Multi-perspective-based student characteristic analysis system and method

The multi-perspective student characteristic analysis system integrates self-, peer-, and teacher-reported data to objectively analyze and visually present student characteristics, addressing the limitations of single-perspective evaluations by providing actionable educational insights.

KR103046029B1Active Publication Date: 2026-11-10주식회사 클래즈
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020260049458
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Priority Date
2026-02-04
Filing Date
2026-03-19
Publication Date
2026-11-10
Estimated Expiration
2046-03-19

AI Technical Summary

Technical Problem

Conventional student evaluation methods relying on single perspectives, such as self-assessment, teacher, or parent evaluations, fail to provide a comprehensive understanding of a student's behavioral characteristics or emotional state, leading to biased or inconsistent results.

Method used

A multi-perspective-based student characteristic analysis system that integrates data from self-report, peer-report, and teacher-report, converting scores into standard scores, determining level values, and classifying patterns of agreement and discrepancy to generate educational interpretations and guidance plans.

Benefits of technology

Enables a more objective, three-dimensional understanding of student characteristics by quantitatively analyzing and visually presenting the agreement and discrepancy between different perspectives, facilitating efficient educational support and counseling.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112026033446260-PAT00001_ABST
    Figure 112026033446260-PAT00001_ABST
Patent Text Reader

Abstract

The present invention relates to a multi-perspective-based student characteristic analysis system comprising: a data collection unit that collects evaluation data from at least two perspectives among self-report, peer-report, parent-report, or teacher-report regarding the student characteristics of the same student and generates student characteristic scores by perspective; a standard score calculation unit that converts the student characteristic scores generated by the data collection unit into standard scores based on class standards; a level determination unit that determines the level value of the student characteristics based on the standard scores calculated by the standard score calculation unit; a match determination unit that determines whether there is a match by comparing the level values ​​by perspective determined by the level determination unit; an interpretation generation unit that generates educational interpretations and guidance plans according to at least one of the standard scores, level values, or whether there is a match; and a visualization unit that generates a visualization screen displaying at least one of the standard scores, level values, whether there is a match, the educational interpretation, or the guidance plans. According to the present invention, student characteristic evaluation data collected from different perspectives, such as the student's own evaluation, peer evaluation, parental evaluation, and teacher evaluation, can be analyzed integrally, thereby enabling a more objective and three-dimensional understanding of the student's behavioral characteristics and emotional state compared to a single-perspective evaluation.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a multi-perspective-based student characteristic analysis system and method, and more specifically, to a student characteristic analysis system and method that integrates and analyzes student characteristic evaluation data collected from different perspectives, such as the student's own evaluation, peer evaluation, parent evaluation, and teacher evaluation, determines the degree of agreement and inconsistency patterns between perspectives, and visually provides the results. Background Technology

[0002] Various assessment methods are being utilized in educational settings to identify students' dispositions, behavioral characteristics, and learning engagement. Such student assessments are generally conducted based on questionnaires, observation records, teachers' subjective judgments, or psychological test results, and serve as foundational data for understanding students' emotional state, social skills, and learning attitudes.

[0003] However, conventional student evaluations often rely on specific perspectives, which presents a problem in that it is difficult to comprehensively grasp a student's actual state. For example, relying solely on a student's own self-assessment can lead to biases in self-perception, while evaluations by teachers or parents may fail to adequately reflect the behavioral characteristics a student exhibits in peer relationships. As such, evaluations based on a single perspective have limitations in understanding a student's behavioral characteristics or emotional state from multiple angles.

[0004] Accordingly, to understand students' characteristics more objectively, multi-perspective assessment methods—which evaluate students from different viewpoints including not only the student themselves but also peers, parents, and teachers—are being utilized in the fields of education and psychological assessment. This multi-perspective assessment has the advantage of enabling a more three-dimensional understanding of a student's behavioral traits or emotional state by comparing different assessment results regarding the same student characteristics.

[0005] However, results collected through multi-perspective evaluations often vary depending on the evaluator, making it difficult to interpret these differences and grasp the student's actual status. In particular, systematic methods of analysis and presentation are frequently not sufficiently provided to comprehensively compare evaluation results collected from multiple viewpoints and to intuitively understand their meaning.

[0006] Therefore, there is a need for technology that can systematically analyze student characteristic assessment data collected from different perspectives and provide an intuitive understanding of the relationships or differences between those perspectives. Prior art literature

[0007] Republic of Korea Registered Patent No. 10-2295503 The problem to be solved

[0008] The objective of the present invention is to provide a multi-perspective-based student characteristic analysis system and method that solves the aforementioned conventional problems by integrally analyzing student characteristic evaluation data collected from different perspectives, such as student self-evaluation, peer evaluation, parent evaluation, and teacher evaluation, determining the degree of agreement and inconsistency patterns between perspectives, and visually providing the results. means of solving the problem

[0009] The above objective is achieved by a multi-perspective-based student characteristic analysis system according to the present invention, comprising: a data collection unit that collects evaluation data from at least two perspectives among self-report, peer report, parent report, or teacher report regarding the student characteristics of the same student and generates student characteristic scores by perspective; a standard score calculation unit that converts the student characteristic scores generated by the data collection unit into standard scores based on class standards; a level determination unit that determines the level value of the student characteristics based on the standard scores calculated by the standard score calculation unit; a match determination unit that determines whether there is a match by comparing the level values ​​by perspective determined by the level determination unit; an interpretation generation unit that generates educational interpretations and guidance plans based on at least one of the standard scores, level values, or whether there is a match; and a visualization unit that generates a visualization screen displaying at least one of the standard scores, level values, whether there is a match, the educational interpretation, or the guidance plans.

[0010] In addition, the data collection unit performs the peer report in a peer nomination manner, aggregates the number of nominations by peer students for a specific student, and calculates the nomination frequency as a ratio to the total number of respondents in the class to generate the student characteristic score, and the level determination unit can determine the level value as low if the standard score is less than 40 points, as average if it is 40 points or more and 60 points or less, and as high if it is greater than 60 points.

[0011] Additionally, the system further includes a pattern classification unit that classifies patterns corresponding to combinations of level values ​​by perspective based on the above-mentioned matching status, and the pattern classification unit can classify the patterns into a plurality of patterns including a self-reporting only low pattern, a self-reporting only high pattern, a parent-reporting only high pattern, a parent-reporting only low pattern, a teacher-reporting only high pattern, a teacher-reporting only low pattern, a peer-reporting only different pattern, a pattern where all are consistent, and a 2:2 separation pattern.

[0012] Additionally, the visualization unit may display a visualization screen including at least one of a summary view or a detailed view, the summary view may display visual identification elements indicating a percentage of agreement and whether there is an agreement by perspective, and the detailed view may display a matrix heatmap, a horizontal gauge, the educational interpretation, and the guidance plan.

[0013] In addition, the method is achieved by a multi-perspective-based student characteristic analysis method comprising: a collection step for collecting evaluation data from at least two perspectives among self-report, peer report, parent report, or teacher report regarding the student characteristics of the same student to generate student characteristic scores for each perspective; a standard score calculation step for converting the student characteristic scores generated in the collection step into standard scores based on class standards; a level determination step for determining the level value of the student characteristics based on the standard scores calculated in the standard score calculation step; a consistency determination step for determining whether there is consistency between perspectives by comparing the level values ​​for each perspective determined in the level determination step; a pattern classification step for classifying patterns based on the consistency determined in the consistency determination step and the combination of the level values ​​for each perspective; an interpretation step for generating educational interpretations and guidance plans according to the patterns classified in the pattern classification step; and a visualization step for generating a visualization screen that displays the results generated in the interpretation step. Effects of the invention

[0014] According to the present invention, student characteristic evaluation data collected from different perspectives, such as the student's own evaluation, peer evaluation, parental evaluation, and teacher evaluation, can be analyzed integrally, thereby enabling a more objective and three-dimensional understanding of the student's behavioral characteristics and emotional state compared to a single-perspective evaluation.

[0015] In addition, since evaluation data collected from each perspective is converted into standard scores based on common criteria, and the level values ​​of student characteristics are determined based on this, and patterns of agreement, consistency, and discrepancy between perspectives can be analyzed, the difference between students' self-perception and perception of others can be identified quantitatively and systematically.

