KR20190072292A - Apparatus and method for body growth prediction modeling - Google Patents

Abstract

Disclosed are a device and a method for modeling body growth prediction. The method receives body growth data for an arbitrary user; models a growth prediction model according to any one of the similarity of a user with the body growth data or the conditional probability of the body growth data of the user, based on the body growth data of the user and body growth data of a plurality of previously stored subjects to be compared; models the growth prediction model based on at least one of a plurality of kinds of parameters included in the body growth data; and predicts and outputs a body growth value corresponding to the user through the growth prediction model. The body growth data of a plurality of previously stored subjects to be compared is recorded in a time-series manner for each subject to be compared for an arbitrary period. In addition, in the arbitrary period corresponding to two or more subjects to be compared among the plurality of subjects to be compared, at least one of the range and the interval thereof may be different from each other, or the interval thereof may be partially overlapped. The present invention can improve accuracy of growth prediction result.

Description

[0001] APPARATUS AND METHOD FOR BODY GROWTH PREDICTION MODELING [0002]

The present invention relates to an apparatus and a method for modeling a body growth prediction that predicts a physical growth value based on fragment data for which an overall trend is not known.

Predictive analysis is a technique of statistical and data mining areas that extract information from data and use it to predict trends and behavior patterns. This predictive analysis can be applied to all areas of decision making based on the information obtained from the data. The core of predictive analysis is to predict unknown variables after understanding the relationship between variables. To this end, various approaches are used depending on the characteristics of the data and the objects to be predicted.

Predictive analysis is important to choose the right approach depending on the nature of the results you want to achieve. Statistical techniques such as regression analysis and machine learning techniques are being used as an approach that can be applied to predictive analysis. The prediction analysis process through the existing regression analysis focused on inducing the interactions between variables into mathematical equations. In addition, the prediction analysis process through the existing machine learning method is focused on classifying the data group through the learning algorithm without relying on mathematical equations like rule - based modeling. However, in general, the accuracy and usefulness of the prediction results are characterized by the analysis method and the background knowledge of the analyst. Therefore, the predictive analysis process should include a reasonable hypothesis, model generation (ie, modeling), and data analysis methods based on result validation to minimize the uncertainty of the predicted results. However, the conventional predictive analysis process was focused on improving the quality of the predicted results regardless of the level of the analyst, but did not focus on each analysis process and uncertainty of the original data and the predicted results. Uncertainty of the original data is mainly caused by missing data, lack of data and noise, and should be controlled during the preprocessing. However, the conventional prediction analysis process can not solve the uncertainty of the prediction result caused when the data is insufficient (that is, when the general tendency of the input data can not be known).

On the other hand, among the various fields requiring predictive analysis, social interest has recently been given to the body growth of adolescents. In particular, there is a growing demand from users who want to know the predicted results of when and how the key growth will grow. Researchers dealing with this discipline or technique focus on analyzing the relationship between body growth and genetic / environmental factors and explore the growth limit curve determined by genetic factors to analyze when growth is accelerating, And the environmental factors that can maximize the environmental impact.

As a method for predicting this growth potential, there is a method of using skeleton information and a method of using statistical data modeling. Using the skeleton information can lead to the most accurate prediction result, but it takes a lot of time and cost. And, in the method of using statistical data modeling, it is very important to analyze the trend of the sample with the biological pattern of the subject to analyze, and to apply the growth prediction model suitable for the biological and growth pattern. However, data such as youth physical growth data have difficulty classifying patterns because the overall trend is unknown. In other words, when the time range covered by the growth prediction model is exceeded, the uncertainty of the prediction model increases, and the reliability of the prediction result may be very low. In order to solve this problem, it is possible to handle the growth prediction using the whole student body growth data, but it is impossible to predict the body growth by reflecting the individual biological characteristics alone. In order to minimize the uncertainty of the prediction model, it is necessary to repeat the model generation and verification every time the new data is updated, even if it is assumed that the growth prediction model that is optimized for each pattern is classified by classifying similar growth patterns in the body growth data of all the students There is a limit to do.

Korean Patent Publication No. 10-2014-0045759 (entitled " Growth Management Service Method and System)

According to an aspect of the present invention, there is provided a method for predicting a physical body growth value of an arbitrary user using time-series fragment data whose overall trend is unknown, such as student body growth data, The present invention provides a body growth prediction modeling apparatus and method for estimating a data distribution over time and searching for similar data and providing a data-oriented approach for linking the relationships among them.

In addition, one embodiment of the present invention provides a method and apparatus for performing a visual analysis process supporting an interactive data search function so that an analyst can easily organize and minimize uncertainties in a process of modeling a body growth prediction model. A modeling device and a method thereof.

It should be understood, however, that the technical scope of the present invention is not limited to the above-described technical problems, and other technical problems may exist.

According to an aspect of the present invention, there is provided a body growth prediction modeling apparatus comprising: a database storing body growth data for a plurality of comparison subjects; An input module for receiving body growth data for an arbitrary user; A memory in which a growth prediction modeling program is stored; And a processor for executing a program stored in the memory, wherein, in accordance with the execution of the growth prediction modeling program, the processor calculates, based on the body growth data of the inputted user and the body growth data of a plurality of pre- A model of a growth prediction model according to any one of the similarity degree with the user's body growth data or the conditional probability with respect to the body growth data of the user based on at least one of a plurality of kinds of parameters included in the body growth data, The body growth model is modeled, and the body growth value corresponding to the user is predicted and output through the growth prediction model. At this time, the body growth data recorded in a time series for an arbitrary period is stored for each comparison subject in the database. The arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects may be different from each other in at least one of the range and the interval, or the interval may partially overlap.

According to another aspect of the present invention, there is provided a body growth prediction modeling method including: receiving body growth data for an arbitrary user; A growth prediction model based on the user's body growth data and a plurality of pre-stored body growth data for each of the comparison subjects based on the similarity of the user with the body growth data or the conditional probability of the user's body growth data, Modeling; And estimating and outputting a body growth value corresponding to the user through the growth prediction model. At this time, modeling the growth prediction model models the growth prediction model based on at least one of a plurality of kinds of parameters included in the body growth data. In addition, the plurality of body growth data for each of the plurality of comparison subjects to be compared is recorded in a time-series manner for each of the comparison subjects for an arbitrary period, and the arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects, At least one of them is different from each other, or a part thereof is partially overlapped.

According to any one of the above-mentioned objects of the present invention, by performing growth prediction modeling using a data-centric approach connecting similar data at the present time point, time-series fragment data whose global trend is unknown can be used as reference data The uncertainty of the predicted results is greatly reduced compared to the conventional simple machine learning or regression analysis.

Further, according to any one of the tasks of the present invention, it is possible to provide a visual analytics process capable of supporting the body growth prediction modeling of any object so that the analyst can identify the relationship between the growth prediction variables, The validity of the model can be directly confirmed, and the accuracy of the growth prediction result can be improved.

1 is a configuration diagram of a body growth prediction modeling apparatus according to an embodiment of the present invention.
FIGS. 2 and 3 are examples of reference data applied to growth prediction modeling according to an embodiment of the present invention.
4 is a diagram for explaining a growth prediction model based on a distribution model according to an embodiment of the present invention.
5 is a diagram for explaining a growth prediction model based on a Bayesian inference model according to an embodiment of the present invention.
FIG. 6 is a graph showing a result of verification of accuracy of a growth prediction result according to a body growth prediction modeling method according to an embodiment of the present invention.
7 is an example of a visual analysis user interface supporting growth prediction modeling according to an embodiment of the present invention.
Figure 8 is an example of a visual analytic user interface that supports analysis of growth prediction results for a user in accordance with an embodiment of the present invention.
9 is a flowchart for explaining a body growth prediction modeling method according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings, which will be readily apparent to those skilled in the art. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description in the drawings are omitted, and like parts are denoted by similar reference numerals throughout the specification. In the following description with reference to the drawings, the same reference numerals will be used to designate the same names, and the reference numerals are merely for convenience of description, and the concepts, features, and functions Or the effect is not limited to interpretation.

Throughout the specification, when a part is referred to as being "connected" to another part, it includes not only "directly connected" but also "electrically connected" with another part in between . Also, when a component is referred to as "comprising ", it is understood that it may include other components as well as other components, But do not preclude the presence or addition of a feature, a number, a step, an operation, an element, a component, or a combination thereof.

Herein, the term " part " or " module " means a unit realized by hardware or software, a unit realized by using both, and a unit realized by using two or more hardware Or two or more units may be realized by one hardware.

1 is a configuration diagram of a body growth prediction modeling apparatus according to an embodiment of the present invention.

Hereinafter, an object to which an actual body growth value is to be predicted will be referred to as a "user ", and an object to which body growth data as reference data necessary for predicting body growth has been collected will be referred to as a" subject to be compared ".

The body growth prediction modeling apparatus 100 according to an exemplary embodiment of the present invention may be configured to estimate a body growth of a user using body growth data of a plurality of comparison subjects collected in advance, And outputs the result of predicting the body growth value corresponding to the user.

