target user's symptoms may be performed. If the equol production ability of the user responding to the questionnaire is known, the intake of isoflavone-containing foods or isoflavone-containing supplements may be converted to equol production amount, and the objective variable may be unified to equol intake. Alternatively, rules based on the relationship between symptom trends and intake amounts may be created in advance, and rule-based recommendation processing may be performed.

[0043] Furthermore, it is preferable to perform recommendation processing based on the intake of isoflavones, etc., of users with similar equol production capabilities. That is, equol production capabilities vary from person to person, and the preferred intake of isoflavones, etc., varies depending on production capabilities. Therefore, in this embodiment, recommendation processing is performed for each user with similar equol production capabilities based on the similarity of symptoms.

[0044] <Learning Process> Fig. 11 is a process flow diagram illustrating the learning process according to this embodiment. In the learning process, the server 1 collects an index value representing each user's equol production capacity and information related to the user's condition, and performs preprocessing (Fig. 11: S41). Fig. 12 is a process flow diagram illustrating the details of data collection and preprocessing. The processor 11 of the server 1 acquires one piece of unprocessed data (Fig. 12: S51).

[0045] FIG. 13 is a diagram showing an example of data stored in the storage device 12. Note that the table shown in this embodiment is an example of a database, and information may be properly normalized and stored in multiple tables, or denormalized and stored collectively in a single table. The table in FIG. 13 includes the attributes "User ID," "Production Capacity," "Condition," and "Intake." The "User ID" field contains identification information for uniquely identifying the user. The "Production Capacity" field contains an index value representing the user's level of equol production capacity. The index value may be a value corresponding to a measured value of equol excretion previously measured by the user, or the measured value itself. The index value may also be calculated, for example, as the ratio of equol excretion (mg) to isoflavone intake (mg). The "Condition" field contains attributes representing symptoms such as "stiff shoulders," "easily fatigued," ..., "lower back pain," "sweating," and "insomnia." Each "Condition" field contains a value entered by the user on a five-point scale, ranging from "1" indicating poor condition to "5" indicating good condition. The "intake" field stores the amount of isoflavones the user ingests on a daily basis, or a converted value for the amount of equol produced from isoflavones. For example, if a user responds to the questionnaire that they consume one pack of natto and one cup (200g) of soy milk per day, a value converted to the daily isoflavone intake based on a predetermined food-to-isoflavone conversion formula is registered in the "intake" field. In S51 of FIG. 12, the processor 11 acquires data corresponding to one record shown in FIG. 13, for example.

[0046] After S51, processor 11 determines whether the user indicated by the acquired data has the ability to produce equol ( FIG. 12 : S52). In this step, it is determined whether the productivity index value of the acquired data is equal to or greater than a predetermined threshold. Note that a predetermined value close to 0, for example, is stored as the threshold for determining whether or not the user has productivity. If the productivity index value is less than the threshold, it is determined that the user does not have productivity ( S52: NO), and processor 11 classifies the data as first learning data (also referred to as learning data 1) ( FIG. 12 : S53). For example, user 1 in FIG. 13 has a productivity of 0, and the record is classified as learning data 1.

[0047] On the other hand, if the productivity index value is equal to or greater than the threshold, it is determined that the user has productivity (S52: YES), and the processor 11 determines whether the user's productivity is equal to or greater than a predetermined standard ( FIG. 12: S54). It is assumed that a second threshold value greater than the threshold value of S52 is set as the predetermined standard. For example, the predetermined standard is set to 0.2. If the productivity index value does not meet the predetermined standard ( S54: NO), the processor 11 classifies the data as second learning data (also referred to as learning data 2) ( FIG. 12: S55). User 2 in FIG. 13 has a productivity of 0.1, which is lower than the standard, and the record is classified as learning data 2.