[0016] In addition, since educational interpretation information and guidance plans can be automatically generated based on classified patterns, teachers, guardians, or counselors can more efficiently understand the characteristics and status of students and establish educational support or counseling directions suitable for each student's situation.

[0017] In addition, since the analysis results can be provided as a visualization screen including summary and detailed views, evaluation results from multiple perspectives, consistency, discrepancy patterns, and interpretation information can be intuitively compared and understood, thereby improving the communicability and usability of the student characteristic analysis results. Brief explanation of the drawing

[0018] FIG. 1 shows the connection configuration of a multi-perspective-based student characteristic analysis system according to one embodiment of the present invention, and FIG. 2 shows the detailed configuration of a multi-perspective-based student characteristic analysis system according to an embodiment of the present invention, and FIG. 3 shows the overall flow of a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, and FIG. 4 is a diagram showing an example of criteria for determining level values ​​based on standard scores in a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, and FIG. 5 is a diagram showing an example of generating educational interpretation information based on a match judgment result in the case of two perspectives in a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, and Figure 6 is a diagram showing an example of generating a guidance plan based on the pattern classification results in the case of four perspectives in a multi-perspective-based student characteristic analysis method according to one embodiment of the present invention. Specific details for implementing the invention

[0019] Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. It should be noted that in assigning reference numerals to the components of each drawing, the same components are given the same reference numeral whenever possible, even if they are shown in different drawings.

[0020] Furthermore, in describing the embodiments of the present invention, if it is determined that a detailed description of related known configurations or functions would hinder understanding of the embodiments of the present invention, such detailed description is omitted.

[0021] In addition, terms such as first, second, A, B, (a), (b), etc., may be used when describing the components of the embodiments of the present invention. These terms are used merely to distinguish the components from other components, and the essence, order, or sequence of the components is not limited by the terms used.

[0022] Additionally, terms such as '~part' or 'system' used in this specification may be interpreted to refer to computer-related entities, such as hardware, a combination of hardware and software, or software currently in execution. For example, the data collection unit (110), standard score calculation unit (120), level determination unit (130), match determination unit (140), pattern classification unit (150), interpretation generation unit (160), and visualization unit (170) according to embodiments of the present invention may be implemented in the form of software modules executed by at least one computation processing unit, or may be implemented as electrical hardware devices comprising one or more processors and memory for storing programs executed by said processors. Additionally, each of these components may be configured to transmit and receive electrical data and to perform computation and processing through a communication bus or network.

[0023] Now, with reference to the attached drawings, a multi-perspective-based student characteristic analysis system (100) according to one embodiment of the present invention will be described in detail.

[0024] Figure 1 shows the connection configuration of a multi-perspective-based student characteristic analysis system according to one embodiment of the present invention.

[0025] As illustrated in FIG. 1, a multi-perspective-based student characteristic analysis system (100) according to one embodiment of the present invention may be configured to be connected to a user terminal (200).

[0026] A user terminal (200) may be a device used by a student, peer student, guardian, teacher, counselor, or other user to input student characteristic evaluation data or to check student characteristic analysis results. For example, the user terminal (200) may be a computing device such as a smartphone, tablet, laptop, or desktop computer. The user terminal (200) may be configured to communicate with a multi-perspective-based student characteristic analysis system (100) via a network and may be configured to display evaluation questions, analysis results, a summary view, and a detailed view.

[0027] A multi-perspective-based student characteristic analysis system (100) may be configured to receive evaluation data on student characteristics from a user terminal (200), process the evaluation data to analyze student characteristics, and generate a visualization screen based on the analysis results.

[0028] FIG. 2 shows the detailed configuration of a multi-perspective-based student characteristic analysis system (100) according to one embodiment of the present invention.

[0029] As illustrated in FIG. 2, the multi-perspective-based student characteristic analysis system (100) more specifically includes a data collection unit (110), a standard score calculation unit (120), a level determination unit (130), a match determination unit (140), a pattern classification unit (150), an interpretation generation unit (160) and / or a visualization unit (170).

[0030] The data collection unit (110) may be configured to collect evaluation data regarding student characteristics from at least two perspectives among self-report, peer-report, parent-report, or teacher-report.

[0031] Here, "student characteristics" may refer to traits that indicate a student's disposition or behavioral patterns, such as behavioral characteristics, emotional state, social interaction characteristics, or learning participation characteristics. For example, student characteristics may include, but are not limited to, prosociality, aggression, emotional stability, adaptability to peer relationships, learning participation, or adaptability to school life.

[0032] Furthermore, perspectives for evaluating student characteristics may include self-reporting, peer reporting, parent reporting, or teacher reporting. Self-reporting may be a method in which the student themselves evaluates their own behavioral characteristics or emotional state, while peer reporting may be a method in which peers belonging to the same class or group evaluate the behavioral characteristics of the subject student. Additionally, parent reporting may be a method in which a guardian evaluates the student's behavioral characteristics observed in the home environment, and teacher reporting may be a method in which a teacher evaluates the student's behavioral characteristics observed during classroom situations or school life.

[0033] The data collection unit (110) may be configured to collect evaluation data by providing a survey or evaluation form consisting of multiple evaluation questions to a user terminal (200) and allowing evaluators of each perspective to input responses to the evaluation questions. Each evaluation question may consist of a question about behavior or attitude related to a specific student characteristic, and the evaluator may respond to the question by selecting one of multiple score scales.

[0034] Meanwhile, peer reporting can be performed using a peer nomination method as an example. The peer nomination method may be a method in which multiple students within a class nominate a single student or multiple students who are judged to correspond to a specific behavioral characteristic. For example, if a specific student characteristic is prosociality, students within the class may select a single student or multiple students who correspond to that characteristic. The data collection unit (110) may be configured to aggregate the number of times each student is nominated and to generate a student characteristic score by converting the nomination frequency into a ratio relative to the total number of respondents or a normalized value. Accordingly, peer reporting can be implemented not only as a simple survey response method but also as a sociometry-based quantitative evaluation method that reflects mutual perception within the group.

[0035] Additionally, the data collection unit (110) may be configured to collect evaluation data through a user terminal (200) used by the student, peer student, guardian, or teacher. Evaluation questions are provided through the user terminal (200), and each evaluator can input a response to the question.

[0036] Additionally, the data collection unit (110) may be configured to store evaluation data collected from different perspectives regarding the same student characteristics in conjunction with a single student identification information. Accordingly, evaluation results between perspectives can be compared and analyzed in a subsequent step.

[0037] Additionally, the data collection unit (110) may be configured to generate student characteristic scores by aggregating response values ​​for multiple evaluation questions or calculating average values ​​in order to calculate evaluation data collected from each perspective as a score for each student characteristic, and when a peer nomination method is applied, the data collection unit (110) may be configured to generate student characteristic scores by aggregating the number of times a specific student is nominated by peer students and converting the nomination frequency into a ratio relative to the total number of respondents or a normalized value.

[0038] The standard score calculation unit (120) can be configured to convert the student characteristic score generated by the data collection unit (110) into a standard score based on the class.

[0039] Here, the student characteristic score may be a score calculated based on evaluation data collected from each perspective, or a raw score generated by summing or averaging the response values ​​for multiple evaluation items corresponding to a specific student characteristic.

[0040] The standard score calculation unit (120) can be configured to convert the evaluation results by perspective into a form comparable to a common standard by normalizing the student characteristic scores based on the distribution of multiple students belonging to the same class. Accordingly, the standard score calculation unit (120) can generate a standard score based on the class standard to indicate where the student characteristic score of a specific student is located relative within the corresponding class group.

[0041] For example, the standard score calculation unit (120) can calculate the mean, variance, and standard deviation of student characteristic scores collected for students belonging to the same class, and generate a standard score by normalizing the difference between a specific student's student characteristic score and the mean value using the variance or standard deviation. Accordingly, evaluation results collected from different perspectives can be compared on the same reference axis. At this time, the standard score can be calculated using the T-score method, and in this case, it can be calculated according to the formula T = 50 + 10 × (X - M) / SD based on a specific student's student characteristic score X, class mean value M, and standard deviation SD. However, the method of calculating the standard score is not limited to this, and various normalization methods may be applied.

[0042] Additionally, the standard score calculation unit (120) may be configured to calculate standard scores independently for each student characteristic. For example, a student characteristic score for prosociality may be converted into a standard score based on the score distribution within the class corresponding to prosociality, and a student characteristic score for aggression may be converted into a standard score separately based on the score distribution within the class corresponding to aggression. Accordingly, the relative level of each characteristic can be accurately reflected without confusion between different student characteristics.

