Learning support systems and learning support methods

Abstract

This system provides a learning support system that can indicate to learners which topics in a subject they should study. [Solution] The artificial intelligence comprises: a correct answer information creation unit 32 that creates correct answer information, which is the correct answer to the question information; an estimated thinking process creation unit 33 that generates an estimated thinking process consisting of multiple estimated thinking steps that are estimated to be considered consciously or unconsciously from the question information to the correct answer information; a correct / incorrect judgment unit 34 that outputs a correct / incorrect judgment result indicating which estimated thinking step of the estimated thinking process the learner made a mistake in, or whether the learner arrived at the correct answer information through the estimated thinking process; and a learning element extraction unit 35 that extracts learning elements to recommend to the learner from a list of learning elements included in a predetermined learning subject, based on the correct / incorrect judgment result.

Description

Technical Field

[0001] The present invention relates to a learning support system for supporting the learning of learners, particularly a learning support system that analyzes the answers of learners and the thinking processes leading to the answers, extracts and proposes individual learning tasks for learners, and a learning support method.

Background Art

[0002] Conventionally, various systems for supporting learning have been provided. For example, Patent Document 1 discloses a system including one or more computer processors for supporting learning. The one or more computer processors perform steps of setting a curriculum to each of a plurality of learning users, to which a plurality of learning contents are assigned; setting target amount information indicating a target amount of learning to each of the plurality of learning users; identifying a plurality of target learning contents from among the plurality of learning contents assigned to the curriculum set for the learning user included in the plurality of learning users based on the target amount information set for the learning user; and providing the plurality of target learning contents to the learning user. The curriculum has the plurality of learning contents assigned to each of a plurality of target periods, and the step of providing the plurality of target learning contents provides the plurality of target learning contents identified from among the plurality of learning contents assigned to the target period corresponding to that point in time among the plurality of target periods to the learning user, thereby supporting the achievement of an appropriate learning amount for each learning user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, conventional systems had problems such as being insufficient in understanding learners' learning challenges and visualizing their learning progress, and being unable to provide learning support according to the learner's learning situation. Furthermore, conventional systems lacked a mechanism for utilizing learners' answers to improve learning content and provide feedback. In addition, conventional systems had problems such as being unable to provide cross-curricular learning that transcended the boundaries of grade levels in the school system, and being unable to facilitate learning that crossed subjects.

[0005] Furthermore, conventional systems had problems such as not being able to be used, for example, for teachers to grasp the learning situation within a class and provide appropriate instruction to each student. Also, from the perspective of educators such as teachers who are responsible for guiding students, since there are many students in one class, conventional systems made it difficult to know which items of the subject matter to focus on for each individual student in order to improve the learning progress of the entire class.

[0006] In view of the circumstances described above, the present invention aims to provide a learning support system that indicates which items of a subject a learner should study (which items to return to and which to move on to) (presenting individually optimized learning). Specifically, it can recommend items to be studied across different levels of school, such as class, grade, subject, and school type (both longitudinally and trans-subjectally).

[0007] Furthermore, these problems (issues) do not preclude the existence of other problems. Also, the embodiments of the present invention described later do not need to solve all of these problems (issues). Moreover, it is possible to extract other problems (issues) from the description, drawings, or claims. [Means for solving the problem]

[0008] As a result of diligent research into the aforementioned problems, the inventors of this invention have discovered the following groundbreaking learning support system and learning support method.

[0009] A first aspect of the present invention for solving the above problems is a learning support system in which artificial intelligence assists a learner in learning a predetermined learning subject by inputting the learner's answer information to question information relating to a predetermined learning subject, wherein the artificial intelligence comprises: a correct answer information creation unit that creates correct answer information which is the correct answer to the question information; an estimated thinking process creation unit that generates an estimated thinking process consisting of a plurality of estimated thinking steps which are estimated to be considered consciously or unconsciously from the question information to the correct answer information; a correct / incorrect judgment unit that outputs a correct / incorrect judgment result indicating which estimated thinking step of the estimated thinking process the learner made a mistake in, or whether the learner arrived at the correct answer information through the estimated thinking process; and a learning element extraction unit that extracts learning elements to recommend to the learner from a list of learning elements included in the predetermined learning subject based on the correct / incorrect judgment result.

[0010] According to this first embodiment, a learning support system can be provided that indicates which items of a subject a learner should study (which items to return to and which to move on to) (providing individually optimized learning). Specifically, it can recommend items to be studied across different levels of school, such as class, grade, subject, and school type (both longitudinally and trans-subjectally).