[0048] On the other hand, if the productivity index value is equal to or greater than the predetermined standard (S54: YES), the processor 11 classifies the data into third learning data (also referred to as learning data 3) ( FIG. 12 : S56). User 3 in FIG. 13 has a productivity of 0.3, which is above the standard, and the record is classified into learning data 3.

[0049] Then, after S53, S55, or S56, the processor 11 determines whether to end the pre-processing (S57 in FIG. 12). For example, if all the records shown in FIG. 13 have been processed, the processor 11 determines to end the processing (YES in S57) and returns to the processing in FIG. 11. On the other hand, if there are unprocessed records, the processor 11 determines not to end the processing (NO in S57) and returns to S51 to repeat the processing in FIG. 12.

[0050] By performing the preprocessing described above, the training data can be classified into groups of users with similar productivity. Note that the number of groups for classification is not limited to three.

[0051] After S41 in Fig. 11, the processor 11 constructs a model for each of the classified data (S42 in Fig. 11). In this step, a model is created when performing model-based collaborative filtering or recommendation processing using deep learning. Note that, for example, when performing user-based collaborative filtering, S42 may be omitted.

[0052] 14 is a process flow diagram showing an example of the recommended amount output process. The recommended amount output process is started, for example, periodically or when a request is received from the terminal 2 by a user operation.

[0053] The processor 11 of the server 1 acquires the productivity index value and information about the user's condition (e.g., information about symptoms) for the user for whom the recommended amount is to be proposed ( FIG. 14 : S61). This information may be received from the terminal 2, for example, as a response to a questionnaire for the user.

[0054] After S61, processor 11 determines whether the target user has a productivity equal to or greater than a predetermined standard value ( FIG. 14 : S62). This step performs the same determination as S52 in FIG. 12. That is, processor 11 determines whether the index value stored in association with the target user's identification information is equal to or greater than a predetermined threshold. If the index value is less than the threshold, it is determined that the target user has no productivity ( S62: NO), and information recommending an equol-containing supplement equivalent to a predetermined target amount is transmitted to the user's terminal 2 ( FIG. 14 : S63). For users whose equol productivity is lower than the standard, the process of S63 is repeated in the recommended amount output process.

[0055] On the other hand, if the index value is equal to or greater than the threshold, it is determined that the target user has equol production capacity (S62: YES), and the processor 11 determines a recommended amount for the target user based on the intake amounts of other users with similar information regarding the user's condition (FIG. 14: S64). In this step, the components and recommended amounts to be ingested by the target user are determined using, for example, the model created in S42 of FIG. 11. That is, the recommended amounts of components to be ingested by the target user are determined based on the components and intake amounts ingested by other users with similar information regarding equol production capacity and condition to the target user. Note that when performing user-based collaborative filtering, instead of using a trained model, other users with similar information regarding equol production capacity and condition to the target user are extracted from a table such as that shown in FIG. 13, and the components to be ingested and recommended amounts are determined based on the components and intake amounts ingested by the extracted users. Note that when multiple users are extracted, the recommended amounts may be determined based on statistics such as the average intake amounts of the extracted users.

[0056] After S64, processor 11 suggests at least one of isoflavone-containing foods, isoflavone-containing supplements, and equol-containing supplements to the user ( FIG. 14 : S65). Recommended intake patterns include at least one of (1) equol-containing supplements only, (2) isoflavone-containing supplements only, (3) isoflavone-containing foods only, (4) equol-containing supplements and isoflavone-containing supplements, (5) equol-containing supplements and isoflavone-containing foods, (6) isoflavone-containing supplements and isoflavone-containing foods, and (7) equol-containing supplements, isoflavone-containing supplements, and isoflavone-containing foods. Processor 11 transmits information suggesting the ingredients to be ingested and the recommended amounts determined in S64 to user terminal 2 via network IF 13 and network 3.

[0057] <Effects> According to the above embodiment, it is possible to suggest an intake amount for each individual user based on the similarity of their productivity and condition with other users, thereby making it possible to improve the equol intake of each individual user.