1, the body growth prediction modeling apparatus 100 includes an input module 110, a memory 120, a processor 130, and a database 140. [

The input module 110 receives input data for predicting growth of any user and transmits the input data to the processor 130. At this time, the input data may be body growth data, and may include values of past or present body components of the user (hereinafter referred to as "body component data").

Further, the input data may include at least one parameter of a plurality of kinds of parameters included in the body growth data. The parameters of such body growth data are used as a condition for modeling the body growth prediction model.

For example, Table 1 below shows the types of multiple parameters included in the body growth data.

<Table 1>

As shown in Table 1, the body growth data includes body mass index (BMI), basal metabolic rate (BMR), body weight, skeletal muscle mass, body fat mass, body water mass, muscle mass, At least one of a mineral size, an inorganic number, a body fat percentage, an abdominal fat percentage, a neck circumference dimension, a chest circumference dimension, an abdominal circumference dimension, a right arm circumference dimension, an arm circumference dimension, a hip circumference dimension, a right thigh circumference dimension, , And 'user information' including user identification information (ID), gender, age, and data measurement time (date).

For reference, the parameter types of the body growth data in the embodiment of the present invention are not limited to the above Table 1, and various kinds of parameters related to the body components and growth can be added or changed. Also, in one embodiment of the present invention, 'user information' of body growth data is a concept that is commonly applied not only to users who want to predict body growth value but also to a comparison object.

In the database 140, a plurality of body growth data for individual subjects to be compared are stored. The body growth data of the plurality of comparison subjects are used as reference data for growth prediction modeling.

Specifically, in the database 140, the body growth data recorded in a time series for an arbitrary period is stored for each of the comparison subjects, and an arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects includes the range and the interval At least one of them is different from each other or a part thereof is partially overlapped.

FIGS. 2 and 3 are examples of reference data applied to growth prediction modeling according to an embodiment of the present invention.

For example, referring to FIG. 2 and FIG. 3, 29969 students between the ages of 7 and 18 are designated as the subjects to be compared. The body composition and the student information (Ie, user information). For these subjects, the number of measurements of body components is 2 to 17 times (average 6 times) per year. The measurement period ranges from 1 to 4 years (average 2 years), and the measurement interval is 3 to 24 give. Examples of recorded body growth data of the subjects to be compared as shown in FIGS. 2 and 3 are irregular in the number of data, interval, and period because of limited data collection environment.

Figure 2 shows the number of body growth data collected for each grade. 2 and 3, an entire school year is classified into a 12th grade, and classified into an elementary school corresponding to the first to sixth grade, a middle school corresponding to the 7th to 9th grade, and a high school corresponding to the 10th to 12th grade. In the example of FIG. 2, it can be seen that the variation of the data count per elementary school, middle school, and high school is very large due to the limit of the data collection environment. In junior high school, 69% of the data of elementary school students is reduced, and in high school, 94% of the data of elementary students is decreased.

FIG. 3 shows the number of data for each segment of a student whose measurement period of the body growth data is 3 years. For example, referring to FIG. 3, there are 1924 persons who recorded physical growth data for three years from the first grade to the third grade, and 161 persons recorded physical growth data for three years from the sixth grade to the eighth grade. In other words, the body growth data collected for each subject to be compared has a measurement period of 3 years, and even though the range itself is the same, the intervals may be different from each other or partially overlap each other. It is also possible that the range of the measurement period of the body growth data of the subject to be compared is measured in a time-series manner. Any comparator may record multiple body growth data continuously for 3 years, and the other comparator may record multiple body growth data continuously for 5 years.

For reference, the database 140 may be read and written according to the control of the processor 130 and may be included in a portion of the memory 120 or may be included in the body growth prediction modeling device 100 As shown in FIG.

In the memory 120, a body growth prediction modeling program for generating a body growth prediction model is stored. Also stored in memory 120 is a visual analysis program that further provides a visual analytics process that visually supports the information for setting the conditions necessary to model the body growth prediction model or analyzing growth prediction results have. For reference, the body growth prediction modeling program and the visual analysis program can operate in conjunction with each other as separate programs. It is also possible that a visual analysis program is included and operated as a subprogram of the body growth prediction modeling program.

This memory 120 is collectively referred to as a non-volatile storage device that keeps stored information even when power is not supplied, or a volatile storage device that requires power to maintain stored information.

The processor 130 executes the program stored in the memory 120, and can perform the following processing.

The processor 130 models the growth prediction model based on the body growth data of the input user and the body growth data of a plurality of previously stored comparison subjects, in accordance with the execution of the growth prediction modeling program. At this time, the processor 130 models a growth prediction model according to one of the similarity with the user's body growth data or the conditional probability with respect to the body growth data of the user, and at least one of a plurality of kinds of parameters included in the body growth data Model the body growth prediction model based on one. The processor 130 predicts and outputs the body growth value corresponding to the user through the modeled growth prediction model.

In addition, the processor 130 may provide a visual analytics user interface to support the modeling of the growth prediction model.

In addition, the processor 130 may provide a visual analysis user interface that supports analysis of predicted results when outputting a body growth value corresponding to a user.

Hereinafter, with reference to Figs. 4 to 8, each operation to be processed through the processor 130 will be described in detail.

First, with reference to FIGS. 4 and 5, a method of modeling a growth prediction model by the processor 130 will be described.

4 is a diagram for explaining a growth prediction model based on a distribution model according to an embodiment of the present invention. And FIG. 5 is a diagram for explaining a growth prediction model based on a Bayesian inference model according to an embodiment of the present invention.

The processor 130 may model the growth prediction model based on two techniques (i.e., similarity and conditional probability).

Referring to FIG. 4, a description will be given of how the processor 130 models a growth prediction model based on a distribution model.

One or more of the plurality of body growth factors has a characteristic that follows a normal distribution. For example, height is representative data that follows a normal distribution. Figure 4 shows that both the height distribution of n-th grade and the height distribution of n + t grade follow normal distribution. Accordingly, the following hypothesis can be applied to the distribution model.

Hypothesis 1: The height data of each grade follows a normal distribution.

Hypothesis 2: n is similar in graders A regular bunposang position grade _{n + t (age n + t} ) in the (Student _A) to (age _n).

Based on this, the processor 130 models the growth prediction model based on the distribution model when using the similarity with the user's body growth data. That is, processor 130 searches for the student (i.e., the comparator) most similar to the student to be predicted in the next step, in order to achieve the effect of actually tying similar data using the distribution model. At this time, any one of Euclidean distance method, Manhattan distance method, and cosine similarity method can be used as a distribution model for measuring similarity from data. However, the technique for measuring the degree of similarity is not limited thereto.

In one embodiment of the invention, it is illustrated that processor 130 performs growth prediction modeling using Euclidean distance techniques. For reference, the Euclidean distance can be calculated by the following equation (1).

&Quot; (1) &quot;

In Equation (1), d is the Euclidean distance between p and q, q _i is the physical constituent of the student to be predicted, and p _i is the physical constituent of the other student to be compared.

Algorithm 1 below is an example of an algorithm that processor 130 performs to model a body growth prediction model for any user.

<Algorithm 1>

First, n to calculate the grade (or age) standard scores (z-score) of A student (Student _A) (age _n) prediction (ie, users) in. Then, in the n + t grade (age _{n + t} ), search for candidates that are closest to the standard score of Student _A (ie, the student to be compared). Referring to FIG. 4, candidates are Student _B , Student _C , and Student _D. Then, the student B who has the closest value to the standard score of the student A is detected through the equation (1). As a result of this growth prediction modeling, the key value (ie, body growth value) of student A in n + t grade (age _{n + t} ) is replaced by the key value of student B

Thus, it has been described that the "age" and "sex" in the user information and the "key" in the body component of the parameters of the body growth data are selected as conditions of the growth prediction modeling. However, the number and types of parameters of the body growth data selected as conditions of the growth prediction modeling in the embodiment of the present invention are not limited. Meanwhile, according to an embodiment of the present invention, an analyst (a user performing a body growth prediction analysis) through a visual analysis user interface to be described with reference to FIG. 6 will be described with reference to various types of parameters, And can directly select the conditions necessary for the prediction analysis.

Referring to FIG. 5, a method for modeling a growth prediction model based on a Bayesian inference model will be described.

5 is a diagram for explaining a growth prediction model based on a Bayesian inference model according to an embodiment of the present invention.

The processor 130 models the growth prediction model based on the Bayesian inference model when using a conditional probability for the user's body growth data. At this time, the probability that the event A occurs when the event B occurs (i.e., the conditional probability) can be expressed by the following equation (2).

&Quot; (2) &quot;

Bayesian inference is a statistical approach to infer posterior probabilities based on prior probabilities for objects to be inferred and likelihood values that are observations at the current point in time.

The relationship between the prior probability and the posterior probability according to the Bayesian inference technique can be summarized as Equation 3 below.

&Quot; (3) &quot;

At this time, P (H) is a prior probability and P (E) is a likelihood value which is an observation value.

The following hypothesis can be applied to the body growth prediction model based on Bayes' inference.

Hypothesis: The student 's physical growth trend follows the average growth trend of a group of students with similar physical characteristics at the present time.

Based on this, the following growth prediction modeling is performed. First, body growth data of a plurality of comparison subjects are classified into a plurality of groups based on age. At this time, the plurality of groups can be classified based on the characteristics of the predetermined (or selected by the analyst) parameter, the relationship value between the selected parameters, and the like. Then, a prior probability which is a probability that the user's body growth data is included in an arbitrary group is calculated, and a posterior probability is calculated based on the calculated prior probability and the likelihood value of the current time to estimate a body growth state path . Accordingly, the body growth value of the user is predicted based on at least one group of body growth data included in the estimated body growth state path, and the result is output as a result of the growth prediction modeling.

이다. 이때, 1 학년(age₁)에서 임의의 학생(즉, 사용자)이 S_1,2로 분류될 확률은 P(E)=P(S_1,2)=0.4이다. 이러한 경우, P(E) 값은 S_1,2에 속한 학생 수를 1학년(age₁)에 속한 전체 학생 수로 나눈 값이다. 학생의 신체 성장 추세가 S_1,2에서 S_2,1로 변이될 확률인 H₁는 0.11이다. 그리고 학생의 신체 성장 추세가 S_1,2에서 S_2,2로 변이될 확률인 H₂는 0.5이다. 이에 따라, H₁에서 우도 함수 P(E|H)=0.17이며, H₂에서 우도 함수 P(E|H)=0.25이다.For example, in Figure 5, S represents the probability that a student will be included in any growth group

to be. In this case, the probability that an arbitrary student (ie, user) is classified as S _1,2 in the first year (age ₁ ) is P (E) = P (S _1,2 ) = 0.4. In this case, the value of P (E) is the number of students in S _1,2 divided by the total number of students in the first year (age ₁ ). H ₁ is the probability that the growing trend of the student body to be mutated in S _1,2 to S _2,1 is 0.11. And the probability that the growing trend of the student body to be mutated in S _1,2 to S _2,2 H ₂ is 0.5. Accordingly, in the likelihood function H ₁ P | and (E H) = 0.17, in H ₂ likelihood function P | a (E H) = 0.25.

The posterior probabilities P (H | E) for the prior probabilities H ₁ and H ₂ are calculated by the following equations (4) and (5), respectively.

&Quot; (4) &quot;

Equation (5)

Thus, the body growth state path for the student is included in the transition path from growth group S _1,2 to S _2,2 . In this manner, a plurality of groups included in the body growth status path of the student among the plurality of growth groups can be identified.

Meanwhile, FIG. 6 is a graph showing accuracy verification results of growth prediction results according to a body growth prediction modeling method according to an embodiment of the present invention.

Referring to FIG. 6, it is shown that 5767 students (25% of the total students) used the data for collecting body growth information for three years or more in accuracy verification.

At this time, the data measured in the first year are compared with the actual data for the next two years.

, MAPE (Mean Absolute Percentage Error):

, MdAPE (Median relative Absolute Percentage Error):

를 사용하였다. RMSE는 모델이 예측한 값과 실제 환경에서 관찰되는 값의 차이를 정량화한 것으로 정밀도를 표현하는데 적합하다. MAPE는 실제 값에 대한 상대 오차 비율을 나타낸다. 그리고 MdAPE는 RMSE와 MAPE와 달리 중간 값을 이용하기 때문에 특이 값에 의한 왜곡에 강인하다. 각 모델별 수식에서 f_t는 예측 값을 의미하며, o_t 는 실제 값을 의미한다. In order to evaluate the growth prediction model from various perspectives, the Root Mean Square Error (RMSE)

, Mean Absolute Percentage Error (MAPE):

, Median relative Absolute Percentage Error (MdAPE):

Were used. The RMSE is a quantification of the difference between the predicted value of the model and the observed value in the real environment, and is suitable for expressing the precision. MAPE represents the relative error ratio to the actual value. And, MdAPE is robust against distortion due to singular value because it uses median value, unlike RMSE and MAPE. In each model expression, f _t denotes the predicted value, and o _t denotes the actual value.

The error rate according to the growth prediction value using the distribution model and the Bayes reasoning model according to an embodiment of the present invention and the error rate according to the growth prediction value predicted through the linear regression model and the logarithm regression model Respectively.

As shown in FIG. 6, it can be seen that the growth prediction value through the growth prediction model modeled using the distribution model and the Bayes inference model according to an embodiment of the present invention is much higher.

Next, with reference to FIG. 7, a method of providing a visual analysis user interface that supports modeling of a growth prediction model by the processor 130 will be described.

7 is an example of a visual analysis user interface supporting growth prediction modeling according to an embodiment of the present invention.

For reference, a visual analysis user interface supporting growth prediction modeling as shown in FIG. 7 can be implemented as a web-based framework. The reason for applying visual analysis techniques to support growth prediction modeling is to recognize and minimize the uncertainties that may arise in the implementation of predictive models. In this case, the uncertainty factor can be divided into data loss, outliers, and variable selection independent of prediction model. Thus, through the visual analytic user interface, it is easy to understand growth prediction modeling conditions and processes, and analysts who lack statistical knowledge can easily use it. Various views and perspectives are provided to fully understand the data, and noise can be controlled in the data to improve predictive modeling performance. In addition, the uncertainty of the predictive model to be implemented can be quantified and easily recognized by analysts.

Referring to FIG. 7, the user interface for supporting growth prediction modeling may be composed of five interaction modules.

Specifically, a parameter selector for selecting at least one of a plurality of kinds of parameters included in the body growth data as shown in FIG. 7 (a) through a visual analysis user interface supporting modeling of a growth prediction model, a histogram selector for outputting a distribution for each body growth data as in (b), (c) and (d), a parallel coordinate view and a scatter plot for outputting the degree of correlation between parameters of the body growth data in the form of coordinates and provides an algorithm modeling module for modeling a growth prediction model to which a selected condition is applied, as shown in FIG. 7 (e).

At this time, the parallel coordinate view as shown in (c) of FIG. 7 can visualize the multivariate data according to the parallel coordinate plot (PCP) technique. PCP converts high dimensional specific geometric characteristics into 2D patterns and outputs them. On the other hand, different results can be derived by the order of axis alignment, axis rotation and scaling in PCP, so the analyst can perform growth prediction modeling after checking various conditions through parallel coordinate view. In addition, the scatter plot view as shown in FIG. 7 (d) can visualize the relationship between two variables using an orthogonal coordinate system.

To explain in more detail, it is assumed that an algorithm is implemented to predict the degree of key growth over time. To analyze the relationship between the data, the analyst can select the type of data to analyze in the parameter selector, (c) the parallel coordinate view on the visual analytical user interface and (d) the relationship between the data in the scatter view And output. For detailed analysis, a range of data to be analyzed can be filtered using a brush in the (b) histogram selector on the visual analysis user interface and (c) in the parallel coordinate view. At this time, the analyzer can remove the noise by using the brush function, and information about the removed noise can be outputted at the bottom of (c). The analyst can model the growth prediction algorithm by drag and drop between (a) and (e) on the visual analysis user interface. For example, as shown in FIG. 7 (e), a grade (grade) on the x-axis and a height (key) on the y-axis can be input and parameters (height, gender, protein) to be used in the next prediction model can be input. In this case, the growth prediction model predicts the height using three dimensions. Also, the degree of uncertainty of the growth prediction model can be quantified and visualized in the progress bar of (e) through the visual analytic user interface. That is, it can be understood that the growth prediction model modeled using the parameter 'height, gender, protein_mass' in FIG. 7 can account for 90.2% of the total data and has a degree of uncertainty of about 9.8%.

Next, referring to FIG. 8, a description will be given of a method of providing a visual analysis user interface that the processor 130 supports analysis of the body growth prediction result.

Figure 8 is an example of a visual analytic user interface that supports analysis of growth prediction results for a user in accordance with an embodiment of the present invention.

For reference, a visual analysis user interface supporting analysis of growth prediction results as shown in FIG. 8 can be implemented as a web-based framework. The visual analysis user interface for analyzing the body growth prediction result shown in FIG. 8 provides information that can comprehensively analyze the growth status of the adolescents, comparison with similar group of students, and growth prediction.

Referring to FIG. 8, the user interface for supporting the growth prediction result analysis may be composed of five interaction modules.

Specifically, a body state view, in which a student's physical state to be analyzed can be known as shown in FIG. 8 (a) through a visual analysis user interface supporting analysis of a growth prediction result, and (b) (Cluster view), which can identify students with similar body growth patterns, (c) Radar chart, which can analyze the body balance as in (c), and (d) A history tracker that can be analyzed, and a prediction view capable of analyzing growth prediction results as shown in (e).

Hereinafter, a body growth prediction modeling method according to an embodiment of the present invention will be described in detail with reference to FIG.

9 is a flowchart for explaining a body growth prediction modeling method according to an embodiment of the present invention.

First, body growth data for a user to be a body growth prediction target is input (S910).