[0011] A second aspect of the present invention is a learning support system according to the first aspect, characterized in that the artificial intelligence has a question information creation unit that creates question information according to a pre-set learning prompt, and the learning prompt includes learning subject information which is a predetermined learning subject, theme information relating to the predetermined learning subject, role information indicating the role of the artificial intelligence, and restriction information necessary when creating the question information.

[0012] According to this second embodiment, artificial intelligence can be made to create question information.

[0013] A third aspect of the present invention is a learning support system according to the first or second aspect, characterized in that the estimation thought process creation unit creates an estimation thought process taking into consideration at least one of metacognition, critical thinking, logical thinking, intuitive thinking, and thought-utterance methods.

[0014] According to this third embodiment, a learning support system can be provided that more appropriately indicates which items of a subject a learner should study.

[0015] A fourth aspect of the present invention is a learning support system according to the first aspect, characterized in that the artificial intelligence includes an achievement status determination unit that outputs an achievement status indicating the extent to which the learner has mastered the estimation thinking steps necessary from question information to correct answer information, based on the estimation thinking process and the correct / incorrect judgment result.

[0016] According to this fourth aspect, the learner's learning progress can be appropriately grasped.

[0017] A fifth aspect of the present invention is a learning support system according to the first aspect, characterized in that the learner is a plurality of learners, and the artificial intelligence further comprises a common learning element extraction unit that collects the correct / incorrect judgment results for each learner and extracts common learning elements that are commonly recommended for the plurality of learners from a list of learning elements based on the correct / incorrect judgment results of the plurality of learners.

[0018] According to this fifth embodiment, common learning elements that are recommended in common to multiple learners can be extracted, and these multiple learners can be instructed to focus their learning on those common learning elements.

[0019] A sixth aspect of the present invention is a learning support system according to the second aspect, characterized in that the subject information is English, the theme information is a topic for English conversation, the role information is an English teacher, and the restriction information is one question.

[0020] According to such a sixth aspect, it is possible to provide a learning support system that can efficiently present to a learner which items should be learned regarding English.

[0021] A seventh aspect of the present invention is that the artificial intelligence further includes a learning promotion information creation unit that creates learning promotion information based on a learning element and predetermined learning promotion instruction information required for the artificial intelligence to create learning promotion information for promoting the learning of a learner with respect to the learning element. The learning support system according to claim 1 is characterized in that.

[0022] According to such a seventh aspect, for each learner, based on a learning element and predetermined learning promotion instruction information, the artificial intelligence can create appropriate learning promotion information.

[0023] An eighth aspect of the present invention is that the learning promotion instruction book includes at least one of explanation information of terms related to the learning element and information related to problems related to the learning element. The learning support system according to the seventh aspect is characterized in that.

[0024] According to such an eighth aspect, for each learner, based on a learning element and predetermined learning promotion instruction information, the artificial intelligence can create more appropriate learning promotion information.

[0025] The ninth aspect of the present invention is a learning support method in which an artificial intelligence supports a learner's learning of a predetermined learning subject by inputting the learner's answer information in response to question information regarding the predetermined learning subject. The artificial intelligence includes a first step of creating correct answer information that is the correct answer to the question information, a second step of generating an estimated thinking process composed of a plurality of estimated thinking steps that are presumed to be considered consciously or unconsciously from the question information to the correct answer information, a third step of outputting a correct / incorrect judgment result indicating at which estimated thinking step of the estimated thinking process the learner made a mistake or reached the correct answer information through the estimated thinking process, and a fourth step of extracting learning elements to be recommended to the learner from a list of learning elements included in the predetermined learning subject based on the correct / incorrect judgment result. This is the learning support method characterized by the above.

[0026] According to such a ninth aspect, it is possible to present which items of the learning subject the learner should learn.

[0027] The tenth aspect of the present invention is that the artificial intelligence has a first step of creating question information according to a preset learning prompt. The question information is created by the artificial intelligence according to a preset learning prompt. The learning prompt includes learning subject information that is a predetermined learning subject, theme information regarding the predetermined learning subject, role information indicating the role of the artificial intelligence, and restriction information required when creating the question information. This is the learning support method according to the ninth aspect, characterized by the above.

[0028] According to such a tenth aspect, the artificial intelligence can be made to create question information.