[0058] <Variation 1> Next, a first variation will be described, focusing on differences from the above-described embodiment. In this variation, a user carries a wearable device, such as a smartwatch or activity tracker, called terminal 2, and data obtained from the wearable device is used as information representing the user's condition. The wearable device is equipped with a biosensor and outputs vital data including at least one of the user's electrocardiogram, heart rate, blood oxygen concentration, information on sleep quality, body temperature (skin temperature), sweat rate, sweat components, etc. Preferably, the vital data is data that reflects the effects of menopausal symptoms, which can be expected to improve with equol. It is also desirable to standardize the conditions under which the user measures the vital data, such as after a predetermined period of rest.

[0059] The vital data may be data directly obtained from the wearable device, or may be input by the user as a five-point scale indicating the health status. FIG. 15 is a diagram showing an example of data stored in the storage device 12 in this modification. The table in FIG. 15 includes the following attributes as "conditions": "electrocardiogram," "heart rate," ..., "sleep quality," "body temperature," and "amount of sweat." Each "condition" field may contain a value entered by the user using a five-point scale ranging from "1" indicating poor health to "5" indicating good health, or a value automatically converted from the vital data. The vital data itself may be registered in each "condition" field.

[0060] Furthermore, by using vital data according to the modified example instead of or in addition to the information on the severity of symptoms collected by questionnaires or the like in the above-described embodiment, the components and recommended amounts to be ingested by the target user are determined based on the components and intake amounts ingested by other users whose information on equol production capacity and condition is similar to that of the target user. Even in this way, it is possible to suggest components and intake amounts to be ingested for each individual user based on the similarity of production capacity and condition with other users. Furthermore, using the vital data itself can reduce the hassle of users having to answer questionnaires.

[0061] Next, a second modification will be described, focusing on the differences from the above-described embodiment. In this modification, the conditions and intake amounts of other users at multiple points in time are stored in the storage device 12, and recommendation processing is performed for other users whose conditions have improved based on the ingredients and intake amounts they were taking before the improvement.

[0062] FIG. 16 is a diagram showing an example of data stored in the storage device 12 in this modified example. User 4's condition on May 25th is generally higher than that on March 20th. On the other hand, User 5's condition on April 10th is not considered to have improved overall compared to that on August 12th. Whether the condition has improved can be determined based on any predetermined criteria. The processor 11 of the server 1 can determine whether the condition has improved based on, for example, the average increase or decrease in the condition for each condition, or can further set conditions such as whether the value has not decreased by more than a predetermined standard. In this modified example, a recommendation process is performed for a user whose condition has improved, such as User 4, based on the ingredients and intake amounts they were consuming before the improvement.

[0063] That is, in the process of Fig. 12, not only the magnitude of productivity but also the data of users whose condition is judged to have improved is classified into learning data. Also, in the process of Fig. 14, the components to be ingested by the target user and the recommended amounts are determined based on the components and intake amounts ingested by other users whose condition is judged to have improved before the improvement.

[0064] According to this modification, it is possible to suggest an intake amount based on the similarity of productivity and condition with users who have a track record of improving their condition. Therefore, it is possible to suggest ingredients and recommended amounts that are more likely to improve. Note that the first and second modifications may be combined to suggest an intake amount based on the similarity of productivity and condition with users whose vital data evaluation has improved.

[0065] <Variation 3> If the condition of the target user does not improve even after ingesting the recommended amount of a component for a predetermined period of time, the recommended amount proposed in S65 of Fig. 14 may be changed by modifying the model created in S42 of Fig. 11 or adjusting a coefficient for correcting the determined recommended amount. For example, if a correction coefficient is used, the determined recommended amount is further multiplied by the coefficient in S65 of Fig. 14 to calculate the recommended amount to be output. Furthermore, for example, after increasing or decreasing the recommended amount, the recommended amount may be repeatedly changed depending on whether the condition has improved after a predetermined period of time.