Next, a growth prediction model according to any one of the similarity with the user's body growth data or the conditional probability with respect to the user's body growth data is modeled using a plurality of body growth data for each comparison subject collected and stored in advance (S920).

Wherein the plurality of previously stored body growth data for each of the comparison subjects is recorded in a time series for each of the comparison subjects in an arbitrary period and the arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects includes at least one of a range and an interval Are different from each other, or the intervals are partially overlapped.

Also, the body growth prediction model is modeled based on at least one of a plurality of kinds of parameters included in the body growth data. At this time, the method of modeling the growth prediction model using the similarity degree or conditional probability is the same as the method described above with reference to FIG. 4 and FIG. 5, so that redundant description will be omitted.

A visual analytic user interface can be provided to support such body growth prediction modeling. A parameter selector for selecting at least one of a plurality of types of parameters included in the body growth data through the visual analysis user interface, a histogram selector for outputting a distribution for each body growth data, a degree of correlation between parameters of the body growth data A parallel coordinate view and a scatter view in which the selected condition is applied, and an algorithm modeling module for modeling a growth prediction model to which the selected condition is applied.

Then, the body growth value corresponding to the user is predicted and output through the growth prediction model (S930).

At this time, it is possible to provide a visual analysis user interface that supports analysis of predicted body growth prediction values through the growth prediction model. Through such a visual analysis user interface, a physical state view for confirming the user's physical status information, a cluster view for identifying objects having similar body growth patterns to a user and a plurality of comparison subjects, A radar chart, a history tracker for analyzing past growth records, and a predictive view for analyzing user growth predictions.

The above-described body growth prediction modeling method according to an embodiment of the present invention may be implemented in the form of a recording medium including instructions executable by a computer such as a program module executed by a computer. Such a recording medium includes a computer-readable medium, which may be any available medium that can be accessed by a computer, and includes both volatile and non-volatile media, both removable and non-removable media. The computer-readable medium may also include computer storage media, which may be volatile and non-volatile, implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data , Both removable and non-removable media.

It will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the present invention as defined by the following claims. will be. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. For example, each component described as a single entity may be distributed and implemented, and components described as being distributed may also be implemented in a combined form.

Furthermore, while the methods and systems of the present invention have been described in terms of specific embodiments, some or all of those elements or operations may be implemented using a computer system having a general purpose hardware architecture.

It is intended that the present invention covers the modifications and variations of this invention provided they come within the scope of the appended claims and their equivalents. .

100: Body growth prediction modeling device
110: input module
120: Memory
130: Processor
140: Database

Claims (14)

