KR20200137097A - Method for organizing and recommending diet based on personal preference - Google Patents

Abstract

The present invention relates to a technology for organizing and recommending a diet for health. A method for organizing the diet based on personal preference extracts diet information from food intake records input from a plurality of people, clusters the extracted diet information to produce at least one preferred dietary groups, and calculates the average of dietary information for each preferred dietary group using a health index calculated from medical information or a health index calculated from statistical health information of multiple people as a weight to be set up as a representative healthy diet of the group.

Description

Method for organizing and recommending diet based on personal preference}

The present invention relates to a technology for selecting a diet for health, and in particular, in consideration of an individual's preferred diet, the current preferred diet and health effects are comprehensively considered to construct a diet that can contribute to health or recommend a healthy diet. It relates to an apparatus, a method, and a recording medium recording the methods.

Food is the most primitive and essential element for human survival, and the composition of an appropriate diet was one of the main concerns of mankind. Along with the advancement of technology, various technologies have been introduced to provide a customized diet suitable for users. Looking at the direction, it can be broadly summarized into the following two levels.

First, various services have emerged that provide customized healthy diets in consideration of the user's physical ability or health status. The prior art literature cited below introduces a customized diet recommendation service that can improve the user's health by evaluating the user's health status or nutritional status. However, these services have become far from the act of enjoying food by focusing too much on the point of view of health, resulting in an inappropriate diet that does not take into account the eating habits.

Second, there is a service that analyzes the user's taste or consumption pattern and provides a customized meal based on it. This is a kind of personalized service derived from the user's purchase history and operates in a manner such as recommending similar foods or suggesting alternative foods according to extremely consumption trends. However, these services are also undesirable in that they are far from the viewpoint of inducing healthy eating habits.

In summary, according to the westernized eating habits, it is a reality that a diet focused solely on health from a traditional point of view is not welcomed, and it is true that there are side effects even in the opposite case.

Therefore, it is necessary to look at how to define the characteristics of a group similar to one's favorite diet, and in particular, a technical means is required to derive a diet that is good for health while being preferred by users of these groups.

Korean Patent Publication No. 2009-0046994, published on May 12, 2009, "Customized meal recommendation service"

The technical problem to be solved by the present invention is that the conventional healthy diet recommendation techniques focus only on providing an ideal diet based on the individual's health status or new conditions, thereby ignoring the individual's preferred diet and enjoying food intake for health. In solving the problem of losing a person, and in matching the health status and diet, the goal is to overcome the limitation of not being able to provide a representative healthy diet suitable for each individual because it is only buried in the viewpoint of nutritional credits in food.

In order to solve the above technical problem, a method for configuring a diet based on personal preference according to an embodiment of the present invention includes: (a) extracting dietary information from food intake records input from a plurality of people; (b) clustering the extracted dietary information to generate at least one preferred dietary group; And (c) a health index calculated from medical information or a health index calculated from a number of people's statistical health as a weight, and calculating the average of the diet information for each of the preferred diet groups to set as the representative healthy diet of the group. Step; includes.

In the method for constructing a diet based on personal preference according to an embodiment, the (a) extracting dietary information comprises: (a1) calculating a total intake from the product of the intake frequency and intake amount for each food according to the food intake record. ; (a2) normalizing the calculated total intake based on the maximum intake for each food; And (a3) generating an array or a vector including the total intake amount normalized for each food intake of each person, and setting it as dietary information.

In the method of constructing a diet based on personal preference according to an embodiment, the (b) generating a preference diet group may include: (b1) clustering the extracted diet information according to similarity of diet information; And (b2) identifying at least one preferred dietary group by assigning a cluster identifier to which dietary information belongs to each person's ingested food.

In the method for constructing a diet based on personal preference according to an embodiment, the (c) setting as a representative health diet for the group includes: (c1) a health index derived from medical information or statistical health information of a plurality of people. Calculating a health index corresponding to a statistical calculation criterion; (c2) receiving a list of preferred dietary groups that match and store diet information for each person and the preferred diet group to which the person belongs, and mapping the preferred diet group to a representative healthy diet one-to-one; And (c3) multiplying each diet information and the corresponding health index as a weight for each individual person data belonging to the preferred diet group, multiplying it with diet information, and dividing it by the sum of the health index to represent the group. Deriving a healthy diet; may include.

In the method for constructing a diet based on personal preference according to an embodiment, the health index is any one of a health level based on medical information, a diversity of intestinal microbes of a number of people, similarity with a standard intestinal microbiota, and a combination thereof. It is preferable that the intestinal health is also calculated based on.