[0029] The eleventh aspect of the present invention is that the second step creates an estimated thinking process in consideration of at least one of thinking based on metacognition, critical thinking, logical thinking, intuitive thinking, and think-aloud method. This is the learning support method according to the ninth or tenth aspect, characterized by the above.

[0030] According to this eleventh embodiment, it is possible to more appropriately indicate which items of the subject matter learners should study.

[0031] A twelfth aspect of the present invention is a learning support method according to the ninth aspect, characterized in that the artificial intelligence has a fifth step of outputting an achievement status indicating the extent to which the learner has mastered the estimation thinking steps necessary from question information to correct answer information, based on the estimation thinking process and the result of correct / incorrect judgment.

[0032] According to this twelfth embodiment, it is possible to more appropriately indicate which items of the subject matter learners should study.

[0033] A thirteenth aspect of the present invention is the learning support method according to claim 9, characterized in that the first to third steps are repeated 10 to 30 times.

[0034] According to this 13th embodiment, it is possible to more efficiently indicate to learners which items of the subject matter they should study.

[0035] A fourteenth aspect of the present invention is a learning support method according to claim 9, characterized in that the artificial intelligence creates learning promotion information based on a learning element and learning promotion instruction information necessary for creating learning promotion information that the artificial intelligence creates to promote the learner's learning with respect to the learning element.

[0036] According to this 14th embodiment, the artificial intelligence can create appropriate learning guidance information for each learner based on learning elements and predetermined learning guidance information.

[0037] A fifteenth aspect of the present invention is a learning support method according to the fourteenth aspect, characterized in that the learning instruction sheet includes at least one of descriptive information on terms related to the learning elements and information on problems related to the learning elements.

[0038] According to this 15th embodiment, the artificial intelligence can create more appropriate learning guidance information for each learner based on learning elements and predetermined learning guidance information.

[0039] In this invention, "database," "system," and "part" do not merely refer to physical means, but also include cases where the functions of the "database," "part," or "system" are realized by software. Furthermore, even if the functions of one "database," "system," or "part" are realized by two or more physical means or devices, the functions of two or more "databases," "systems," or "parts" may be realized by one physical means or device. [Brief explanation of the drawing]

[0040] [Figure 1] Figure 1 is a schematic diagram of the learning support system according to Embodiment 1. [Figure 2] Figure 2 is a schematic diagram of the learner terminal of Embodiment 1. [Figure 3] Figure 3 is a schematic diagram of the learning support server according to the embodiment. [Figure 4] Figure 4 shows an example of a learning prompt in Embodiment 1. [Figure 5] Figure 5 shows the question information created according to the learning prompts in Figure 4. [Figure 6] Figure 6 is a flowchart showing the operation of the learning support system according to the embodiment. [Figure 7] Figure 7 shows the question information in Example 1. [Figure 8] Figure 8 shows the learner's answer information in Example 1. [Figure 9] Figure 9 shows the correct answer information and estimation thought process in Example 1. [Figure 10] Figure 10 shows the correct / incorrect judgment results for each learner (students A-C) in Example 1. [Figure 11] Figure 11 shows the learning elements recommended for each learner (students A to C) in Example 1. [Figure 12]Figure 12 shows the question information in Example 2. [Figure 13] Figure 13 shows the learner's answer information in Example 2. [Figure 14] Figure 14 shows the correct answer information and estimation thought process in Example 2. [Figure 15] Figure 15 shows the learners' correct / incorrect judgment results in Example 2. [Figure 16] Figure 16 shows the learning elements recommended for learners in Example 2. [Figure 17] Figure 17 shows the question information for Example 3. [Figure 18] Figure 18 shows the answer information for each learner (students A to E) in Example 3. [Figure 19] Figure 19 shows the correct answer information (example solution) and estimation thought process in Example 3. [Figure 20] Figure 20 shows the correct / incorrect judgment results for each learner (students A to E) in Example 3. [Figure 21] Figure 21 shows the recommended learning elements for each learner (students A and B) in Example 3. [Figure 22] Figure 22 shows the recommended learning elements for each learner (students C to E) in Example 3. [Figure 23] Figure 23 is a schematic diagram of the learning support system according to Embodiment 3. [Figure 24] Figure 24 is a schematic diagram of the learning support system according to Embodiment 4. [Figure 25] Figure 25 shows an example of predetermined learning promotion instruction information. [Figure 26] Figure 26 shows an example of predetermined learning promotion instruction information. [Figure 27] Figure 27 shows an example of a learning promotion information creation unit. [Figure 28] Figure 28 shows an example of a learning promotion information creation unit. [Figure 29] Figure 29 shows an example of a learning element. [Figure 30] Figure 30 shows an example of learning promotion information. [Figure 31] Figure 31 shows an example of learning promotion information. [Figure 32] Figure 32 shows an example of learning promotion information. [Figure 33] Figure 33 shows an example of learning promotion information. [Figure 34] Figure 34 shows an example of a learning element. [Figure 35] Figure 35 shows an example of learning promotion information. [Figure 36] Figure 36 shows an example of learning promotion information. [Figure 37] Figure 37 shows an example of learning promotion information. [Figure 38] Figure 38 shows an example of learning promotion information. [Figure 39] Figure 38 shows an example of learning promotion information. [Modes for carrying out the invention]

