Educational assessment system using artificial intelligence (AI) technology

Abstract

This system provides an educational assessment method that efficiently evaluates children's intellectual abilities and developmental status based on their interaction history with AI, and then provides feedback on the results to parents and educators. [Solution] The system comprises a storage means for saving dialogue history, a summary generation means for generating a summary of the dialogue history after a predetermined period of time has elapsed, an analysis means for integrating and analyzing new dialogue history with the saved summaries, and an evaluation generation means for generating an educational evaluation based on the analysis results. The analysis means performs the analysis based on rubrics related to language patterns, knowledge, and interests. Rubrics can be set by parents or teachers. This makes it possible to efficiently conduct long-term learning evaluations and realize multifaceted educational evaluations.

Description

Technical Field

[0001] The present invention relates to an educational evaluation system using artificial intelligence (AI) technology, and particularly to a system that analyzes the interaction history between children and AI to evaluate children's intellectual abilities and developmental status, and provides feedback on the results to guardians and educators.

Background Art

[0002] In recent years, the use of interactive generative AI such as chatbots has been advancing in the educational field. These AI systems support children's learning and provide educational assistance by answering various questions.

[0003] Conventional systems lacked the function of analyzing the interaction content between children and AI from an educational perspective and providing feedback on the results to guardians and teachers. There was also a problem that rich information regarding children's intellectual abilities and developmental status included in the interaction history could not be utilized.

[0004] Furthermore, in existing evaluation systems, the analysis of language patterns and the understanding of knowledge structures were fragmentary, making it difficult to conduct comprehensive educational evaluations. There was also a technical problem that as the learning period became longer, the interaction history became longer, making it difficult to input all the history into the AI.

Summary of the Invention

Problems to be Solved by the Invention

[0005] The present invention aims to solve the following problems. (1) To provide a means for guardians and teachers to appropriately grasp the interaction content between children and AI (2) To provide a technical means that enables efficient analysis and evaluation even when the interaction history becomes long (3) To realize an objective evaluation of children's intellectual abilities and developmental status from the interaction history (4) To effectively visualize the evaluation results and utilize them for specific educational support

Means for Solving the Problems

[0006] To solve the above problems, the present invention employs the following technical means.

[0007] The educational evaluation system of the present invention comprises a storage means for storing dialogue history, a summary generation means for generating a summary of the dialogue history after a predetermined period of time has elapsed, an analysis means for integrating and analyzing new dialogue history and the stored summary, and an evaluation generation means for generating an educational evaluation based on the analysis results.

[0008] Furthermore, in this invention, the analytical means performs the analysis based on a rubric. The rubric includes the following elements:

[0009] (a) Language patterns - Multilayered analysis of text complexity - Quantitative evaluation of conceptual understanding - Time-series analysis of dialogue development skills

[0010] (b) knowledge - Extraction and classification of knowledge elements by domain - Analysis of the hierarchical structure of knowledge and assessment of understanding - Detection of misconceptions and estimation of learning progress

[0011] (c) Interests - Measuring topic frequency and duration - Tracking inquiry-based behavior and evaluating learning motivation [Effects of the Invention]

[0012] The present invention provides the following effects: (1) Realization of multifaceted educational evaluation (2) Streamlining long-term learning assessment (3) Objective and comprehensive competency assessment (4) Providing practical educational support (5) Ensuring the reliability of the evaluation [Modes for carrying out the invention]

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0014] Figure 1 is a block diagram showing the overall configuration of the education evaluation system of the present invention. System 100 includes a dialogue history storage unit 110, a summary generation unit 120, an analysis unit 130, and an evaluation generation unit 140.

[0015] The dialogue history storage unit 110 stores the dialogue history between a child and AI in a database. The stored dialogue history includes, in addition to text data, time information of the dialogue, context information, and the like.

[0016] The summary generation unit 120 generates a summary of the dialogue history for that period after a predetermined period (for example, one month) has elapsed. The summary generation includes the following steps: (1) Extraction of important keywords (2) Context analysis (3) Generation of summary sentences

[0017] The analysis unit 130 integrates and analyzes the new dialogue history and the stored summary. The analysis is carried out based on a rubric, and evaluations are made from the perspectives of language patterns, knowledge, and interests.

[0018] The evaluation generation unit 140 generates an education evaluation report based on the analysis results. The report includes quantitative evaluations, time-series changes, improvement proposals, and the like.

Example

[0019] Example 1: Application in school education Figure 2 shows an example of application in an elementary school. Teachers can view the evaluation results of all the students in the class in the form of a dashboard.

[0020] Example 2: Application in home learning Figure 3 shows an example of application at home. Guardians can view the learning progress of their children over time.

Industrial Applicability

[0021] This invention can be used in a wide range of fields, including educational institutions, home education support, and educational consulting. [Brief explanation of the drawing]

[0022] [Figure 1] This is a block diagram showing the overall configuration of the educational evaluation system of the present invention, illustrating the detailed internal structure and data flow of each component: the dialogue history storage unit (110), the summary generation unit (120), the analysis unit (130), and the evaluation generation unit (140). The dialogue history storage unit (110) includes a dialogue data DB, a summary data DB, and an evaluation result DB; the summary generation unit (120) is equipped with processing modules for keyword extraction, contextual analysis, and summary text generation; the analysis unit (130) has functions for language pattern analysis, knowledge analysis, and interest analysis; and the evaluation generation unit (140) has output functions for evaluation reports, improvement suggestions, and progress management. [Figure 2] This diagram shows the processing flow in the educational evaluation system of the present invention. It shows the process starting with the acquisition of dialogue data, followed by a period determination, and then branching processing based on the determination that a predetermined period has elapsed. When the predetermined period has elapsed, a summary generation process is executed, and after an integrated analysis process of the results and new data, the evaluation results are generated. The processing path when the period has not elapsed is also shown by a dashed arrow. [Figure 3] This figure shows an example of the evaluation UI display in the educational evaluation system of the present invention. The learning evaluation dashboard shows a layout with basic information (learning period, total number of conversations, average conversation time, etc.) on the left, ability evaluation graphs (language ability, comprehension, expression ability, etc. displayed as radar charts) in the center, and improvement suggestions on the right. This illustrates a method for visually understanding evaluation results and presenting specific improvement suggestions. [Explanation of Symbols]

[0023] 100 Educational Evaluation Systems 110 Dialogue History Storage Unit - Dialogue Data Database - Summary Data DB - Evaluation Results Database 120 summary generator - Keyword extraction -Contextual analysis -Summary sentence generation 130 Analysis Department - Language pattern analysis -Knowledge analysis - Interest Analysis 140 Evaluation Generation Unit - Evaluation Report -Improvement suggestions - Progress management

Claims

1. This system analyzes the history of conversations between children and AI to perform educational evaluations. A storage means for saving the dialogue history, A summary generation means that generates a summary of the dialogue history after a predetermined period of time has elapsed, An analytical method that integrates and analyzes new dialogue history and saved summaries, An evaluation generation means for generating educational evaluations based on analysis results, An educational evaluation system characterized by having the following features.

2. The educational evaluation system according to claim 1, characterized in that the analysis means performs the analysis based on a pre-set rubric.

3. The aforementioned rubric is, Analysis items for language patterns, Knowledge analysis items, Analysis items for interests and concerns, The educational evaluation system according to claim 2, characterized by including the following:

4. The analysis items for the aforementioned language patterns are: Multilayered analysis of text complexity, Quantitative evaluation of conceptual understanding, A time-series analysis of dialogue development skills, The educational evaluation system according to claim 3, characterized by including the following:

5. The analysis items of the aforementioned knowledge are: Extraction and classification of knowledge elements by domain, Analysis of the hierarchical structure of knowledge and assessment of understanding, Detection of misconceptions and estimation of learning progress, The educational evaluation system according to claim 3 or 4, characterized by including the following:

6. The aforementioned analysis items for interests are: Measurement of topic frequency and duration, Tracking inquiry behavior and evaluating learning motivation, An educational evaluation system according to any one of claims 3 to 5, characterized in that it includes the following:

7. The summary generation means is Extract important keywords from the conversation history, By analyzing the frequency and context of the extracted keywords, An educational evaluation system according to any one of claims 1 to 6, characterized in that it generates a summary of the dialogue history based on the analysis results.

8. The educational evaluation system according to any one of claims 2 to 7, characterized in that the rubric can be set by a parent or teacher.

9. A method for evaluating education by analyzing the history of conversations between children and AI, A memory step to save the dialogue history, A summary generation step that generates a summary of the dialogue history after a predetermined period of time has elapsed, An analysis step that integrates and analyzes new dialogue history and saved summaries, An evaluation generation step that generates educational evaluations based on the analysis results, An educational evaluation method characterized by including the following.

10. Computers, A storage means for saving dialogue history, Summary generation means for generating a summary of the dialogue history after a predetermined period of time has elapsed. An analytical tool that integrates and analyzes new dialogue history and saved summaries. An evaluation generation means that generates educational evaluations based on analysis results. A program designed to function as such.

Images