[0043] In addition, the standard score calculation unit (120) can be configured to enable direct comparison between self-reported scores, peer-reported scores, parent-reported scores, and teacher-reported scores by processing student characteristic scores collected by perspective according to the same conversion standard. For example, when the self-reported score and the teacher-reported score for the same student characteristic are each converted into standard scores based on class standards, the difference between the two scores can be used to compare and analyze the difference between the student's own perception and the teacher's perception.

[0044] The level determination unit (130) may be configured to determine the level value of the student characteristics based on the standard score calculated by the standard score calculation unit (120).

[0045] Here, the level value may be a value indicating the relative level at which a specific student characteristic is located within a class group, and may be a value classified into one of multiple level intervals such as low, average, or high.

[0046] The level determination unit (130) may be configured to determine the level value of a student's characteristics by comparing the standard score calculated by the standard score calculation unit (120) with a plurality of pre-set standard intervals. For example, the level determination unit (130) may be configured to determine a low level if the standard score is less than the first standard value, determine a normal level if it is greater than or equal to the first standard value and less than or equal to the second standard value, and determine a high level if it exceeds the second standard value. However, the number of level intervals or the method of setting the standard values ​​are not limited to this and may be set in various ways depending on the educational setting or the purpose of evaluation.

[0047] Additionally, the level determination unit (130) may be configured to determine level values ​​independently for each student characteristic. For example, separate level values ​​may be determined based on standard scores for each of different student characteristics, such as prosociality, aggression, emotional stability, or learning participation. Accordingly, level values ​​corresponding to each of multiple student characteristics may be generated for a specific student.

[0048] Additionally, the level determination unit (130) may be configured to independently determine level values ​​for standard scores calculated from different perspectives. For example, for the same student characteristics of the same student, it may determine level values ​​corresponding to the standard score based on self-report, the standard score based on peer report, the standard score based on parent report, and the standard score based on teacher report. Accordingly, the level of student characteristics perceived from each perspective can be individually identified.

[0049] Additionally, the level determination unit (130) may be configured to store or provide the determined level value so that it can be used for comparative analysis between perspectives in a subsequent stage. The level value may be used as an input value to determine whether there is agreement or a pattern of disagreement between perspectives in a subsequent stage. Through this, differences in perception between different perspectives regarding the same student's characteristics can be systematically analyzed.

[0050] Additionally, the level determination unit (130) may be configured to determine the standard score as a low, normal, or high level value according to a preset score range. For example, the level determination unit (130) may be configured to determine the standard score as low if it is less than 40 points, determine the standard score as normal if it is 40 points or more and 60 points or less, and determine the standard score as high if it exceeds 60 points. Accordingly, standard scores calculated from different perspectives can be converted into a common level system and compared, so that the consistency or inconsistency pattern between perspectives can be more easily determined in subsequent steps.

[0051] The agreement judgment unit (140) may be configured to determine whether the evaluation results between perspectives are consistent by comparing the level values ​​for each perspective calculated by the level judgment unit (130).

[0052] The agreement judgment unit (140) may be configured to compare the level values ​​of each perspective and determine that if they are judged to be at the same level, it is considered an agreement, and if they are judged to be at different levels, it is considered an inconsistency. For example, if there are two input perspectives, it may be determined that there is an agreement if the level values ​​of the two perspectives are the same, and an inconsistency if the level values ​​of the two perspectives are different. In addition, if there are four input perspectives, it may be determined that there is an agreement across all perspectives if the same level value is produced in self-report, peer report, parent report, and teacher report for the same student characteristic, and if different level values ​​are produced in some perspectives, it may be determined that there is an inconsistency between perspectives.

[0053] Additionally, the agreement judgment unit (140) may be configured to determine whether there is an agreement based on the degree of difference between level values ​​for each perspective. For example, if the difference between level values ​​calculated from different perspectives is below a preset allowable range, it may be considered substantially the same level and judged as an agreement, and if it exceeds the allowable range, it may be judged as an inconsistency. Accordingly, it is possible to prevent excessive judgment of inconsistency regarding minute differences between evaluation perspectives.

[0054] Additionally, the agreement judgment unit (140) may be configured to calculate the degree of agreement between views in order to comprehensively determine whether there is an agreement between multiple views. For example, if there are two input views, the degree of agreement may be calculated based on whether there is an agreement between the two views, and if there are three or more input views, the degree of agreement may be calculated by dividing the number of views having the same level value by the total number of views, or by assigning weights based on the difference in levels between views.

[0055] Additionally, the agreement judgment unit (140) may be configured to generate judgment results for each student characteristic, and accordingly, whether or not there is an agreement or the degree of agreement between perspectives corresponding to each of multiple student characteristics for a specific student may be calculated. The results calculated in this way may be used in a subsequent step to classify patterns of mismatch between perspectives or to visualize the results of the student characteristic analysis.

[0056] The pattern classification unit (150) may be configured to classify into one of a plurality of patterns based on a combination of level values ​​by perspective, based on the type of mismatch or the result of the match judgment determined by the match judgment unit (140).

[0057] Here, the term "discrepancy type" may refer to cases where level values ​​calculated from different perspectives differ regarding the same student's characteristics. For example, if a student is judged to be at a high level in a self-report but at an average level in a teacher report, or at a low level in a peer report, the discrepancy may be identified as a discrepancy type when level values ​​differ between perspectives.

[0058] The pattern classification unit (150) may be configured to classify the type of mismatch into one of a plurality of predefined patterns according to a combination of level values ​​for each perspective. Additionally, the pattern classification unit (150) may classify it as a matching pattern if all perspectives are determined to have the same level value by the matching judgment unit (140), and may be configured to classify it as a partial mismatch pattern if a difference occurs only in some perspectives.

[0059] Additionally, the pattern classification unit (150) may be configured to apply different detailed pattern systems depending on the number of input perspectives. For example, if there are two input perspectives, it may be classified as a matching pattern if the level values ​​of the two perspectives are the same, and as a mismatched pattern if the level values ​​of the two perspectives are different. Additionally, if there are four input perspectives, it may be classified into multiple patterns including a self-report only low pattern, a self-report only high pattern, a parent-report only high pattern, a parent-report only low pattern, a teacher-report only high pattern, a teacher-report only low pattern, a peer-report only different pattern, an all-matching pattern where the level values ​​of all perspectives are the same, or a 2:2 separation pattern where two perspectives and two different perspectives are distinguished from each other, depending on the combination of level values ​​for each perspective.

[0060] Accordingly, the pattern classification unit (150) can more specifically identify which perspectives cause a difference in perception regarding the same student's characteristics based on the number of perspectives input, and can generate more precise educational interpretation and guidance plans in subsequent steps.

[0061] Additionally, the pattern classification unit (150) may be configured to generate classified pattern information according to student characteristics, and pattern information corresponding to each of multiple student characteristics may be generated for a specific student. The pattern information generated in this way can be used as basic information for interpreting or visually representing the results of the student characteristic analysis in a subsequent step.

[0062] Meanwhile, the pattern classification unit (150) may be configured to additionally determine whether a high-risk state exists by comparing the difference in standard scores between self-report and other perspective reports when a pattern is identified where the student characteristic is a positive variable and only self-report is high. For example, if the standard score of self-report is determined to be at a high level and the difference in standard scores with peer reports, parent reports, and teacher reports is greater than or equal to a preset difference in scores, the pattern classification unit (150) may classify the student characteristic as a high-risk state in which there is a large discrepancy between self-perception and other perceptions. Accordingly, the pattern classification unit (150) can go beyond simple discrepancy pattern classification and separately identify situations where the degree of bias in the student's self-perception is large.

[0063] The interpretation generation unit (160) can be configured to generate educational interpretations and guidance plans based on the patterns classified by the pattern classification unit (150).

[0064] Here, educational interpretation may refer to information explaining the meaning of evaluation results presented from different perspectives regarding specific student characteristics, and guidance plan may refer to educational measures or guidance methods that teachers or guardians can refer to in order to improve or support the student's behavioral characteristics or emotional state based on the educational interpretation.