In order to solve the above technical problem, a method of recommending a diet based on personal preference according to another embodiment of the present invention includes: (a) extracting dietary information from food intake records input from a plurality of people; (b) clustering the extracted dietary information to generate at least one preferred dietary group, wherein the list of preferred dietary groups matches and stores dietary information for each person and a preferred dietary group to which they belong; (c) A health index calculated from medical information or a health index calculated from statistical health information of a number of people is used as a weight and the average of the diet information is calculated for each of the preferred diet groups, thereby setting it as the representative healthy diet for the group. step; (d) receiving an individual's dietary information and determining a preferred dietary group to which the individual belongs by referring to the at least one preferred dietary group; And (e) providing a representative healthy diet for the individual.

In the diet recommendation method based on personal preference according to another embodiment, the diet information of the plurality of people or individuals is calculated by calculating a total intake amount from the product of the intake frequency and intake amount for each food according to the food intake record, and the calculated total The intake amount is normalized based on the maximum intake amount for each food, and may be set by generating an array or vector including the total intake amount normalized for each food intake of each person.

In the method for recommending a diet based on personal preference according to another embodiment, the (c) setting as a representative healthy diet for the group includes: (c1) a health index derived from medical information or statistical health information of a plurality of people. Calculating a health index corresponding to a statistical calculation criterion; (c2) receiving a list of preferred dietary groups that match and store diet information for each person and the preferred diet group to which the person belongs, and mapping the preferred diet group to a representative healthy diet one-to-one; And (c3) multiplying each diet information and the corresponding health index as a weight for each individual person data belonging to the preferred diet group, multiplying it with diet information, and dividing it by the sum of the health index to represent the group. Deriving a healthy diet; may include.

In the method for recommending a diet based on personal preference according to another embodiment, the health index is any one of a health level based on medical information, a diversity of intestinal microbes of a number of people, similarity to a standard intestinal microbiota, and a combination thereof. It is preferable that the intestinal health is also calculated based on.

In the method for recommending a diet based on personal preference according to another embodiment, the step of (d) determining a preferred diet group to which the individual belongs may include generating at least one preferred diet group in the step of (b) creating a preferred diet group. By classifying the individual's dietary information according to the clustering method, a group most similar to the individual's preferred diet can be selected.

In the method for recommending a diet based on personal preference according to another embodiment, the (e) providing a representative healthy diet for an individual may include, in the step (c) setting a representative healthy diet for the corresponding group, the preference diet is set for each group. The representative healthy diet corresponding to the preferred diet group to which the individual belongs may be read by referring to the representative healthy diet of the corresponding group.

A method for recommending a diet based on personal preference according to another embodiment further includes: (f) calculating a difference between the input diet information of the individual and a representative healthy diet of the group to which the individual belongs to induce diet improvement; can do. In addition, the (f) inducing dietary improvement may include: (f1) calculating a difference between the input dietary information of the individual and the provided representative healthy diet for the individual; And (f2) selecting a diet in which the deviation from the food elements constituting the calculated difference is minimized from a preset recommended diet list.

The method of recommending a diet based on personal preference according to another embodiment may further include (g) updating the dietary information of the individual according to improvement in diet.

Further, the following provides a computer-readable recording medium in which a program for executing the above-described personal preference-based dietary composition method and personal preference-based dietary recommendation method on a computer is recorded.

In order to solve the above technical problem, a diet recommendation device based on personal preference according to another embodiment of the present invention includes a health index calculated from medical information or a health index calculated from food intake records and statistical health information from multiple people. An input unit for receiving an index; A memory for dividing an individual's preferred dietary group based on the preferred food using the food intake record and the health index and storing a program constituting a representative healthy diet of the group; And at least one processor for driving the program; wherein the program stored in the memory extracts dietary information from the food intake record and clusters the extracted dietary information to generate at least one preferred dietary group, and And a command for setting a representative healthy diet of the group by calculating an average of the dietary information for each of the preferred dietary groups using the health index as a weight.

In the apparatus for recommending a diet based on personal preference according to another embodiment, the program stored in the memory receives the diet information of the individual and determines a preferred diet group to which the individual belongs by referring to the at least one preference diet group, and the A representative healthy diet for an individual is provided, and a command for inducing diet improvement by calculating a difference between the input diet information of the individual and the representative healthy diet of the group to which the individual belongs may be further included.

In the apparatus for recommending a diet based on personal preference according to another embodiment, the program stored in the memory calculates a health index derived from medical information or a health index corresponding to a statistical calculation standard from statistical health information of a plurality of people, and , By receiving a list of preferred dietary groups that match and store dietary information for each person and the preferred dietary group to which they belong, map the preferred dietary group one-to-one with a representative healthy diet, and for each individual person data belonging to the preferred dietary group. The health index corresponding to the dietary information may be multiplied by the dietary information as a weight, summing all of them, and dividing it by the sum of the health indexes, thereby deriving a representative healthy diet of the corresponding group.