[0066] <Other Modifications> The configurations and combinations thereof in each embodiment are merely examples, and additions, omissions, substitutions, and other modifications of the configurations are possible as appropriate without departing from the spirit of the present disclosure. The present disclosure is not limited by the embodiments, but only by the scope of the claims. Furthermore, each aspect disclosed in this specification can be combined with any other feature disclosed in this specification.

[0067] At least some of the functions of the server 1 may be distributed to multiple devices, or multiple devices may provide the same functions in parallel. Furthermore, the prediction model described in the second embodiment may be a rule-based prediction model rather than one created by machine learning.

[0068] The present disclosure also includes a method and a computer program for executing the above-described process, and a computer-readable recording medium having the program recorded thereon. The recording medium having the program recorded thereon enables the above-described process by causing a computer to execute the program.

[0069] Here, a computer-readable recording medium refers to a recording medium that stores information such as data and programs electrically, magnetically, optically, mechanically, or chemically and can be read by a computer. Among such recording media, those that can be removed from a computer include flexible disks, magneto-optical disks, optical disks, magnetic tapes, memory cards, etc. Furthermore, recording media that are fixed to a computer include HDDs, SSDs (Solid State Drives), ROMs, etc.

[0070] The present application also includes the following supplementary matters: <Supplementary Note 1> A recommended amount output system including one or more computers that execute the following steps: acquire an index value representing the level of a target user's equol production ability and information representing the target user's condition; determine a recommended amount of equol and / or isoflavones to be consumed by the target user based on the equol and / or isoflavone intakes of other users whose information representing equol production ability and condition is similar to that of the target user, using information of other users, including index values ​​representing the level of equol production ability of other users, information representing the other users' conditions, and information representing the other users' equol and / or isoflavone intakes, which has been stored in advance in a storage device; and output the determined recommended amount. <Supplementary Note 2> The recommended amount output system according to Supplementary Note 1, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user abstained from dietary isoflavone intake for at least a predetermined period of time and ingested a predetermined amount of an isoflavone-containing supplement. <Supplementary Note 3> The recommended amount output system according to Supplementary Note 1 or 2, wherein the information representing the condition is information representing the degree to which the user has answered about symptoms that may appear due to menopausal disorders. <Supplementary Note 4> The recommended amount output system according to Supplementary Note 1 or 2, wherein the information representing the condition includes at least one of the user's electrocardiogram, heart rate, blood oxygen concentration, an index representing sleep quality, body temperature, sweat amount, and sweat components. <Supplementary Note 5> The recommended amount output system according to Supplementary Note 1 or 2, wherein the information of the other user includes information representing the other user's condition at multiple time points and information representing the other user's intake of equol and / or isoflavones, and the determination process uses the information of the other user whose information representing the other user's condition at a second time point after the first time point is improved compared to the information representing the other user's condition at the first time point, to determine a recommended amount of equol and / or isoflavones to be consumed by the target user based on the equol and / or isoflavone intake of the other user at the first time point, whose equol production ability and information representing the condition at the first time point are similar to those of the target user.<Supplementary Note 6> The recommended amount output system according to Supplementary Note 1 or 2, wherein if the target user has taken the output recommended amount of equol and / or isoflavones for a predetermined period of time and the information representing the target user's condition is not determined to have improved based on predetermined criteria, the recommended amount is corrected in the determination process. <Supplementary Note 7> A recommended amount output method, performed by one or more computers, comprising: acquiring an index value representing the target user's level of equol production ability and information representing the target user's condition; determining a recommended amount of equol and / or isoflavones to be consumed by the target user based on the equol and / or isoflavone intakes of other users whose information representing equol production ability and condition is similar to that of the target user, using information of other users, including information representing the other users' level of equol production ability, information representing the other users' conditions, and information representing the other users' equol and / or isoflavone intakes, which has been stored in advance in a storage device; and outputting the determined recommended amount. <Supplementary Note 8> The recommended amount output method according to Supplementary Note 7, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user abstained from dietary isoflavone intake for a predetermined period of time or longer and ingested a predetermined amount of an isoflavone-containing supplement. <Supplementary Note 9> A program for causing one or more computers to execute the following steps: acquiring an index value representing the level of equol production ability of a target user and information representing the target user's condition; determining a recommended amount of equol and / or isoflavones to be consumed by the target user based on the equol and / or isoflavone intakes of other users whose information representing equol production ability and condition is similar to that of the target user, using information of other users that has been stored in advance in a storage device and includes information representing the other users' levels of equol production ability, information representing the other users' conditions, and information representing the other users' equol and / or isoflavone intakes; and outputting the determined recommended amount.<Appendix 10> The program according to Appendix 9, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user refrained from consuming isoflavones through diet for a predetermined period of time or longer and took a predetermined amount of an isoflavone-containing supplement.