[0065] The interpretation generation unit (160) may be configured to generate different educational interpretations based on the number of input perspectives and the patterns classified by the pattern classification unit (150). For example, if there are two input perspectives, the educational interpretation may be generated based on combinations formed when the level value of the first perspective and the level value of the second perspective are each determined to be low, normal, or high. That is, the interpretation generation unit (160) may be configured to derive a total of nine combination patterns when three level values ​​are possible for each of the two perspectives, and to generate judgment results, interpretation directions, and guidance plans corresponding to each combination pattern. For example, if the level values ​​of both perspectives are low, it may be determined as a matching pattern and interpreted as a situation requiring joint intervention, and if the level value of the first perspective is high and the level value of the second perspective is low, it may be determined as a mismatched pattern and interpreted as a situation requiring more objectification of self-awareness. In addition, if the level value of the first perspective is low and the level value of the second perspective is high, it is determined as a discrepancy pattern and can be interpreted as a situation requiring self-esteem improvement.

[0066] In addition, if there are four input perspectives, a total of 81 combination patterns are derived where three level values ​​are possible for each, and the system can be configured to generate judgment results, interpretation directions, and guidance plans corresponding to each combination pattern. For example, a pattern of "low self-reporting only" or "high self-reporting only" can be interpreted as indicating a difference between a student's self-perception and other perspectives, while a pattern of "high parent-reporting only" or "low parent-reporting only" can be interpreted as suggesting a difference in perception between home and school or specific situations in the home environment. Furthermore, a pattern of "high teacher-reporting only" or "low teacher-reporting only" can be interpreted as suggesting specific situations based on the classroom environment or the teacher's observation context, and a pattern of "different peer-reporting only" can be interpreted as a situation reflecting the characteristics of interaction in peer relationships. Additionally, a pattern of unanimous agreement can be interpreted as a case where the same state perception is formed from multiple perspectives, and a 2:2 separation pattern can be interpreted as indicating a situation where a difference in perception exists between a student group and an adult group or between generations.

[0067] Additionally, the interpretation generation unit (160) may be configured to generate guidance plans corresponding to the number of input perspectives and detailed patterns. For example, if there are two input perspectives, guidance plans including the need for joint intervention, the need for objectification of self-awareness, or the need for self-esteem improvement may be generated depending on the combination of level values ​​of the two perspectives. Additionally, if there are four input perspectives, guidance plans including self-esteem support, feedback provision, or self-reflection activities may be generated for patterns where a difference appears between self-awareness and other perspectives, and guidance plans including parent counseling, parent interviews, or home environment exploration may be generated for patterns where the parent perspective is different. Additionally, guidance plans including re-examination of the teacher's observation results or exploration of special situations in the classroom may be generated for patterns where the teacher perspective is different, and guidance plans including improvement of peer relationships or support for peer interaction may be generated for patterns where the peer perspective is different. In addition, for the pattern of full agreement, guidance plans for maintaining the current state or joint intervention by relevant parties can be generated, and for the 2:2 separation pattern, guidance plans for strengthening communication or promoting mutual understanding to alleviate the difference in perception between the student group and the adult group can be generated.

[0068] Additionally, the interpretation generation unit (160) may be configured to automatically generate educational interpretation sentences and guidance plan sentences based on the pattern classification results, variable types of student characteristics, and level values ​​by perspective. Here, the variable type may be distinguished according to whether the student characteristics are positive variables representing positive behavioral characteristics or negative variables representing negative behavioral characteristics, and the interpretation generation unit (160) may apply different interpretation logic depending on the variable type.

[0069] Additionally, the interpretation generation unit (160) may be configured to generate educational interpretation sentences and guidance plan sentences by referring to pre-set interpretation rules or sentence templates based on the pattern classification results, variable types of student characteristics, and level values ​​by perspective. The interpretation generation unit (160) may store interpretation sentence templates and guidance sentence templates that are pre-mapped by pattern, and generate a final sentence by selecting or combining parts of sentences corresponding to the corresponding pattern. Additionally, the interpretation generation unit (160) may be configured to generate educational interpretation sentences and guidance plan sentences using a natural language generation module or an artificial intelligence model.

[0070] Additionally, the interpretation generation unit (160) may be configured to provide comprehensive interpretation information and guidance plans regarding the characteristics of students by synthesizing multiple pattern information generated according to student characteristics.

[0071] In addition, the educational interpretation and guidance plan generated by the interpretation generation unit (160) can be provided in a later stage along with the results of the student characteristic analysis on a visualization screen so that they can be checked through a user terminal (200).

[0072] Meanwhile, the interpretation generation unit (160) may be configured to additionally generate educational interpretations and guidance plans corresponding to the high-risk state when the pattern classification unit (150) determines that the high-risk state is determined. For example, the interpretation generation unit (160) may generate interpretation information indicating that there is a high possibility that the student perceives their positive characteristics as excessively high compared to reality and that there is a significant discrepancy with other perspectives. Additionally, the interpretation generation unit (160) may be configured to generate guidance plans regarding the high-risk state that include intensive feedback for objectifying self-perception, joint intervention between guardians and teachers, continuous observation, or referral to expert counseling.

[0073] The visualization unit (170) may be configured to generate a visualization screen to visually represent the analysis results generated by the interpretation generation unit (160).

[0074] Here, a visualization screen may refer to a screen that presents analysis results in a visual form so that users can intuitively understand them. The visualization screen may be provided to enable teachers, guardians, or counselors to quickly identify the characteristics of a specific student and easily check the differences in evaluation results between perspectives.

[0075] The visualization unit (170) may be configured to generate a summary view. The summary view may be configured to allow an overview of the results of the overall student characteristic analysis for a specific student on one screen. For example, the summary view may be configured to visually display whether or not there is a match or degree of match between perspectives for each student characteristic, and to display the pattern classification results corresponding to each student characteristic together.

[0076] Additionally, the summary view can display agreement information indicating the proportion of views that match each other out of the total views. For example, if three out of four views are determined to have the same level value, the agreement percentage or ratio information can be displayed in the form of 75% or 3 / 4 match.

[0077] Additionally, the summary view may include icons or display elements for each perspective to visually identify whether each perspective matches, and may distinguish and display matching perspectives in a first visual display form and mismatching perspectives in a second visual display form. For example, matching perspectives may be displayed with a green icon and mismatching perspectives with an orange icon.

[0078] Additionally, the visualization unit (170) can be configured to switch to a corresponding detail view when the user selects a specific student characteristic or a specific display area of ​​the summary view.

[0079] Additionally, the visualization unit (170) may be configured to generate a detailed view. The detailed view may be a screen that specifically displays evaluation results by perspective for specific student characteristics. For example, the detailed view may display level values ​​or standard scores by perspective, such as self-report, peer-report, parent-report, and teacher-report, so as to be comparable, and may be configured to allow visual confirmation of differences in evaluation results from each perspective.

[0080] Additionally, the detailed view may include a visualization in the form of a matrix heatmap to intuitively grasp the distribution of level values ​​by perspective. For example, the matrix heatmap may be configured to allow for intuitive verification of the distribution between perspectives by arranging the perspectives of self-report, peer-report, parent-report, and teacher-report in the row direction, arranging the levels of low, medium, and high in the column direction, and placing indicator symbols at the level positions corresponding to each perspective.

[0081] Additionally, the detail view may include a visualization in the form of multiple horizontal gauges arranged in parallel to allow for side-by-side comparison of level values ​​for each perspective. If necessary, the detail view may further include a graph-type visualization that simultaneously displays level values ​​from multiple perspectives.

[0082] Additionally, the visualization unit (170) may be configured to display pattern information generated by the pattern classification unit (150) and educational interpretation sentences or guidance plan sentences generated by the interpretation generation unit (160) together on the visualization screen. For example, if the evaluations by peers, parents, and teachers regarding a specific student characteristic differ from the self-report, an interpretation sentence explaining the difference may be displayed together in the detailed view. Accordingly, the user can understand the meaning of the result and educational utilization methods together, going beyond simply checking the score results.

[0083] Additionally, the visualization unit (170) can be configured to allow the analysis results of multiple student characteristics to be selectively viewed in a single interface, and can be configured so that when a user selects a specific student characteristic or a specific perspective, a corresponding detailed view is displayed. Accordingly, the explorability and usability of the student characteristic analysis results can be improved.

[0084] Additionally, the visualization unit (170) may be configured to provide the generated visualization screen to a user terminal (200) used by a teacher, guardian, or counselor. For example, the visualization screen may be displayed on the user terminal (200) through a web-based interface or an application interface, and the user may view the student characteristic analysis results through the user terminal (200) and navigate while selectively switching between a summary view and a detailed view.

[0085] Meanwhile, the visualization unit (170) may be configured to provide an urgent indication on the visualization screen for student characteristics determined to be in a high-risk state by the pattern classification unit (150). For example, the visualization unit (170) may display student characteristics corresponding to a high-risk state in the summary view so that the user can immediately identify them through a warning icon, an accent color, or a separate urgent indication area. Additionally, the visualization unit (170) may be configured to display the difference in standard scores between self-reporting and other perspective reporting that serve as the basis for determining the high-risk state, the meaning of the corresponding pattern, educational interpretation information, and response guidance plan together in the detail view. Accordingly, the user can prioritize identifying student characteristics with a high risk level and quickly identify specific response directions.

[0086] According to a multi-perspective-based student characteristic analysis system (100) according to an embodiment of the present invention, which includes a data collection unit (110), a standard score calculation unit (120), a level judgment unit (130), a match judgment unit (140), a pattern classification unit (150), an interpretation generation unit (160), and / or a visualization unit (170) as described above, student characteristic evaluation data input from different perspectives, such as the student's own evaluation, peer evaluation, parent evaluation, and teacher evaluation, can be collected and processed in an integrated manner, thereby allowing the student's behavioral characteristics and emotional state to be understood more objectively and three-dimensionally compared to a single-perspective evaluation.

[0087] In addition, since evaluation data collected from each perspective is converted into standard scores based on common criteria and can automatically determine the consistency, degree of agreement, and patterns of discrepancy between perspectives based on this, the difference between students' self-perception and perception of others can be analyzed in a systematic and consistent manner.

[0088] In addition, educational interpretation information and guidance plans can be automatically generated based on classified patterns and provided through a user terminal (200), so that a teacher, guardian, or counselor can more efficiently understand the student's condition and establish an appropriate educational intervention or counseling direction.

[0089] In addition, since the analysis results can be provided as a visualization screen including summary and detailed views, evaluation information from multiple perspectives can be intuitively compared and explored, thereby improving the communicability and usability of the student characteristic analysis results.

[0090] Now, with reference to the attached drawings, a multi-perspective-based student characteristic analysis method (S100) according to one embodiment of the present invention will be described in detail.

[0091] FIG. 3 shows the overall flow of a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, FIG. 4 is a diagram showing an example of a criterion for determining a level value according to a standard score in a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, FIG. 5 is a diagram showing an example of generating educational interpretation information based on a matching judgment result in the case of two perspectives in a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention, and FIG. 6 is a diagram showing an example of generating a guidance plan based on a pattern classification result in the case of four perspectives in a multi-perspective-based student characteristic analysis method according to an embodiment of the present invention.

[0092] As illustrated in FIGS. 3 to 6, a multi-perspective-based student characteristic analysis method (S100) according to one embodiment of the present invention includes a collection step (S110), a standard score calculation step (S120), a level determination step (S130), a match determination step (S140), a pattern classification step (S150), an interpretation step (S160), and / or a visualization step (S170).

[0093] The collection step (S110) may consist of collecting data on student characteristics from at least two perspectives among self-report, peer-report, parent-report, or teacher-report.

[0094] Here, "student characteristics" may refer to traits that indicate a student's disposition or behavioral patterns, such as behavioral characteristics, emotional state, social interaction characteristics, or learning participation characteristics. For example, student characteristics may include, but are not limited to, prosociality, aggression, emotional stability, adaptability to peer relationships, learning participation, or adaptability to school life.

[0095] Furthermore, perspectives for evaluating student characteristics may include self-reporting, peer reporting, parent reporting, or teacher reporting. Self-reporting may be a method in which the student themselves evaluates their own behavioral characteristics or emotional state, while peer reporting may be a method in which peers belonging to the same class or group evaluate the behavioral characteristics of the subject student. Additionally, parent reporting may be a method in which a guardian evaluates the student's behavioral characteristics observed in the home environment, and teacher reporting may be a method in which a teacher evaluates the student's behavioral characteristics observed during classroom situations or school life.

[0096] The collection step (S110) can be performed by providing a survey or evaluation form consisting of multiple evaluation items and collecting evaluation data by having evaluators of each perspective input responses to the evaluation items. Each evaluation item may consist of a question about a behavior or attitude related to a specific student characteristic, and the evaluator may respond to the question by selecting one of multiple score scales.

[0097] Meanwhile, peer reporting may be performed using a peer nomination method as an embodiment. The peer nomination method may be a method in which multiple students within a class nominate a single student or multiple students who are judged to correspond to a specific behavioral characteristic. For example, if a specific student characteristic is prosociality, students within the class may select a single student or multiple students who match that characteristic. The collection step (S110) may be configured to generate a student characteristic score by aggregating the number of times each student is nominated and converting the nomination frequency into a ratio relative to the total number of respondents or a normalized value. Accordingly, peer reporting can be implemented not only as a simple survey response method but also as a sociometry-based quantitative evaluation method that reflects mutual perception within the group.

[0098] Additionally, the collection step (S110) can be performed through a user terminal (200) used by a student, peer student, guardian, or teacher. Evaluation questions are provided through the user terminal (200), and each evaluator can input a response to the corresponding question. The input response data is transmitted via a network to a multi-perspective-based student characteristic analysis system (100) and can be used for analysis in subsequent steps.

[0099] In addition, the collection stage (S110) can be configured to store evaluation data collected from different perspectives regarding the same student characteristics in conjunction with a single student identification information, thereby enabling comparative analysis of evaluation results between perspectives in subsequent stages.

[0100] Additionally, in the collection stage (S110), student characteristic scores may be generated by aggregating response values ​​for multiple evaluation items or calculating average values ​​to calculate evaluation data collected from each perspective as scores for each student characteristic. In the case where a peer nomination method is applied, the collection stage (S110) may be configured to generate student characteristic scores by aggregating the number of times a specific student is nominated by peer students and converting the nomination frequency into a ratio relative to the total number of respondents or a normalized value.

[0101] The standard score calculation step (S120) may consist of a step of converting the student characteristic scores generated in the collection step (S110) into standard scores based on class standards.

[0102] Here, the student characteristic score may be a score calculated based on evaluation data collected from each perspective, or a raw score generated by summing or averaging the response values ​​for multiple evaluation items corresponding to a specific student characteristic.

[0103] The standard score calculation step (S120) can convert the evaluation results by perspective into a form comparable to a common standard by normalizing the student characteristic scores based on the distribution of multiple students belonging to the same class. Accordingly, the standard score calculation step (S120) can generate a standard score based on the class standard to indicate where the student characteristic score of a specific student is located relatively within the corresponding class group.

[0104] For example, the standard score calculation step (S120) can calculate the mean value and variance value or standard deviation of student characteristic scores collected for students belonging to the same class, and calculate the standard score by normalizing the difference between a specific student's student characteristic score and the mean value using the variance value or standard deviation. Accordingly, evaluation results collected from different perspectives can be compared on the same reference axis. At this time, the standard score can be calculated using the T-score method, and in this case, it can be calculated according to the formula T = 50 + 10 × (X - M) / SD based on a specific student's student characteristic score X, the class mean value M, and the standard deviation SD. However, the method of calculating the standard score is not limited to this, and various normalization methods may be applied.

[0105] In addition, the standard score calculation step (S120) can calculate standard scores independently for each student characteristic. Student characteristic scores regarding prosociality can be converted into standard scores based on the in-class score distribution corresponding to prosociality, and student characteristic scores regarding aggression can be separately converted into standard scores based on the in-class score distribution corresponding to aggression. Accordingly, the relative level of each characteristic can be accurately reflected without confusion between different student characteristics.

[0106] In addition, the standard score calculation step (S120) can be configured to enable direct comparison between self-reported scores, peer-reported scores, parent-reported scores, and teacher-reported scores by processing student characteristic scores collected by perspective according to the same conversion criteria. For example, when the self-reported score and the teacher-reported score for the same student's same characteristic are each converted into standard scores based on class standards, the difference between the two scores can be used to compare and analyze the difference between the student's own perception and the teacher's perception.