[0041] Embodiments of the learning support system and learning support method according to the present invention will be described below with reference to the attached drawings. However, the present invention is not limited to the following embodiments.

[0042] (Embodiment 1) The learning support system 1 of this embodiment supports a learner's learning of a predetermined learning subject by allowing the learner to input answer information (including response information) in response to question information. As shown in Figure 1, the learning support system 1 consists of a plurality of learner terminals 10 and a learning support server 30 connected to these learner terminals 10 via a network 20.

[0043] Here, the learner may be one person or multiple people. Furthermore, the prescribed learning subjects refer to the subjects predetermined by elementary schools, junior high schools, high schools, universities, etc., such as English, Japanese language, mathematics (arithmetic), science, and social studies (geography, history, and civics).

[0044] First, let's describe the learner terminal 10. The learner terminal 10 is used by learners and, as shown in Figure 2, has an output unit 11 that displays the output of the learning support system 1 and an input unit 12 that allows learners to input information into the learning support system 1.

[0045] The output unit 11 is not particularly limited as long as it can output the output of the learning support system 1 (for example, text, video, images, audio, etc.), and examples include liquid crystal displays, braille displays, speakers, headphones, and projectors.

[0046] The input unit is not particularly limited as long as it allows learners to input information (e.g., text, video, images, QR codes®, audio, etc.) into the learning support system 1. Examples include keyboards, touch panels, microphones, cameras, styluses, QR code® readers, and scanners. Note that QR code® readers and scanners are used when learners input information written on paper or elsewhere into the learning support system 1.

[0047] The learner terminal 10 is not particularly limited as long as it has the output unit 11 and input unit 12 described above, and examples include personal computers, tablet computers, and smartphones.

[0048] Next, the learning support server 30 will be described. As shown in Figure 3, the learning support server 30 is equipped with artificial intelligence. This artificial intelligence has been trained on at least the following four things.

[0049] The first stage of learning involves training the system to generate questions and their correct answers for a given subject. The second stage of learning involves training the system to determine whether the learner's answer to a question is correct or incorrect. The third stage of learning involves training the system to recommend specific learning elements included in the given subject if the learner's answer is incorrect. The fourth stage of learning involves training the system to determine which learning elements should be studied in each grade level, from the first grade of elementary school to the third year of high school, for a given subject. These learning methods are not particularly limited, and for example, artificial intelligence may be trained using a large-scale language model. Here, a "large-scale language model" refers to a natural language model constructed with a large "computational complexity," "data volume," and "number of parameters." For example, ChatGPT® and Google Gemini® have already been trained on these aspects.

[0050] Through this learning process, artificial intelligence (AI) learns, for example, the content of the curriculum guidelines for elementary, junior high, and high schools in Japan, and learns what knowledge and skills need to be acquired at each grade level. As a result, based on the analysis of learners' answer information, it can identify the learning elements necessary to supplement any missing knowledge or skills. Furthermore, by analyzing learners' answer information and recognizing the differences from the correct answers, AI can identify the problems (challenges) that learners are facing. As a result, AI can extract the learning elements necessary to solve these problems from past learning elements (a list of learning elements) or connect them to learning elements that will be needed in the future.

[0051] Furthermore, artificial intelligence possesses advanced natural language processing capabilities, enabling it to analyze learners' responses from various perspectives, including grammar, vocabulary, expression, and logical structure. As a result, it can gain a detailed understanding of the learner's estimated thought process and areas for improvement in their expressive abilities, and then present appropriate learning elements.

[0052] Furthermore, artificial intelligence has learned from a vast amount of text data and possesses knowledge from various fields. As a result, it can provide relevant learning elements regardless of the learner's answers, and support the deepening of the learner's understanding.

[0053] