WO2024106805A1

WO2024106805A1 - Method and apparatus for comprehensive diagnosis of latent ability through observation of child

Info

Publication number: WO2024106805A1
Application number: PCT/KR2023/017120
Authority: WO
Inventors: 이종열
Original assignee: 주식회사 생각의탄생
Priority date: 2022-11-14
Filing date: 2023-10-31
Publication date: 2024-05-23
Also published as: KR102575696B1

Abstract

The present invention relates to a method and an apparatus for comprehensive diagnosis of a child. In the present invention, a plurality of pieces of observation data generated from observation of a child by means of different entities are used to perform diagnosis such that child characteristics that cannot be perceived through a general test and the like can be objectively identified. Additionally, in the present invention, questions and answers of observation data are complexly connected to and correspond to one or more diagnoses, and thus the psychology and behavior of a child can be integrally analyzed and child characteristics can be comprehensively diagnosed.

Description

Method and device for comprehensive diagnosis of potential through child observation

The present invention relates to a comprehensive diagnostic method and device for children.

In the field of education, the need for individually tailored education for learners is continuously increasing. However, for such individually tailored education, it is essential to first understand the learner's characteristics. In other words, the learner's characteristics must first be identified before customized education can be provided.

The characteristics of adolescents and adults over a certain age can be easily identified through various tests. Furthermore, its characteristics can also be determined through methods using electronic devices. This is because adolescents and adults already have cognitive abilities, so there is not much distortion in the test results.

On the other hand, children do not yet have a certain level of cognitive ability, so there is a lot of distortion in test results. Therefore, there is a problem in that it is generally impossible to properly determine a child's characteristics through tests or methods using electronic devices. In addition, the existing test adopted a single, independent analysis method for each test based on a statistical approach, making it impossible to analyze the child's comprehensive psychological characteristics and behavior. As a result, comprehensive characteristic diagnosis and individually tailored education are not properly provided to children.

Objectively identify the child's characteristics that cannot be identified through tests, etc., and comprehensively analyze the child's psychology and behavior to comprehensively diagnose the child's characteristics.

The comprehensive diagnosis method according to the present invention includes the steps of (1) receiving observation data including responses to questions corresponding to individual areas in one or more diagnoses; and (2) generating each diagnostic result for one or more diagnoses using observation data.

A comprehensive diagnostic device for children according to the present invention includes a storage unit that stores observation data including responses to questions corresponding to individual areas in one or more diagnoses; and a generating unit that generates each diagnostic result for one or more diagnoses using observation data.

The present invention performs diagnosis using a plurality of observation data generated from observation of a child by different subjects, making it possible to objectively identify characteristics of a child that cannot be identified through general tests, etc.

In addition, the present invention can comprehensively analyze the child's psychology and behavior and comprehensively diagnose the child's characteristics by linking and corresponding to one or more diagnoses in a complex manner by the questions and responses of the observation data.

Even if the effect is not directly described herein, the effect predicted or anticipated by the description of the invention should be understood as described in the effect of the invention.

1 is a diagram showing the idea of the present invention.

Figure 2 is a diagram showing a system in which a diagnostic method according to an embodiment of the present invention is executed.

Figure 3 is a diagram showing the configuration of a diagnostic method according to an embodiment of the present invention.

4 to 6 are diagrams illustrating observation data being input in a diagnostic method according to an embodiment of the present invention.

Figures 7 and 8 are diagrams showing potential diagnosis results in the diagnosis method according to an embodiment of the present invention.

Figures 9 and 10 are diagrams showing the determination of a child's strengths and supplementary areas in the diagnostic method according to an embodiment of the present invention.

Figure 11 is a diagram illustrating determining a detailed type according to a strength area in the diagnostic method according to an embodiment of the present invention.

Figure 12 is a diagram showing a potential diagnosis model in the diagnosis method according to an embodiment of the present invention.

Figure 13 is a diagram showing the configuration of a diagnostic device according to an embodiment of the present invention.

The present invention is not limited to the embodiments described in the description of the invention and can be implemented in various ways. Terms used in the description of the invention are for describing embodiments and are not intended to limit the invention. In the description of the invention, the term 'comprising' is intended to designate the existence of a configuration in which the described features are combined. Therefore, it should be understood that it does not exclude the possibility of the presence or addition of one or more other features or combinations thereof. Hereinafter, detailed descriptions of the disclosed configurations and functions will be omitted as they are obvious to those skilled in the art.

1 is a diagram showing the idea of the present invention. Referring to Figure 1, the present invention relates to a system in which teachers and parents observe a child's behavior and comprehensively diagnose the child's characteristics using the observed data. Diagnosing these characteristics of children can provide convenience to teachers, objectivity to parents, and provide a common theme for children's education between teachers and parents.

Unlike adults, it is very difficult to diagnose the characteristics of children using data generated through tests or sensors. This is because children are influenced by various internal and external influences in the process of expressing their thoughts or judging something. Therefore, in order to properly diagnose a child's characteristics, the child's psychology and behavior must be observed. However, there is a problem in that observation can involve subjectivity. In order to eliminate these problems as much as possible, in the present invention, teachers and parents each observe children and accumulate observation data to ensure objectivity.

Observing a child's psychology and behavior takes a lot of time and effort, but the present invention can easily generate observation data through records using tags or responses to questions. Therefore, both teachers and parents can easily generate observation data in various areas about children just by selecting tags or answering questions.

Meanwhile, parents often have subjective biases about their children. Because of this, there are frequent cases where opinions differ between teachers and parents. The present invention can generate comprehensive diagnostic results for children by using observation data from teachers and parents, respectively. This allows parents to have an objective perspective on their children and judge their characteristics. And through the diagnosis results, teachers and parents have a common theme about the child, helping them understand the child's psychology and behavior more deeply.

Figure 2 is a diagram showing a system in which a diagnostic method according to an embodiment of the present invention is executed. Referring to Figure 2, a comprehensive diagnosis method of potential through observation of a child (hereinafter referred to as 'diagnosis method') according to an embodiment of the present invention is executed in the diagnosis device 100. Here, the diagnostic device 100 is a computing device that performs calculation and storage and is capable of processing data. The diagnostic device 100 transmits and receives data with the teacher device 200 and the parent device 300 to provide a child's diagnosis result. The diagnostic device 100 may transmit and receive data with the teacher device 200 and the parent device 300 through various methods including wired/wireless networks. Here, the teacher device 200 is a term referring to a computing device used by a teacher, and the parent device 300 is a term referring to a computing device used by parents.

Figure 3 is a diagram showing the configuration of a diagnostic method according to an embodiment of the present invention. Referring to FIG. 3, the diagnosis method according to an embodiment of the present invention includes the steps of (1) receiving observation data (S100) and (2) generating a diagnosis result (S200). Each step may be performed by a computing device, and subjects may be omitted from the description of each step.

In step (1) (S100), observation data including one or more of text and responses to questions corresponding to individual areas in one or more diagnoses is input.

Observation data includes character and numeric data. Observation data is generated from observations of children by different entities. Observers can generate observation data through records or responses to questions. Observation data has the characteristic that the objectivity and reliability of children's observations increase as the number of people and the amount of data generated increases. In an embodiment of the present invention, it is created by a teacher and a parent, but depending on the embodiment, it may also be created by a third party.

4 to 6 are diagrams illustrating observation data being input in a diagnostic method according to an embodiment of the present invention. Referring to FIG. 4, step (1) (S100) receives responses to questions in a Likert scale. Depending on the embodiment, on a 5-point Likert scale, responses can be assigned as 1 (not at all) - 2 (not much) - 3 (averagely) - 4 (usually) - 5 (extremely). And through the Likert scale and corresponding numbers, it is possible to score and add up scores for each area of diagnosis.

In the example of Figure 4, questions and responses to diagnose one child are input. On the other hand, in the example of FIG. 5, each response for a plurality of children is input. Referring to FIG. 5, in step (1) (S100), responses from a plurality of children can be received for questions randomly selected from among the questions. If a respondent answers questions about one child and then answers about another child, bias may occur in the respondent's answers due to previous responses. This is because observational data is generated after observing the child's psychology and behavior, and subjective involvement intervenes in the respondent's memory and judgment due to responses to multiple questions. Therefore, in the present invention, responses for several children are input based on one question. In other words, responses for each child are entered in parallel based on one question. In particular, these questions are selected randomly. Because respondents can focus on responding to one randomly selected question, they can respond to multiple children using the same criteria. And since only one observation result is entered when responding, respondents can respond much faster. Moreover, since a long time is not required for response, respondents can be encouraged to respond frequently in a short period of time, and a large number of observation data can be generated. Therefore, in the present invention, a large amount of observation data corresponding to various areas can be input in diagnosis.

Additionally, in step (1) (S100), a response selected from several examples can be input. Referring to Figure 6, after selecting a child for whom observation data is to be entered, a detailed response can be entered by selecting the observed area in the observation area. At this time, example sentences can be automatically generated according to preset tags or levels (scales). At this time, examples of multiple responses can be included even at the same level (scale). Accordingly, it is possible to more easily and quickly write observations or opinions recorded by the observer among the observation data.

Meanwhile, depending on the embodiment, in step (1) (S100), one or more of the groups and subgroups are selected, and observation data for a plurality of children included in the selected groups and subgroups is input. In general, observations of children are often conducted in educational institutions. And in educational institutions, several children are grouped into groups through classes or groups. At this time, observers frequently observe children in groups during the educational process. After the training process is over, if the observation unit and input unit match during the process of entering observation data, the observer can more easily and accurately generate observation data. For example, assume that there are children 'A', 'B', and 'C' in group 'B' of class 'A'. If observation data is entered after the education process for the entire class is completed, the observer can select class 'A' (group) and generate observation data for all children in class 'A'. In addition, if observation data is entered after the training course for group 'B' of class 'A' is over, the observer selects both class 'A' (group) and group 'B' (subgroup) and selects 'A' Observation data can be generated for children in group ‘B’ of the class. In the present invention, observation units and input units can be matched in the process of inputting observation data according to the observer's selection, making it possible to generate observation data more accurately and quickly.

Depending on the embodiment, the present invention may include various diagnostics. Diagnosis may include, but is not limited to, basic diagnosis, potential diagnosis, and self-esteem diagnosis. Meanwhile, each diagnosis includes several areas that classify the properties to be diagnosed. And the questions correspond to one or more areas covered by the diagnosis. Therefore, the attribute to be diagnosed can be measured based on the responses to the questions.

Referring again to FIG. 3, in step (2) (S200), each diagnosis result for one or more diagnoses is generated using observation data. The diagnosis result is the result of analyzing the properties to be diagnosed in each diagnosis. Therefore, diagnosis results may differ for each diagnosis. Basic diagnosis is a preliminary diagnosis for other diagnoses, and at the same time, it is a diagnosis to understand the parents' level of prior awareness and parenting attitude toward the child. Potential diagnosis is a diagnosis to identify areas of strength and weakness according to the Multiple Intelligence Theory (Multiple Intelligence tests for Aptitude Assessment). And the self-esteem diagnosis is a diagnosis to identify factors in a child's positive self-esteem. In the present invention, questions are linked to each diagnosis and used as data to generate complex diagnostic results.

Depending on the embodiment, during basic diagnosis, observation data includes basic data that are responses to questions corresponding to learning characteristics, motivation characteristics, creative characteristics, leadership characteristics, communication characteristics, spatial and abstract thinking characteristics, and adaptive abilities. In addition, when diagnosing self-esteem, observation data includes self-esteem data that is responses to questions corresponding to positive self-image, emotional stability, freedom and independence, empathy, and human friendliness. In basic diagnosis and self-esteem diagnosis, the scores for responses to questions corresponding to individual domains are added up, and then diagnostic results are generated according to the scores for each domain and the distribution of scores. Although the expression of terms for classifying individual areas in a diagnosis may be different, the configuration of classifying each diagnosis into multiple areas should be viewed as a technical feature of the present invention.

Additionally, observational data includes unstructured data. Depending on the embodiment, unstructured data may be text data. Observation results are not expressed in predetermined categories or numbers, but can be generated as text with meaning. At this time, the predetermined text may correspond to an individual area in one or more diagnoses. Therefore, observation data can include both categorical data based on responses to questions and text data based on records. And this observation data can be used as raw data to understand the characteristics of children.

Figures 7 and 8 are diagrams showing potential diagnosis results in the diagnosis method according to an embodiment of the present invention. Referring to Figures 7 and 8, when diagnosing potential, observation data corresponds to the child's verbal intelligence, logical-mathematical intelligence, kinesthetic intelligence, visuospatial intelligence, musical intelligence, natural exploration intelligence, interpersonal intelligence, and self-reflection intelligence. It includes multi-intelligence data that is the answers to the questions asked. Potential diagnosis analyzes strengths and complementary areas in eight intelligence areas. At this time, linguistic intelligence and logical-mathematical intelligence can be classified as series intelligence, body-kinesthetic intelligence, spatio-temporal intelligence, musical intelligence, and natural exploration intelligence can be classified as domain-specific intelligence, and interpersonal intelligence and self-reflection intelligence can be classified as relational intelligence.

The multiple intelligence data of the diagnostic method according to an embodiment of the present invention includes false regions. The questions corresponding to the false domain measure the child's desired and unconscious responses. In the present invention, if the response to the questions in the false domain is higher than a predetermined score, the diagnosis result is determined to be unreliable. Accordingly, as a result of the diagnosis, a re-diagnosis can be requested or the diagnosis can be discontinued. Depending on the embodiment, the diagnosis method of the present invention may include the step (3) of generating a re-diagnosis request or a request to stop diagnosis if the response to the questions in the false domain is more than a predetermined score (S300).

At this time, in the multi-intelligence data, each intelligence variable may follow a normal distribution. When the number of data exceeds a certain amount, the probability distribution of the random variable is approximated as a normal distribution according to the Central Limit Theory. Each intelligence variable in the multi-intelligence data is a random variable, and when the number of data exceeds a certain amount, it is approximated as a normal distribution according to the central limit theorem. In the present invention, since a large number of observation data are input, each intelligence variable of the multiple intelligence data also follows a normal distribution. If each intelligence variable in the multiple intelligence data follows a normal distribution, comparisons can be made for each intelligence variable. Therefore, it is possible to determine the relative characteristics of a child's areas of strength or weakness. Moreover, even if the absolute value of the score for each area is low, relative comparisons can be made within it, allowing the individual strengths and weaknesses of the child to be identified. Depending on the embodiment, the normal distribution may be converted to a standard normal distribution, and the score for each variable may be reset to a T score.

Figures 9 and 10 are diagrams showing the determination of a child's strengths and supplementary areas in the diagnostic method according to an embodiment of the present invention. Referring to Figure 9, in step (2) (S200), the strengths and complementary areas are determined by ranking each intelligence from the multiple intelligence data. At this time, the ranking by intelligence can be determined by the sum of scores based on the Likert scale.

Depending on the embodiment, the strength area and the complementary area may each be a combination of two areas. In step (2) (S200), the combination of the 1st and 2nd ranked areas can be judged as a strength area, and the combination of the 7th and 8th ranked areas can be judged as a complementary area. At this time, in the same ranking, series intelligence (linguistic intelligence and logical-mathematical intelligence) takes precedence over domain-specific intelligence (kinesthetic intelligence, visuospatial intelligence, musical intelligence, and natural exploration intelligence). And in serial intelligence, linguistic intelligence takes precedence over logical-mathematical intelligence. In intelligence by domain, natural intelligence, musical intelligence, visuospatial intelligence, and physical-kinesthetic intelligence are prioritized in that order.

Additionally, in step (2) (S200), if 2nd and 3rd place are tied, and if 2nd and 3rd place include self-reflective intelligence, 1st place and self-reflective intelligence are first judged as areas of strength. In particular, in step (2) (S200), if self-reflection is included among the first and second places, it is judged that there is a significantly high possibility that the rest are strengths. On the other hand, if introspective intelligence is not included among the second and third places, but interpersonal intelligence is included, series intelligence (verbal intelligence and logical-mathematical intelligence) takes precedence.

In the example of Figure 9, as a result of the diagnosis, the areas of strength were determined to be 'self-reflection' and 'logic and mathematics', and the areas of supplementation were determined to be 'interpersonal relationships' and 'space-time'. Referring to Figure 8, the first place in T-score is self-reflection (78.6), and the second place is logical mathematics (71.5). 8th place is interpersonal relationships (39.3), while 7th place is space-time (53.6) and natural exploration (53.6). However, in intelligence by domain, natural inquiry intelligence takes precedence, so natural inquiry is ranked 6th and space-time is 7th. Therefore, in the example of FIG. 9, in step (2) (S200), the complementary areas are determined as 'interpersonal relationships' and 'time and space'.

The diagnosis method according to an embodiment of the present invention includes the step (4) of generating a profile result for the diagnosis result (S400). Referring to Figure 10, specific profile results are generated for the determined strength areas and complementary areas. A profile can be created by combining tags for the child's characteristics with a predetermined template for each area. However, it is not limited to this, and if the content can be generated according to the diagnosis results, the content of the profile can be generated in various ways. Therefore, by looking at the profile results, teachers and parents can easily identify the behaviors the child likes, the behaviors the child is good at, activities that strengthen areas of strength, and activities that require supplementary areas. In addition, you can suggest customized education to your child by encouraging activities that can further strengthen areas of strength and supplement areas of weakness.

Figure 11 is a diagram illustrating determining a detailed type according to a strength area in the diagnostic method according to an embodiment of the present invention. Referring to FIG. 11, step (2) (S200) determines the detailed type according to the strength area. Specifically, the detailed type is judged based on the combination of the strength area and the top 3 areas. In the example of FIG. 11, the detailed types include, but are not limited to, mastery type, human-friendly type, understanding type, and self-expression type, and may include various types. In addition, the expression of terms for classifying detailed types may be different, but the configuration of classifying detailed types according to strength areas and their ranks should be viewed as a technical feature of the present invention.

In step (2) (S200), if self-reflection intelligence is included in the strength area or in the top 3, it is judged as mastery type. Alternatively, in step (2) (S200), if interpersonal intelligence is included in the strength area or 3rd place, it is judged to be human-friendly. Alternatively, in step (2) (S200), if logical-mathematical intelligence is included in the area of strength or in the top 3, the person is judged to be an understanding type. Alternatively, in step (2) (S200), if verbal intelligence is included in the strength area or in the top 3, it is judged as a self-expression type. Since the strength area is a combination of two areas, the detailed type can be determined by considering at least one of self-reflective intelligence, interpersonal intelligence, logical-mathematical intelligence, and verbal intelligence, as well as the third area.

Figure 12 is a diagram showing a potential diagnosis model in the diagnosis method according to an embodiment of the present invention. Referring to Figure 12, in step (2) (S200), machine learning is used to create and learn a potential diagnosis model with multiple intelligence data as input variables and strength areas, complementary areas, and detailed types as output variables. can do. At this time, the potential diagnosis model may be a classification model. Depending on the embodiment, the potential diagnosis model may be learned using learning data having input variables and output variables of the same structure. The potential diagnosis model can learn correlations between variables and output strength areas, complementary areas, and detailed types. As the amount of learning data increases, the accuracy of the results output from the potential diagnosis model increases. Depending on the embodiment, the potential diagnosis model may receive multiple intelligence data including false regions. Alternatively, it may include multiple intelligence data and receive one or more of basic data and self-esteem data. As the number of input variables increases, the potential diagnosis model can output more accurate classification results by reflecting complex interrelationships between input variables. Depending on the embodiment, the potential diagnosis model may be verified using verification data having input variables and output variables of the same structure, and fine tuning of the model may be performed to increase accuracy.

In step (2) (S200), when the first result, which is the judgment result according to the potential diagnosis algorithm, and the second result, which is the classification result of the potential diagnosis model, are different from each other, observation data, the first result, and the second result are included. Unjudged data can be generated. The potential diagnosis algorithm corresponds to the strength area and supplement area judgment algorithm described above. If the first and second results are different, an in-depth judgment regarding the diagnosis of the child's potential may be required. Therefore, in step (2) (S200), unjudged data including observation data, which is raw data, and first and second results are generated. An alarm may be generated when such unjudged data is generated, or only the unjudged data may be classified through separate filtering. Afterwards, the different judgment results from the first and second results are compared to conduct an in-depth diagnosis of the child's potential. In in-depth diagnosis, the diagnostic results of both basic diagnosis and self-esteem diagnosis can be used. In particular, in the present invention, it is possible to generate a final diagnosis result of undetermined data by using even unstructured data included in observation data. Depending on the embodiment, the final diagnosis result may be generated by measuring the frequency with which a tag corresponding to a specific area is included among the text data included in the observation data and weighting the area with the mode. Therefore, the present invention can perform in-depth diagnosis by referring to both results even when the results between the existing algorithm and the machine-learned potential diagnosis model are different, and more accurately diagnose the child's potential characteristics. .

Figure 13 is a diagram showing the configuration of a diagnostic device according to an embodiment of the present invention. Depending on the embodiment, the functions performed in each step of the diagnostic method can be converted into each configuration and implemented as a diagnostic device including these configurations. Referring to FIG. 13, the diagnostic device according to an embodiment of the present invention uses (1) a storage unit to store observation data including responses to questions corresponding to individual areas in one or more diagnoses, and (2) the observation data. It includes a generating unit that generates each diagnosis result for one or more diagnoses. At this time, observation data may be generated from observations of children by different subjects. Since each configuration overlaps with the description of the steps of the above-described diagnostic method, detailed description thereof will be omitted.

The contents described in the description of the invention are merely examples, and the invention may be implemented in various modifications by those skilled in the art. Therefore, the scope of protection of the present invention is not limited to the description and expression of the described embodiments.

Claims

(1) receiving observation data including one or more of text and responses to questions corresponding to individual areas in one or more diagnoses; and

(2) generating each diagnostic result for the one or more diagnoses using the observation data,

A comprehensive diagnosis method of potential through observation of a child, characterized in that the observation data is generated from observation of the child by different subjects.
In claim 1,

In step (1) above,

A comprehensive diagnosis method of potential through child observation, characterized in that it receives responses from a plurality of children to questions randomly selected from among the above questions.
In claim 1,

In step (1) above,

A comprehensive diagnosis method of potential through child observation, characterized in that at least one of a group and a subgroup is selected and observation data for a plurality of children included in the selected group and subgroup are input.
In claim 1,

The above diagnosis is,

A comprehensive diagnosis method of potential through observation of a child, characterized in that it includes at least one of the child's basic diagnosis, potential diagnosis, and self-esteem diagnosis.
According to any one of claims 2 to 4,

In step (1) above,

A comprehensive diagnosis method of potential through child observation, characterized by receiving responses to the above questions on a Likert scale.
In claim 5,

When diagnosing the potential, the observation data is,

Contains multiple intelligence data that are responses to questions corresponding to the child's verbal intelligence, logical-mathematical intelligence, physical-kinesthetic intelligence, visuospatial intelligence, musical intelligence, natural exploration intelligence, interpersonal intelligence, and self-reflection intelligence,

The multi-intelligence data is,

A comprehensive diagnosis method of potential through child observation, characterized in that each intelligence variable follows a normal distribution.
In claim 6,

In step (2) above,

A comprehensive diagnosis method of potential through observation of children, characterized in that the ranking of each intelligence is determined from the multiple intelligence data and the areas of strength and complementary areas are determined.
In claim 7,

In step (2) above,

The combination of the 1st and 2nd ranked areas is judged to be a strength area, and the combination of the 7th and 8th ranked areas is judged to be a complementary area.

In the same ranking, the verbal intelligence and the logical-mathematical intelligence take precedence over the kinesthetic intelligence, the visuospatial intelligence, the musical intelligence, and the natural exploration intelligence,

A comprehensive diagnosis method of potential through observation of a child, characterized in that priority is given to the natural exploration intelligence, the musical intelligence, the spatial-temporal intelligence, and the physical-kinesthetic intelligence.
In claim 8,

In step (2) above,

In the case where 2nd and 3rd place are tied, if the 2nd and 3rd places include the self-reflective intelligence, the combination of 1st place and the above self-reflective intelligence is first judged as a strength area, through child observation. Comprehensive diagnosis method of potential
In claim 7,

In step (2) above,

A comprehensive diagnosis method of potential through child observation, characterized by determining detailed types based on a combination of the above strength areas and the top 3 areas.
In claim 10,

In step (2) above,

Using machine learning, create and learn a potential diagnosis model with the multiple intelligence data as input variables and the strength area, the complementary area, and the detailed type as output variables,

A comprehensive diagnosis method of potential through child observation, wherein the potential diagnostic model is a classification model.
In claim 11,

In step (2) above,

When the first result, which is the judgment result according to the potential diagnosis algorithm, and the second result, which is the classification result of the potential diagnosis model, are different from each other, unjudged data including the observation data, the first result, and the second result. A comprehensive diagnosis method of potential through child observation, characterized by generating
In claim 12,

In step (2) above,

Comprehensive analysis of potential abilities through child observation, characterized by measuring the frequency with which tags corresponding to specific areas are included among the text data included in the observation data, and generating a final diagnosis result by weighting the areas with the mode. Diagnosis method.
a storage unit that stores observation data including responses to questions corresponding to individual areas in one or more diagnoses; and

A generator that generates diagnostic results for each of the one or more diagnoses using the observation data,

A comprehensive diagnostic device for potential through observation of a child, characterized in that the observation data is generated from observation of the child by different subjects.