[0107] The level determination step (S130) may consist of a step for determining the level value of student characteristics based on the standard score calculated in the standard score calculation step (S120).

[0108] Here, the level value may be a value indicating the relative level at which a specific student characteristic is located within a class group, and may be a value classified into one of multiple level intervals such as low, average, or high.

[0109] The level determination step (S130) can determine the level value of a student's characteristics by comparing the standard score calculated in the standard score calculation step (S120) with a plurality of preset standard intervals. For example, it may be configured to determine a low level if the standard score is less than the first standard value, determine an average level if it is greater than or equal to the first standard value and less than or equal to the second standard value, and determine a high level if it exceeds the second standard value. However, the number of level intervals or the method of setting standard values ​​are not limited to this and may be set in various ways depending on the educational setting or the purpose of evaluation.

[0110] In addition, the level determination step (S130) can determine level values ​​independently for each student characteristic. That is, separate level values ​​can be determined based on standard scores for each different student characteristic, such as prosociality, aggression, emotional stability, or learning participation. Accordingly, level values ​​corresponding to each of multiple student characteristics can be generated for a specific student.

[0111] In addition, the level determination step (S130) can independently determine level values ​​for standard scores calculated from different perspectives. For example, for the same student characteristics of the same student, level values ​​corresponding to the standard score based on self-report, the standard score based on peer report, the standard score based on parent report, and the standard score based on teacher report can be determined. Accordingly, the level of student characteristics perceived from each perspective can be identified individually.

[0112] Additionally, the level determination step (S130) may store or provide the determined level value so that it can be used for comparative analysis between perspectives in a subsequent step. The level value can be used as an input value to determine whether there is agreement or inconsistency patterns between perspectives in a subsequent step. Through this, differences in perception between different perspectives regarding the same student's characteristics can be systematically analyzed.

[0113] Additionally, the level determination step (S130) may be configured to determine the standard score as a level value of low, normal, or high according to a preset score range. For example, if the standard score is less than 40 points, it may be determined as low; if it is 40 points or more and 60 points or less, it may be determined as normal; and if it exceeds 60 points, it may be determined as high. Accordingly, since the standard scores calculated from different perspectives can be converted into a common level system and compared, it is easier to determine whether there is agreement or inconsistency between perspectives in subsequent steps.

[0114] The agreement determination step (S140) may be composed of a step for determining whether there is an agreement by comparing the level values ​​for each perspective calculated in the level determination step (S130).

[0115] The agreement judgment step (S140) may be configured to compare the level values ​​of each perspective with one another and determine that they are in agreement if they are judged to be at the same level, and inconsistent if they are judged to be at different levels. For example, if there are two input perspectives, it may be determined that they are in agreement if the level values ​​of the two perspectives are the same, and inconsistent if the level values ​​of the two perspectives are different. In addition, if there are four input perspectives, it may be determined that all perspectives are in agreement if the same level value is produced in self-report, peer report, parent report, and teacher report for the same student characteristic, and it may be determined that there is inconsistency between perspectives if different level values ​​are produced in some perspectives.

[0116] Additionally, the agreement judgment step (S140) may be configured to determine whether there is an agreement based on the degree of difference between level values ​​for each perspective. If the difference between level values ​​calculated from different perspectives is below a preset allowable range, they may be considered substantially the same level and judged as an agreement; if they exceed the allowable range, they may be judged as an inconsistency. Accordingly, it is possible to prevent excessive judgment of inconsistency regarding minute differences between evaluation perspectives.

[0117] Additionally, the agreement judgment step (S140) may be configured to calculate the degree of agreement between perspectives in order to comprehensively determine whether there is an agreement between multiple perspectives. For example, if there are two input perspectives, the degree of agreement may be calculated based on whether there is an agreement between the two perspectives, and if there are three or more input perspectives, the degree of agreement may be calculated by dividing the number of perspectives having the same level value by the total number of perspectives, or by assigning weights based on the difference in levels between perspectives.

[0118] Additionally, the agreement judgment step (S140) can generate judgment results for each student characteristic, and accordingly, for a specific student, the agreement status or degree of agreement between perspectives corresponding to each of multiple student characteristics can be calculated. The results calculated in this way can be used in subsequent steps to classify patterns of inconsistency between perspectives or to visualize the results of the student characteristic analysis.

[0119] The pattern classification step (S150) may consist of a step of classifying the types of mismatches calculated in the match judgment step (S140) into patterns according to a combination of level values ​​for each perspective.

[0120] Here, the discrepancy type may refer to cases where level values ​​calculated from different perspectives differ for the same student's characteristics. For example, if the level is judged as high in self-reporting but average in teacher-reporting, or low in peer-reporting, it may be identified as a discrepancy type when level values ​​differ between perspectives.

[0121] The pattern classification step (S150) can classify the type of mismatch into one of a plurality of predefined patterns based on the combination of level values ​​for each perspective. Additionally, the pattern classification step (S150) can classify it as a matching pattern if all perspectives are determined to have the same level value in the matching judgment step (S140), and as a partial mismatch pattern if a difference occurs only in some perspectives.

[0122] Additionally, the pattern classification step (S150) may be configured to apply different detailed pattern systems depending on the number of input perspectives. For example, if there are two input perspectives, it may be classified as a matching pattern if the level values ​​of the two perspectives are the same, and as a mismatched pattern if the level values ​​of the two perspectives are different. Additionally, if there are four input perspectives, it may be classified into multiple patterns including a self-report only low pattern, a self-report only high pattern, a parent-report only high pattern, a parent-report only low pattern, a teacher-report only high pattern, a teacher-report only low pattern, a peer-report only different pattern, an all-matching pattern where the level values ​​of all perspectives are the same, or a 2:2 separation pattern where two perspectives and two different perspectives are distinguished from each other, depending on the combination of level values ​​for each perspective.

[0123] Accordingly, in the pattern classification step (S150), regarding the characteristics of the same student, it is possible to more specifically identify which perspectives cause differences in perception depending on the number of input perspectives, and in subsequent steps, more precise educational interpretation and guidance plans can be generated.

[0124] Additionally, the pattern classification step (S150) can generate classified pattern information according to student characteristics, and pattern information corresponding to each of multiple student characteristics can be generated for a specific student. The pattern information generated in this way can be used as basic information for interpreting or visually representing the results of the student characteristic analysis in subsequent steps.

[0125] Meanwhile, the pattern classification step (S150) may be configured to additionally determine whether a high-risk state exists by comparing the difference in standard scores between self-report and other perspective reports when a pattern is identified where the student characteristic is a positive variable and only self-report is high. For example, if the standard score of self-report is determined to be at a high level and the difference in standard scores with peer reports, parent reports, and teacher reports is greater than or equal to a preset difference, the pattern classification step (S150) may classify the student characteristic in question as a high-risk state with a large discrepancy between self-perception and perception by others. Accordingly, the pattern classification step (S150) can go beyond simple discrepancy pattern classification and separately identify situations where the degree of bias in the student's self-perception is significant.

[0126] The interpretation step (S160) may consist of a step for generating educational interpretations and guidance plans based on the patterns classified in the pattern classification step (S150).

[0127] Here, educational interpretation may refer to information explaining the meaning of evaluation results presented from different perspectives regarding specific student characteristics, and guidance plan may refer to educational measures or guidance methods that teachers or guardians can refer to in order to improve or support the student's behavioral characteristics or emotional state based on the educational interpretation.

[0128] The interpretation step (S160) may be configured to generate different educational interpretations depending on the number of input perspectives and the patterns classified in the pattern classification step (S150). For example, if there are two input perspectives, an educational interpretation may be generated based on combinations formed when the level value of the first perspective and the level value of the second perspective are each determined to be low, normal, or high. That is, the interpretation step (S160) may be configured to derive a total of nine combination patterns when three level values ​​are possible for each of the two perspectives, and to generate judgment results, interpretation directions, and guidance plans corresponding to each combination pattern. For example, if the level values ​​of both perspectives are low, it may be determined as a matching pattern and interpreted as a situation requiring joint intervention; and if the level value of the first perspective is high and the level value of the second perspective is low, it may be determined as a mismatched pattern and interpreted as a situation requiring more objectification of self-awareness. In addition, if the level value of the first perspective is low and the level value of the second perspective is high, it is determined as a discrepancy pattern and can be interpreted as a situation requiring self-esteem improvement.

[0129] In addition, if there are four input perspectives, a total of 81 combination patterns are derived where three level values ​​are possible for each, and the system can be configured to generate judgment results, interpretation directions, and guidance plans corresponding to each combination pattern. For example, a pattern of "low self-reporting only" or "high self-reporting only" can be interpreted as indicating a difference between a student's self-perception and other perspectives, while a pattern of "high parent-reporting only" or "low parent-reporting only" can be interpreted as suggesting a difference in perception between home and school or specific situations in the home environment. Furthermore, a pattern of "high teacher-reporting only" or "low teacher-reporting only" can be interpreted as suggesting specific situations based on the classroom environment or the teacher's observation context, and a pattern of "different peer-reporting only" can be interpreted as a situation reflecting the characteristics of interaction in peer relationships. Additionally, a pattern of unanimous agreement can be interpreted as a case where the same state perception is formed from multiple perspectives, and a 2:2 separation pattern can be interpreted as indicating a situation where a difference in perception exists between a student group and an adult group or between generations.

[0130] Additionally, the interpretation step (S160) may be configured to generate guidance plans corresponding to the number of input perspectives and detailed patterns. For example, if there are two input perspectives, guidance plans including the need for joint intervention, the need for objectification of self-awareness, or the need for self-esteem improvement may be generated depending on the combination of level values ​​of the two perspectives. Additionally, if there are four input perspectives, guidance plans including self-esteem support, feedback provision, or self-reflection activities may be generated for patterns where a difference appears between self-awareness and other perspectives, and guidance plans including parent counseling, parent interviews, or home environment exploration may be generated for patterns where the parent perspective is different. Additionally, guidance plans including re-examination of the teacher's observation results or exploration of special situations within the classroom may be generated for patterns where the teacher perspective is different, and guidance plans including improvement of peer relationships or support for peer interaction may be generated for patterns where the peer perspective is different. In addition, for the pattern of full agreement, guidance plans for maintaining the current state or joint intervention by relevant parties can be generated, and for the 2:2 separation pattern, guidance plans for strengthening communication or promoting mutual understanding to alleviate the difference in perception between the student group and the adult group can be generated.

[0131] Additionally, the interpretation step (S160) can be configured to automatically generate educational interpretation sentences and guidance plan sentences based on the patterns classified in the pattern classification step (S150) and the variable types of student characteristics. Here, the variable types can be distinguished based on whether the student characteristics are positive variables representing positive behavioral characteristics or negative variables representing negative behavioral characteristics, and the interpretation step (S160) can apply different interpretation logic depending on the variable types.

[0132] Additionally, the interpretation step (S160) may be configured to generate educational interpretation sentences and guidance plan sentences by referring to pre-set interpretation rules or sentence templates based on the pattern classification results, variable types of student characteristics, and level values ​​by perspective. The interpretation step (S160) may store interpretation sentence templates and guidance sentence templates that are pre-mapped by pattern, and generate a final sentence by selecting or combining parts of sentences corresponding to the corresponding pattern. Additionally, educational interpretation sentences and guidance plan sentences may be generated using a natural language generation module or an artificial intelligence model.

[0133] In addition, the interpretation stage (S160) can provide a comprehensive set of multiple educational interpretations and guidance plans based on pattern information generated for each student characteristic. Accordingly, interpretation information and guidance plans corresponding to each of the multiple student characteristics can be generated for a specific student, thereby enabling teachers or guardians to more systematically understand the student's behavioral characteristics and emotional state and provide appropriate educational support.

[0134] In addition, the educational interpretation and guidance plan generated in the interpretation stage (S160) can be included in a visualization screen along with the results of the student characteristic analysis in a subsequent stage and provided so that it can be checked through a user terminal (200).

[0135] Meanwhile, the interpretation stage (S160) may be composed of a step for generating additional educational interpretations and guidance plans corresponding to the high-risk state when the pattern classification stage (S150) determines a high-risk state. For example, the interpretation stage (S160) may generate interpretation information indicating that there is a high possibility that the student perceives their positive characteristics as excessively high compared to reality and that there is a significant discrepancy with other perspectives. Additionally, regarding the high-risk state, the interpretation stage (S160) may generate guidance plans that include intensive feedback for objectifying self-perception, joint intervention between guardians and teachers, continuous observation, or referral to expert counseling.

[0136] The visualization step (S170) may consist of a step of creating a visualization screen to visually represent the analysis results generated in the interpretation step (S160).

[0137] Here, a visualization screen may refer to a screen that presents analysis results in a visual form so that users can intuitively understand them. The visualization screen may be provided to enable teachers, guardians, or counselors to quickly identify the characteristics of a specific student and easily check the differences in evaluation results between perspectives.

[0138] In the visualization step (S170), a summary view can be generated. The summary view can be configured to allow an overview of the overall student characteristic analysis results for a specific student on a single screen.

[0139] The summary view can be configured to visually indicate whether or not there is agreement or the degree of agreement between perspectives by student characteristics, and to display the pattern classification results corresponding to each student characteristic together.

[0140] Additionally, the summary view can display agreement information indicating the proportion of views that match each other out of the total views. For example, if three out of four views are determined to have the same level value, the agreement percentage or ratio information can be displayed in the form of 75% or 3 / 4 match.

[0141] Additionally, the summary view may include icons or display elements for each perspective to visually identify whether each perspective matches, and may distinguish and display matching perspectives in a first visual display form and mismatching perspectives in a second visual display form. For example, matching perspectives may be displayed with a green icon and mismatching perspectives with an orange icon.

[0142] In addition, it can be configured to switch to a corresponding detail view when the user selects a specific student characteristic or a specific display area in the summary view.

[0143] Additionally, a detailed view can be generated in the visualization step (S170). The detailed view may be a screen that displays evaluation results by perspective regarding specific student characteristics in more detail. The detailed view may display level values ​​or standard scores by perspective, such as self-report, peer report, parent report, and teacher report, so that they can be compared, and may be configured to allow visual confirmation of differences in evaluation results from each perspective.

[0144] Additionally, the detailed view may include a matrix heatmap visualization to intuitively grasp the distribution of level values ​​by perspective. The matrix heatmap can be configured to allow for intuitive verification of the distribution between perspectives by arranging the perspectives of self-report, peer-report, parent-report, and teacher-report in the row direction, arranging the levels of low, medium, and high in the column direction, and placing indicator symbols at the level positions corresponding to each perspective.

[0145] Additionally, the detail view may include a visualization in the form of multiple horizontal gauges arranged in parallel to allow for side-by-side comparison of level values ​​for each perspective. If necessary, the detail view may further include a graph-type visualization that simultaneously displays level values ​​from multiple perspectives.

[0146] Additionally, the visualization step (S170) may be configured to display pattern information generated in the pattern classification step (S150) and educational interpretation sentences or guidance plan sentences generated in the interpretation step (S160) together on the visualization screen. For example, if the evaluations by peers, parents, and teachers regarding a specific student characteristic differ from the self-report, an interpretation sentence explaining the difference may be displayed together in the detailed view. Accordingly, the user can go beyond simply checking the score results and understand the meaning of the results and methods for educational utilization.

[0147] Additionally, the visualization step (S170) can be configured to allow selective viewing of analysis results for multiple student characteristics in a single interface, and can be configured so that when a user selects a specific student characteristic or a specific perspective, a corresponding detailed view is displayed. Accordingly, the exploreability and usability of the student characteristic analysis results can be improved.

[0148] Additionally, the visualization screen generated in the visualization step (S170) may be provided to a user terminal (200) used by a teacher, guardian, or counselor. For example, the visualization screen may be displayed on the user terminal (200) through a web-based interface or an application interface, and the user may view the student characteristic analysis results through the user terminal (200) and navigate while selectively switching between the summary view and the detailed view.

[0149] Meanwhile, the visualization step (S170) may be composed of a step that provides an urgent indication on the visualization screen for student characteristics determined to be in a high-risk state in the pattern classification step (S150). For example, the visualization step (S170) may display student characteristics corresponding to a high-risk state in the summary view through a warning icon, accent color, or a separate urgent indication area so that the user can immediately identify them. Additionally, the visualization step (S170) may display the difference in standard scores between self-reports and reports from other perspectives that serve as the basis for determining the high-risk state, the meaning of the corresponding pattern, educational interpretation information, and response guidance plans together in the detailed view. Accordingly, the user can prioritize identifying student characteristics with a high risk level and quickly grasp specific response directions.

[0150] According to a multi-perspective-based student characteristic analysis method (S100) according to an embodiment of the present invention, which includes a collection step (S110), a standard score calculation step (S120), a level determination step (S130), a match determination step (S140), a pattern classification step (S150), an interpretation step (S160), and / or a visualization step (S170) as described above, student characteristic evaluation data collected from different perspectives, such as the student's own evaluation, peer evaluation, parent evaluation, and teacher evaluation, can be analyzed step by step, thereby allowing the student's behavioral characteristics and emotional state to be identified from various angles.

[0151] In addition, since the evaluation results of each perspective are converted into standard scores based on class standards, level values ​​are determined, and patterns of agreement and disagreement between perspectives can be classified, the difference between students' self-perception and perception of others can be analyzed quantitatively and systematically.

[0152] In addition, since educational interpretations and guidance plans can be generated based on classified patterns, the meaning of the student characteristic assessment results can be derived more clearly, and useful information can be provided to teachers or counselors to establish directions for educational support appropriate to each student's situation.

[0153] In addition, since analysis results can be provided as a visualization screen including summary and detailed views, evaluation results and pattern information from multiple perspectives can be intuitively compared and understood, thereby enhancing the usability of the student characteristic analysis results.

[0154] Although all components constituting the embodiments of the present invention have been described above as being combined or operating together, the present invention is not necessarily limited to such embodiments. That is, within the scope of the purpose of the present invention, all components may be selectively combined and operated in one or more ways.

[0155] Furthermore, terms such as "include," "compose," or "have" described above, unless specifically stated otherwise, mean that the relevant component may be inherent; therefore, they should be interpreted as allowing for the inclusion of additional components rather than excluding them. All terms, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains, unless otherwise defined. Commonly used terms, such as those defined in advance, should be interpreted in accordance with their meaning in the context of the relevant technology and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the present invention.

[0156] Furthermore, the above description is merely an illustrative explanation of the technical concept of the present invention, and those skilled in the art to which the present invention pertains will be able to make various modifications and variations within the scope of the essential characteristics of the present invention.

[0157] Accordingly, the embodiments disclosed in this invention are intended to illustrate, not limit, the technical concept of the invention, and the scope of the technical concept of the invention is not limited by these embodiments. The scope of protection of this invention shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of this invention. Explanation of the symbols

[0158] 100: Multi-perspective based student characteristics analysis system 110: Data Collection Unit 120 : Standard Score Calculation Unit 130 : Level Assessment Unit 140 : Match determination unit 150 : Pattern Classification Section 160 : Interpretation generation unit 170 : Visualization Section 200 : User terminal S100: Multi-Perspective Based Student Characteristics Analysis Method S110: Collection stage S120: Standard score calculation stage S130: Level Assessment Stage S140: Match determination stage S150: Pattern classification stage S160: Interpretation stage S170: Visualization stage

Claims

Claim 1 A data collection unit that collects evaluation data from multiple perspectives, including self-report, peer report, and teacher report, regarding the student characteristics of the same student and generates student characteristic scores by perspective; a standard score calculation unit that converts the student characteristic scores generated by the data collection unit into standard scores based on class standards; a level determination unit that determines the level value of the student characteristics based on the standard scores calculated by the standard score calculation unit; a match determination unit that determines whether there is a match by comparing the level values ​​by perspective determined by the level determination unit; a pattern classification unit that classifies into patterns corresponding to combinations of the level values ​​by perspective based on whether there is a match; and an interpretation generation unit that generates educational interpretations and guidance plans according to at least one of the standard scores, the level values, the match status, or the patterns.The system includes a visualization unit that generates a visualization screen displaying at least one of the above standard score, above level value, above match status, above educational interpretation, or above guidance plan; the data collection unit performs the above peer report in a peer nomination manner, aggregates the number of nominations by peer students for a specific student, and generates the above student characteristic score by calculating the nomination frequency as a ratio relative to the total number of respondents in the corresponding class; the level determination unit compares the above standard score by perspective with a preset first reference value and a second reference value, and determines the above level value by perspective as low if the standard score is less than the first reference value, average if it is greater than or equal to the first reference value and less than or equal to the second reference value, and high if it exceeds the second reference value; the pattern classification unit classifies the above pattern into a plurality of patterns including a pattern of complete agreement in which the level values ​​of all the plurality of perspectives are the same, and a pattern of partial disagreement in which a difference occurs only in some of the plurality of perspectives; the interpretation generation unit distinguishes the variable type of the above student characteristic according to whether the above student characteristic is a positive variable representing positive behavioral characteristics or a negative variable representing negative behavioral characteristics; and the A multi-perspective-based student characteristic analysis system characterized by referencing a pre-established interpretation rule or sentence template based on the pattern classified in the pattern classification unit, the variable type, and the level value by perspective, and generating the educational interpretation and the guidance plan by selecting or combining parts of sentences corresponding to the pattern from among the interpretation sentence templates and guidance sentence templates pre-mapped by pattern. Claim 2 A multi-perspective-based student characteristic analysis system according to claim 1, characterized in that the first reference value is 40 points and the second reference value is 60 points. Claim 3 A multi-perspective-based student characteristic analysis system according to claim 1, wherein the plurality of perspectives comprises four perspectives: self-report, peer-report, parent-report, and teacher-report, and the pattern classification unit classifies the pattern into a plurality of patterns including a self-report only low pattern, self-report only high pattern, parent-report only high pattern, parent-report only low pattern, teacher-report only high pattern, teacher-report only low pattern, peer-report only different pattern, all consistent pattern, and 2:2 separated pattern. Claim 4 A multi-perspective-based student characteristic analysis system according to claim 1, wherein the visualization unit displays a visualization screen including at least one of a summary view or a detailed view, the summary view displays visual identification elements indicating a percentage of agreement and whether there is an agreement by perspective, and the detailed view displays a matrix heatmap, a horizontal gauge, the educational interpretation, and the guidance plan. Claim 5 A collection step for collecting evaluation data from multiple perspectives, including self-report, peer report, and teacher report, regarding the student characteristics of the same student to generate student characteristic scores for each perspective; a standard score calculation step for converting the student characteristic scores generated in the collection step into standard scores based on class standards; a level determination step for determining the level value of the student characteristics based on the standard scores calculated in the standard score calculation step; a consistency determination step for determining whether there is consistency between perspectives by comparing the level values ​​for each perspective determined in the level determination step; a pattern classification step for classifying patterns based on the consistency determined in the consistency determination step and the combination of the level values ​​for each perspective; and an interpretation step for generating educational interpretations and guidance plans according to the patterns classified in the pattern classification step.The method includes a visualization step for generating a visualization screen that displays the results generated in the interpretation step above, wherein the collection step performs the peer report using a peer nomination method, aggregates the number of nominations by peer students for a specific student, and calculates the nomination frequency as a ratio relative to the total number of respondents in the class to generate the student characteristic score, wherein the level determination step compares the standard score for each perspective with a pre-set first reference value and second reference value, and determines the level value for each perspective as low if the standard score is less than the first reference value, average if it is greater than or equal to the first reference value and less than or equal to the second reference value, and high if it exceeds the second reference value, wherein the pattern classification step classifies the pattern into multiple patterns including a pattern of unanimous agreement in which the level values ​​of all multiple perspectives are the same, and a pattern of partial disagreement in which a difference occurs only in some of the multiple perspectives, wherein the interpretation step distinguishes the variable type of the student characteristic according to whether the student characteristic is a positive variable representing positive behavioral characteristics or a negative variable representing negative behavioral characteristics, and the pattern, variable type, and perspective-specific A multi-perspective-based student characteristic analysis method characterized by referencing a preset interpretation rule or sentence template based on a level value, and generating the educational interpretation and the guidance plan by selecting or combining parts of sentences corresponding to the corresponding pattern among the interpretation sentence templates and guidance sentence templates pre-mapped for each pattern.

Citation Information

Patent Citations

  • Computer-Aided Evaluation System

    JP2024509912A

  • Method and apparatus for providing grade information

    KR1020050095667A

  • System, server and method for customized interview based on artificial intelligence

    KR1020250006664A

  • Class operation system to reduce teacher teaching and evaluation burden in digital classroom environment

    KR1020250177462A