Embodiments of the present invention classify a preferred diet group from dietary information collected from various people and set a representative healthy diet for each group, thereby providing a desirable diet that is a criterion for objectively determining the level of health while being based on personal preference. It can be configured, and the preferred diet group to which the individual belongs is determined from the individual's dietary information, and a recommended diet is provided by using the difference from the representative healthy diet of the group. Optimized diet improvements can be achieved.

1A is a diagram illustrating a method of configuring a representative healthy diet according to personal preference using dietary information proposed by embodiments of the present invention.
1B is a diagram for explaining a method of recommending a diet according to personal preference using the dietary configuration of FIG. 1A proposed by embodiments of the present invention.
2 is a diagram for explaining the relationship between an example of a statistical health index of intestinal microbes and a medical index.
3 is a flowchart illustrating a method of constructing a diet based on personal preference according to an embodiment of the present invention.
4 is a diagram showing an implementation example of a dietary configuration and recommendation method adopted by embodiments of the present invention.
5 to 8 are diagrams showing in more detail the process of constructing a diet based on personal preference in FIG. 4.
9 is a flowchart illustrating a method of recommending a diet based on personal preference according to another embodiment of the present invention.
10 to 12 are diagrams illustrating a process of searching for and recommending a diet based on personal preference in FIG. 4 in more detail.
13 is a block diagram illustrating an apparatus for recommending a diet based on personal preference according to another embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, in the following description and the accompanying drawings, detailed descriptions of known functions or configurations that may obscure the subject matter of the present invention will be omitted. In addition, throughout the specification, "including" a certain component means that other components may be further included, rather than excluding other components unless specifically stated to the contrary.

The terms used in the present invention are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present application, terms such as "comprise" or "comprise" are intended to designate the existence of a set feature, number, step, action, component, part, or a combination thereof, but one or more other features or It is to be understood that the presence or addition of numbers, steps, actions, components, parts, or combinations thereof does not preclude the possibility of preliminary exclusion.

Unless otherwise defined, all terms, including technical and scientific terms, used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning of the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in this application. .

1A is a diagram for explaining a basic idea of constructing a representative healthy diet according to individual preferences using dietary information proposed by embodiments of the present invention.

Referring to FIG. 1A, embodiments of the present invention form a cluster of people's preferred diet by receiving dietary information from a plurality of people and classifying the information through a clustering technique. As illustrated, it is possible to create a'preferred dietary group' composed of people with similar diet preferences from information on food/diet consumed by each of a plurality of people.

Here, the dietary information can be obtained by quantifying the response details of the Food Frequency Questionnaire (FFQ), and such a food intake frequency survey paper is, for example, the'Korean Genome Epidemiological Survey Project' It can be a'Health Nutrition Survey' survey. There are differences in questionnaire items depending on the survey, but in general,'intake frequency' (e.g., once a month, 2-3 times a week, 1-2 times a week, etc.) for diet (e.g., rice) Quantity' (for example, 1/2 air, 1 air, 1 air and a half) is queried.

Dietary information can also be obtained through other methods of collecting individual dietary information. For example, it may be obtained based on various information such as a personal food delivery request details, a restaurant card payment details, and a food purchase details. The information obtained in this process may be converted into information obtained through the food intake frequency survey paper and used.

In the embodiments of the present invention, the diet information is information obtained by quantifying a questionnaire, and may be set as an array or a vector formed by the product of the'intake frequency' and the'intake amount' of each diet. . For example, it may be expressed as a number such as'Bob: 10','Ramen: 5', etc. and included in an array or vector. Since the exemplified'Korean Genome Epidemiology Research Project' survey contains 102 meals, an individual's diet can be expressed as a 102-dimensional array. The dimension of dietary information can be changed depending on the type of survey paper.

Now, the preferred diet group representative value 110 can be calculated from the diet information, and if the assumption that the calculated representative value is a healthy expression capable of representing the group is correct, the calculated representative value 110 It can be seen that it can be set as the'representative healthy diet' of the group. The problem will be a method of calculating the representative value of objective evidence that can determine the representative healthy diet of the group, which can be configured in various ways depending on the goal of the diet to be provided.