[0071] 100: System 1: Server, 11: Processor, 12: Storage device, 13: Communication IF 2: Terminal 3: Network

Claims

1. A recommended amount output system including one or more computers that executes the following steps: calculate a user's equol deficiency using a predicted value of the user's equol production amount and a predetermined target amount; calculate a recommended amount of equol and / or isoflavones that the user should consume based on the calculated equol deficiency and an index value that represents the user's level of equol production ability; and output the calculated recommended amount.

2. The recommended amount output system according to claim 1, wherein the predicted value of the equol production amount is predicted using the equol excretion amount of the user.

3. The recommended amount output system according to claim 1 or 2, wherein the predicted value of the equol production amount is predicted using vital data obtained from the user's wearable device.

4. The recommended amount output system according to claim 3, wherein the vital data includes at least one of the user's electrocardiogram, heart rate, blood oxygen concentration, an index representing sleep quality, body temperature, sweat amount, and sweat components.

5. The recommended amount output system of claim 1, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user refrained from consuming isoflavones through diet for a predetermined period of time or longer and took a predetermined amount of isoflavone-containing supplements.

6. The recommended amount output system of claim 3, wherein the predicted value of the amount of equol production is predicted using a prediction model that has learned the relationship between the vital data and the amount of equol excreted after the user has refrained from consuming isoflavones through food for a specified period of time, and the vital data and the amount of equol excreted after the user has refrained from consuming isoflavones through food for a specified period of time and has taken a specified amount of an isoflavone-containing supplement.

7. The recommended amount output system of claim 6, wherein if the post-ingestion equol deficiency calculated after receiving information indicating that the user has ingested the recommended amount of isoflavone satisfies a predetermined condition, the prediction model or a correction coefficient for the deficiency or the recommended amount is modified based on the relationship between the target amount and the post-ingestion equol deficiency.

8. A recommended amount output method implemented by one or more computers that calculates the user's equol deficiency using a predicted value of the user's equol production amount and a predetermined target amount; calculates a recommended amount of equol and / or isoflavones that the user should ingest based on the calculated equol deficiency and an index value representing the user's level of equol production ability; and outputs the calculated recommended amount.

9. The recommended amount output method according to claim 8, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has elapsed since the user refrained from consuming isoflavones through diet for a predetermined period of time or longer and took a predetermined amount of an isoflavone-containing supplement.

10. A program for causing one or more computers to perform the following steps: calculating a user's equol deficiency using a predicted value of the user's equol production amount and a predetermined target amount; calculating a recommended amount of equol and / or isoflavones that the user should consume based on the calculated equol deficiency and an index value representing the user's level of equol production ability; and outputting the calculated recommended amount.

11. The program described in claim 10, wherein the index value is a value corresponding to the amount of equol excreted or blood concentration after a predetermined time has passed since the user refrained from consuming isoflavones through diet for a predetermined period of time or longer and took a predetermined amount of isoflavone-containing supplements.

Images