In a body growth prediction modeling device,
A database storing a plurality of body growth data for each comparison subject;
An input module for receiving body growth data for an arbitrary user;
A memory in which a growth prediction modeling program is stored; And
And a processor for executing a program stored in the memory,
The processor is configured to perform a growth prediction modeling program based on the user's body growth data of the user and the body growth data of a plurality of previously stored comparison subjects, Modeling the body growth model based on at least one of a plurality of kinds of parameters included in the body growth data by modeling a growth prediction model according to any one of conditional probability for data, Predicts and outputs the body growth value corresponding to the user,
In the database, body growth data recorded in a time series for an arbitrary period is stored for each comparison subject,
Wherein the arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects has at least one of a range and an interval that are different from each other or a section is partially overlapped. The method according to claim 1,
The processor comprising:
Providing a visual analytics user interface that supports modeling of the growth prediction model,
A parameter selector for selecting at least one of a plurality of types of parameters included in the body growth data through the visual analysis user interface, a histogram selector for outputting a distribution for each body growth data, A parallel coordinate view and a scatter plot view outputting the degree of correlation in a coordinate form and an algorithm modeling module for modeling the growth prediction model to which the selected condition is applied. The method according to claim 1,
The processor comprising:
When the degree of similarity with the body growth data of the user is used, the growth prediction model is modeled based on a distribution model,
The body growth data of the comparison object most similar to the user is classified using the Euclidean distance technique, Manhattan distance technique, and cosine similarity technique as the distribution model And predicts a body growth value of the user based on the classified body growth data of the comparison subject. The method according to claim 1,
The processor comprising:
Modeling the growth prediction model based on a Bayesian inference model when the conditional probability of the user's body growth data is used,
Classifying body growth data of the plurality of comparison subjects into a plurality of groups based on age, calculating a prior probability that is a probability that the body growth data of the user is included in an arbitrary group, Estimating a body growth state path by calculating a posterior probability based on a likelihood value, estimating a body growth value of the user based on at least one group of body growth data included in the estimated body growth state path, , A model of body growth prediction. The method according to claim 1,
The body growth data may include:
Body mass index (BMI), basal metabolic rate (BMR), body weight, skeletal muscle mass, body fat mass, body water content, muscle mass, fat mass, protein level, mineral level, mineral level, Body component data including at least one of a fat percentage, a neck circumference dimension, a chest circumference dimension, an abdominal circumference dimension, a right arm circumference dimension, an left arm circumference dimension, a hip circumference dimension, a right thigh circumference dimension and a left thigh circumference dimension,
User identification information, gender, age, and time of data measurement. The method according to claim 1,
The processor comprising:
And selects at least one parameter having a normal distribution characteristic among a plurality of parameters included in the body growth data as a modeling reference of the growth prediction model. The method according to claim 1,
The processor comprising:
Providing a visual analytics user interface that supports analysis of predicted body growth predictions through the growth prediction model,
Through the visual analysis user interface,
A body state view capable of confirming the user's body state information, a cluster view capable of identifying objects having similar body growth patterns to the user and the plurality of comparison subjects, A method for providing physical growth, comprising providing a radar chart capable of being analyzed, a history tracker capable of analyzing a past growth record, and a prediction view capable of analyzing a growth prediction result of the user, Prediction modeling device. A body growth prediction modeling method performed by a body growth prediction modeling apparatus,
Receiving body growth data for an arbitrary user;
A growth prediction model based on the user's body growth data and a plurality of pre-stored body growth data for each of the comparison subjects based on the similarity of the user with the body growth data or the conditional probability of the user's body growth data, Modeling; And
Estimating and outputting a body growth value corresponding to the user through the growth prediction model,
Wherein modeling the growth prediction model comprises:
Modeling the growth prediction model based on at least one of a plurality of types of parameters included in the body growth data,
The body growth data for each of the plurality of comparison subjects is stored in a time-series manner for a predetermined period of time,
Wherein the arbitrary period corresponding to two or more comparison subjects among the plurality of comparison subjects has at least one of a range and an interval that are different from each other or a section is partially overlapped. 9. The method of claim 8,
Wherein modeling the growth prediction model comprises:
Providing a visual analytics user interface that supports modeling of the growth prediction model,
A parameter selector for selecting at least one of a plurality of types of parameters included in the body growth data through the visual analysis user interface, a histogram selector for outputting a distribution for each body growth data, A parallel coordinate view and a scatter plot view outputting the degree of correlation in a coordinate form and an algorithm modeling module for modeling the growth prediction model to which the selected condition is applied. 9. The method of claim 8,
Wherein modeling the growth prediction model comprises:
When the degree of similarity with the body growth data of the user is used, the growth prediction model is modeled based on a distribution model,
The body growth data of the comparison object most similar to the user is classified using the Euclidean distance technique, Manhattan distance technique, and cosine similarity technique as the distribution model And estimating the body growth value of the user based on the body growth data of the classified comparison subjects. 9. The method of claim 8,
Wherein modeling the growth prediction model comprises:
Modeling the growth prediction model based on a Bayesian inference model when the conditional probability of the user's body growth data is used,
Wherein the body growth data of the plurality of comparison subjects is grouped into a plurality of groups based on age and a prior probability that is a probability that the body growth data of the user is included in an arbitrary group is calculated, estimating a body growth state path by calculating a posterior probability based on a likelihood value of the user and estimating a body growth value of the user based on at least one group of body growth data included in the estimated body growth state path, , Body growth prediction modeling method. 9. The method of claim 8,
The body growth data may include:
Body mass index (BMI), basal metabolic rate (BMR), body weight, skeletal muscle mass, body fat mass, body water content, muscle mass, fat mass, protein level, mineral level, mineral level, Body component data including at least one of a fat percentage, a neck circumference dimension, a chest circumference dimension, an abdominal circumference dimension, a right arm circumference dimension, an left arm circumference dimension, a hip circumference dimension, a right thigh circumference dimension and a left thigh circumference dimension,
User identification information, gender, age, and time of data measurement. 9. The method of claim 8,
The step of predicting and outputting a body growth value corresponding to the user may include:
Providing a visual analytics user interface that supports analysis of predicted body growth predictions through the growth prediction model,
Through the visual analysis user interface,
A body state view capable of confirming the user's body state information, a cluster view capable of identifying objects having similar body growth patterns to the user and the plurality of comparison subjects, A method for providing physical growth, comprising providing a radar chart capable of being analyzed, a history tracker capable of analyzing a past growth record, and a prediction view capable of analyzing a growth prediction result of the user, Predictive modeling method. A recording medium on which a computer program for executing the method according to any one of claims 8 to 13 is recorded.

Images