The concept of'health index' used for this is introduced. 'Health index' is an index representing individual health, and a'representative healthy diet' (representative value) can be defined using various types of health index. First, general health indicators such as blood sugar, blood pressure, liver level, and BMI can be used as health indexes. In this case, it can be named as'Healthy Diet for Reduced Blood Sugar','Healthy Diet for Reduced Blood Pressure', and'Health Diet for Improved Liver Value' according to each health index. Second, considering the high diversity of intestinal microbes in healthy people, the diversity of intestinal microbes can be used as a health index. In this case, the representative healthy diet can be called the'representative diet for increasing the diversity of intestinal microbes'. Third, by comparing the composition of intestinal microbes in the general population, the average intestinal microbes can be considered as an indicator of statistical health. Therefore, clustering is performed using the composition of intestinal microorganisms, and the distance to the centroid of each cluster is used as a health index. In this case, the representative healthy diet can be named as the'representative diet for the composition of healthy intestinal microbes'. In addition, it is expected that various health indices can be used to propose a representative healthy diet that meets the purpose. In the embodiments of the present invention described below, a method of determining such a representative value by using the correlation between intestinal microbes and health indicators is proposed, and will be introduced in more detail later in FIG. 7.

On the other hand, the labeling of Korean food and Western food illustrated in FIG. 1A is a name given for convenience by analyzing a preferred diet (eg, rice or bread) of a clustered group using dietary information. In other words, when individuals belonging to a specific cluster rarely eat bread or pizza, whereas individuals belonging to other clusters eat a lot of bread, pizza, etc., each can be classified into Korean food preference and Western food preference. Such labeling is indicated for convenience of explanation, and in the actual clustering process, classification will be made according to the similarity of dietary information.

1B is a view for explaining an idea of recommending a diet according to personal preference using the dietary configuration of FIG. 1A proposed by embodiments of the present invention.

Using the dietary information (or FFQ) 210 of a specific individual, the'preferred dietary group' to which the person belongs is compared with the'preferred dietary groups', and the'representative healthy diet' of this'preferred dietary group' (described above). The difference between the group's representative value) and the individual's dietary information (210) is calculated. Then, recommend a diet that can reduce these differences.

Referring to FIG. 1B, after comparing the dietary information 210 of a given individual with the representative healthy diet (representative value) of the corresponding group, it can be seen that the recommendation of the diet is made in a direction in which this difference is minimized. For example, if a guideline “Eat more rice and reduce meat” is suggested as a dietary recommendation, an appropriate diet may be provided in connection with a food provider associated therewith.

2 is a diagram for explaining the relationship between an example of a statistical health index of intestinal microbes and a medical index. Statistical health can be predicted using the similarity of intestinal microbes, and these predicted values reflect health-related medical information. Accordingly, intestinal and human health can be inferred from information on a number of human intestinal microbes, and using this, a representative healthy diet based on intestinal microbes can be constructed and recommended to users.

Hereinafter, the basic idea of the present invention proposed through FIGS. 1A and 1B and FIG. 2 will be described in more detail with reference to the drawings.

3 is a flowchart illustrating a method of configuring a diet based on personal preference according to an embodiment of the present invention, and each process may be performed through a diet configuration apparatus having at least one processor. At this time, each process is implemented as a program including a series of instructions and stored in a memory, and a defined operation may be performed by driving such a program.

In step S310, the dietary construction device extracts dietary information from food intake records input from a plurality of people. As described above, diet information is information obtained by quantifying a questionnaire input from a user, and can be set as an array or a vector consisting of the product of the'intake frequency' and the'intake amount' of each diet. I can.

In step S320, the dietary construction device generates at least one preferred dietary group by clustering the dietary information extracted in step S310. It is possible to collect arrays or vectors of dietary information for a large number of people, and perform clustering using the similarity between them. In terms of implementation, various methods such as Hierachical clustering, K-means clustering, and DBSCAN can be used, and are not limited to a specific clustering technique.

In step S330, the dietary apparatus includes a health index calculated from medical information such as blood sugar, blood pressure, liver value, and BMI, or a health index calculated from statistical health of a plurality of people as a weight, and the diet for each preferred diet group. By calculating the average of the information, it is set as the representative healthy diet for the group. Since individual dietary information is expressed as an array or vector representing the intake for each diet, the average diet vector can be calculated by weighting the individual's health index to the diet vector of normal people (individuals belonging to the same group) with the same preference diet. I can. The result will be a diet vector that well expresses the diet of healthy people, and this can be defined as a'representative healthy diet'. At this time, the reason why the statistical average value of diet information (for example, FFQ) can be defined as a representative health formula is because the diet information is given an individual's health index as a weight.

As briefly introduced above, the health indicator may be implemented using various indicators as indicators representing health. According to recent studies, it has been found that the intestinal microbes are closely related to the health of individuals. In the embodiments of the present invention, the health index is determined by using'diversity of intestinal microbes' or'similarity with standard intestinal microbes Was derived. If these indicators are used, the representative healthy diet could be more accurately named as the'representative healthy diet for promoting health of intestinal microbes'. Meanwhile, other health indicators (eg, glycemic index) may be used as health indicators. In the case of the glycemic index, a formula can be constructed such as'a representative healthy diet for improving diabetes'.

FIG. 4 is a diagram showing an overall implementation example of a dietary composition and recommendation method adopted by embodiments of the present invention, and largely, a representative healthy diet configuration 410 customized for personal preference and a representative healthy diet search and recommendation 420 are shown. It consists of three systems.

The former 410 is configured to be performed once for an investigation target (eg, all Koreans). The input for the personal preference customized representative healthy diet composition (410) system can be the Korean FFQ database and the intestinal microbial database, and as a result, the FFQ cluster prediction model (426) and the individual preference customized representative healthy diet set { x} is constructed. A series of processes is performed once, but according to the difference in the method of predicting health and increasing data collected later, another type of personalized representative health diet set {x} may be formed.

The latter 420 may make dietary recommendations for each individual using a system configured through the former 410. First, for each individual diet recommendation, an individual FFQ is given as an input, and a representative health diet tailored to personal preference (x) is output as a result value. Next, by using the diet recommendation module 428 to recommend a desirable diet (menu) from the difference with the representative healthy diet, a recommended diet is derived. The recommended diet is derived in a way that reduces the difference between the individual's FFQ vector and the personalized representative healthy diet (x), which can be compared with the diet database of each food company. Furthermore, it is possible to continuously monitor an individual's diet through the diet monitoring function and update 429 diet recommendations accordingly. That is, as the diet is continuously monitored in the future, the individual's FFQ vector is updated and other diets can be continuously provided.

5 to 8 are diagrams showing in more detail the process of constructing a diet based on personal preference in FIG. 4, wherein the same member numbers refer to the same configuration.

Referring to FIG. 5, a process of extracting dietary information from food intake records input from a plurality of people is performed according to an operation of the following preprocessing module 411. First, the total intake amount is calculated from the product of the intake frequency and intake amount for each food according to the food intake record, and the calculated total intake amount is normalized based on the maximum intake amount for each food. According to the illustrated FIG. 5, it can be normalized by dividing the total intake for each food by the maximum intake. Now, an array or vector including the total intake amount normalized for each food intake of each person is generated and set as dietary information. In the'List of FFQ ventors' of FIG. 5, this expression vector is illustrated.

Next, FIG. 6 shows a process of generating at least one preferred dietary group by clustering the extracted dietary information. The FFQ clustering module 412 classifies at least one or more preferred dietary groups by clustering the extracted dietary information according to the similarity of dietary information and assigning a cluster identifier to which dietary information belongs to each person's ingested food. . At this time, a predictor having a parameter as a result of the clustering is generated and transmitted to the FFQ cluster prediction module 426 for subsequent classification of an individual's preference group. As a result, in the list of FFQ clusters, diet information for each person and the cluster identifier to which he belongs are matched and stored.

7 is related to the intestinal health prediction module 414, which receives input into the intestinal microbial database and predicts the intestinal microbial-based health. Referring to FIG. 7, for example, the diversity of intestinal microorganisms may be set as a health index (health index), or the similarity of intestinal microorganisms may be set as a health index. At this time, the health index is preferably an intestinal health level calculated based on any one of the diversity of the intestinal microflora of a number of humans, the similarity with the standard intestinal microbiota, and combinations thereof.

FIG. 8 shows a process of setting a representative healthy diet for the group by calculating the average of diet information for each preferred diet group by using a health index calculated from statistical health information of a number of people as a weight. To this end, the FFQ cluster is mapped one-to-one with the representative healthy diet through the representative healthy diet mapper 413, and the representative healthy diet is weighted as a health index through the representative healthy diet prediction module 415 customized for personal preference. Configure. This representative healthy diet may be calculated as in Equation 1 below.

That is, the health index derived from medical information or the health index corresponding to the statistical calculation standard is calculated from the statistical health information of a number of people, and the diet information for each person and the preferred diet group to which he belongs are matched and stored. A list is received and the preferred diet group is mapped one-to-one with the representative healthy diet, and each diet information and the corresponding health index are multiplied with diet information as a weight for individual person data belonging to the preferred diet group, and all are summed. By dividing by the sum of the health indices, a representative healthy diet for the group can be derived. Through the above series of processes, it is possible to obtain a representative healthy diet set tailored to individual preferences.

In the above, the process of constructing a diet based on personal preference was examined. Hereinafter, a process of retrieving individual dietary information using the previously constructed diet and recommending a desirable diet is suggested.

9 is a flowchart illustrating a method of recommending a diet based on personal preference according to another embodiment of the present invention, and may be performed in succession to the processes S310 to S330 of FIG. 3. Accordingly, in order to avoid duplication of description, the description up to step S330 will be omitted.

In step S340, the dietary recommendation device receives the individual's dietary information and determines a preferred dietary group to which the individual belongs by referring to one or more preferred dietary groups. Previously, since at least one preferred dietary group has been generated by clustering in step S320, it is possible to determine which group the newly inputted dietary information of the individual belongs to.

In step S350, the diet recommendation device provides a representative healthy diet for the individual. Since the preferred dietary group to which the individual belongs is determined in step S340, a representative healthy diet corresponding to the group may be selected from the representative healthy diets for each preferred diet group set in step S330. The representative healthy diet selected in this way has clues about the desirable eating habits or diet that a given individual should aim for.

Then, in step S360, the dietary recommendation device may induce diet improvement by calculating a difference between the input dietary information of the individual and the representative healthy diet of the group to which the individual belongs. In other words, it helps to set a goal that can reduce the difference by finding the difference from the representative healthy diet of the preferred diet group to which the individual belongs using dietary information of a given individual.

Furthermore, in step S370, the dietary recommendation device may update the dietary information of the individual according to dietary improvement. In this process, it is possible to track and manage the improvement of an individual's eating habits through continuous monitoring.

10 to 12 are diagrams illustrating a process of searching for and recommending a diet based on personal preference in FIG. 4 in more detail, and the same member numbers denote the same configuration.

In FIG. 10, the same pre-processing module 411 and a representative healthy diet mapper 413 as described above through the dietary configuration based on personal preference (410 in FIG. 4) are utilized. If the formula based on personal preference is performed for a plurality of people, the difference is that it is performed for a given individual in FIG. 10, but it is natural that the same operation processing is required for analysis and determination of personal data.

That is, in the preprocessing module 411, the dietary information of a number of people (diet composition process) or individual dietary information (dietary recommendation process) calculates the total intake from the product of the intake frequency and intake amount for each food according to the food intake record. The calculated total intake is normalized based on the maximum intake for each food, and can be set by generating an array or vector including the normalized total intake for each person's intake. .

Next, the FFQ cluster prediction model 426 receives the individual's diet information and determines a preferred diet group to which the individual belongs by referring to one or more preferred diet groups. In this process, an individual's dietary information is classified according to a clustering method for generating at least one preference dietary group previously performed, so that the group most similar to the individual's preference diet is selected.

Subsequently, the representative healthy diet mapper 413 provides a representative healthy diet for a given individual. To this end, a representative healthy diet corresponding to the preferred diet group to which the individual belongs is read by referring to the representative healthy diet of the corresponding group previously set for each preferred diet group.

FIG. 11 is a diagram to introduce a process of recommending a diet, inducing diet improvement by calculating a difference between inputted dietary information of an individual and a representative healthy diet of a group to which the individual belongs.

More specifically, the FFQ comparison module 427 calculates a difference between the previously inputted individual's dietary information and the provided representative healthy diet for the individual. Then, the diet recommendation module 428 selects a diet in which the deviation from the factor foods constituting the difference calculated through the FFQ comparison module 427 is minimized from a preset recommended diet list. To this end, various matching algorithms for comparing the calculated difference and a preset diet recommendation list may be used.

As described above, the difference between the representative healthy diet and the individual's diet may be calculated as a vector for each diet. For example {rice: +2, ramen: -2, pork: +2, fish: +2… A'difference vector' such as} can be calculated. From this information, there are many ways to recommend a diet, for example pork could be pork cutlet or pork belly. Such dietary recommendations may be performed by referring to a database owned by a third party. For example, a food company would have to compose a product that corresponds to the above recommended diet among its own products, and a delivery company may need a database consisting of the diets of affiliated stores.

12 is a diagram for post-management after a diet recommendation, and shows a process in which the FFQ vector update module 429 updates the dietary information of the individual according to dietary improvement.

When it comes to diet composition and recommendations, the most desirable direction is for individuals to continually fill in their diet. In this case, the individual's FFQ will be updated in real time, so it is possible to monitor how close or far each individual's diet is from the representative healthy diet. For example, it is expected that the embodiment of FIG. 12 can be effectively applied to individual diet management when linked with technology for continuously monitoring an individual's diet using a mobile device, a smart speaker, or other methods.

FIG. 13 is a block diagram illustrating a diet recommendation device 40 based on personal preference according to another embodiment of the present invention, and a series of processes of FIGS. 3 and 9 are reconstructed from a hardware perspective. Therefore, here, in order to avoid duplication of description, functions of each component will be outlined from the viewpoint of the device.

The input unit 10 is configured to receive a health index calculated from medical information or a health index calculated from food intake records and statistical health information from multiple people.

The memory 20 stores a program for classifying an individual's preferred dietary group based on the preferred food using the food intake record and the health index and configuring a representative healthy diet of the group, and at least one processor 30 Run the program using. At this time, the program stored in the memory extracts dietary information from the food intake record, clusters the extracted dietary information to generate at least one preferred dietary group, and uses the health index as a weight for each preferred dietary group. And a command for setting the average of the dietary information as a representative healthy diet for the group. In this case, the health index is preferably a health level value calculated based on any one of a health level based on medical information, a diversity of intestinal microbes of a number of people, similarity with a standard intestinal microbiota, and a combination thereof. Do.

The program stored in the memory 20 receives an individual's dietary information, determines a preferred dietary group to which the individual belongs, by referring to the at least one preference dietary group, provides a representative healthy diet for the individual, and It may further include a command for inducing diet improvement by calculating a difference between the dietary information of the individual and the representative healthy diet of the group to which the individual belongs.

In addition, the program stored in the memory 20 calculates a health index derived from medical information or a health index corresponding to a statistical calculation standard from statistical health information of a plurality of people, and diet information for each person and a preferred diet to which they belong. Receive a list of preferred dietary groups that match and store groups, map the preferred dietary group to a representative healthy diet one-to-one, and weight the health index corresponding to each dietary information with respect to individual person data belonging to the preferred dietary group. It may include an instruction for deriving a representative healthy diet for the group by multiplying it with dietary information and summing it up by the sum of the health index.

According to the above-described embodiments of the present invention, by classifying a preferred diet group from dietary information collected from various people and setting a representative healthy diet for each group, the standard for objectively determining the level of health while based on personal preference is It is possible to compose a desirable diet to be used, and by determining the preferred diet group to which the individual belongs from the diet information of the individual and providing a recommended diet using the difference from the representative healthy diet of the group, it is not a uniform and uniform diet recommendation. It is possible to achieve dietary improvement that is optimized for individual eating habits.

Meanwhile, the embodiments of the present invention can be implemented as computer-readable codes on a computer-readable recording medium. The computer-readable recording medium includes all types of recording devices storing data that can be read by a computer system.

Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. Further, the computer-readable recording medium is distributed over a computer system connected by a network, so that the computer-readable code can be stored and executed in a distributed manner. In addition, functional programs, codes, and code segments for implementing the present invention can be easily inferred by programmers in the technical field to which the present invention belongs.

In the above, the present invention has been looked at around the various embodiments. Those of ordinary skill in the art to which the present invention pertains will be able to understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered from an illustrative point of view rather than a limiting point of view. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope equivalent thereto should be construed as being included in the present invention.

40: Dietary recommendation device based on personal preference
10: input
20: memory
30: processor

Claims (18)

(a) extracting dietary information from food intake records input from a plurality of people;
(b) clustering the extracted dietary information to generate at least one preferred dietary group; And
(c) setting a representative healthy diet for the group by calculating the average of the diet information for each preferred diet group by using the health index calculated from medical information or the health index calculated from the statistical health of a number of people as a weight. Including; method of constructing a diet based on personal preference. The method of claim 1,
The step (a),
(a1) calculating a total intake amount from the product of the intake frequency and intake amount by food according to the food intake record;
(a2) normalizing the calculated total intake based on the maximum intake for each food; And
(a3) generating an array or vector including the total intake normalized for each food intake for each food and setting it as dietary information; including, personal preference-based diet construction method . The method of claim 1,
The step (b),
(b1) clustering the extracted dietary information according to similarity of dietary information; And
(b2) classifying at least one preferred dietary group by assigning a cluster identifier to which dietary information belongs with respect to each person's ingested food; including, personal preference-based diet construction method. The method of claim 1,
The step (c),
(c1) calculating a health index derived from medical information or a health index corresponding to a statistical calculation criterion from statistical health information of a plurality of people;
(c2) receiving a list of preferred dietary groups that match and store diet information for each person and the preferred diet group to which the person belongs, and mapping the preferred diet group to a representative healthy diet one-to-one; And
(c3) Regarding individual person data belonging to the preferred diet group, each diet information and the corresponding health index are multiplied by diet information as a weight, and all are summed and divided by the sum of the health index, thereby representing the representative health of the group. Deriving a diet; containing, personal preference-based diet construction method. The method of claim 1,
The health index is,
A method of constructing a diet based on personal preference, which is a measure of intestinal health calculated based on any one of a health level based on medical information, the diversity of a large number of human gut microbiota, a similarity to a standard gut microbiota, and a combination thereof. (a) extracting dietary information from food intake records input from a plurality of people;
(b) clustering the extracted dietary information to generate at least one preferred dietary group, wherein the list of preferred dietary groups matches and stores dietary information for each person and a preferred dietary group to which they belong;
(c) A health index calculated from medical information or a health index calculated from statistical health information of a number of people is used as a weight and the average of the diet information is calculated for each of the preferred diet groups, thereby setting it as the representative healthy diet for the group. step;
(d) receiving an individual's dietary information and determining a preferred dietary group to which the individual belongs by referring to the at least one preferred dietary group; And
(e) providing a representative healthy diet for the individual; personal preference-based diet recommendation method comprising. The method of claim 6,
The dietary information of the plurality of people or individuals,
The total intake is calculated from the product of the intake frequency and intake by food according to the food intake record, the calculated total intake is normalized based on the maximum intake by food, and the normalized for each person's intake food. A diet recommendation method based on personal preference, which is set by creating an array or vector containing the total intake for each food. The method of claim 6,
The step (c),
(c1) calculating a health index derived from medical information or a health index corresponding to a statistical calculation criterion from statistical health information of a plurality of people;
(c2) receiving a list of preferred dietary groups that match and store diet information for each person and the preferred diet group to which the person belongs, and mapping the preferred diet group to a representative healthy diet one-to-one; And
(c3) Regarding individual person data belonging to the preferred diet group, each diet information and the corresponding health index are multiplied by diet information as a weight, and all are summed and divided by the sum of the health index, thereby representing the representative health of the group. Deriving a diet; containing, personal preference-based diet recommendation method. The method of claim 6,
The health index is,
A diet recommendation method based on personal preference, which is a measure of intestinal health calculated based on any one of a health level based on medical information, a diversity of intestinal microflora of a large number of humans, similarity to a standard gut microbiota, and combinations thereof. The method of claim 6,
The step (d),
In the step (b), by classifying the dietary information of the individual according to a clustering method for generating at least one preferred dietary group, the group most similar to the individual's preference diet is selected. The method of claim 6,
The step (e),
In the step (c), a representative healthy diet corresponding to the preferred diet group to which the individual belongs is read by referring to the representative healthy diet of the corresponding group set for each preferred diet group. The method of claim 6,
(f) calculating a difference between the input dietary information of the individual and the representative healthy diet of the group to which the individual belongs to induce diet improvement; further comprising, a diet recommendation method based on personal preference. The method of claim 12,
The step (f),
(f1) calculating a difference between the input diet information of the individual and the provided representative healthy diet for the individual; And
(f2) selecting a diet from a preset recommended diet list from a list of recommended diets that minimizes deviations from food elements constituting the calculated difference. Diet recommendation method based on personal preference. The method of claim 12,
(g) updating the dietary information of the individual according to dietary improvement; further comprising, a diet recommendation method based on personal preference. A computer-readable recording medium storing a program for executing the method of claim 1 on a computer. An input unit for receiving a health index calculated from medical information or a health index calculated from food intake records and statistical health information from a plurality of people;
A memory for dividing an individual's preferred dietary group based on a preferred food using the food intake record and the health index and storing a program constituting a representative healthy diet of the group; And
At least one processor for driving the program; Including,
The program stored in the memory extracts dietary information from the food intake record, clusters the extracted dietary information to generate at least one preferred dietary group, and uses the health index as a weight to generate the dietary information for each preferred dietary group. It includes a command to set the representative healthy diet of the group by calculating the average of the information,
The health index is based on personal preference, which is a measure of intestinal health calculated based on any one of a health level based on medical information, a diversity of intestinal microbes of a number of people, similarity with a standard intestinal microbiota, and a combination thereof. Dietary recommendation device. The method of claim 16,
The program stored in the memory,
By receiving the individual's dietary information and referring to the one or more preferred dietary groups, the preferred dietary group to which the individual belongs is determined, a representative healthy diet is provided for the individual, and the input dietary information of the individual and the individual A diet recommendation device based on personal preference, further comprising a command for inducing diet improvement by calculating a difference between the group's representative healthy diet. The method of claim 16,
The program stored in the memory,
A health index derived from medical information or a health index corresponding to the statistical calculation standard is calculated from the statistical health information of a number of people, and a list of preferred diet groups that matches and stores diet information for each person and the preferred diet group to which they belong is created. It receives the input and maps the preferred diet group to the representative healthy diet one-to-one, and multiplies each diet information and the corresponding health index as a weight as a weight for individual person data belonging to the preferred diet group, and sums them all by multiplying it with the diet information. A diet recommendation device based on personal preference, including an instruction for deriving a representative healthy diet for the group by dividing it by the sum of.

Images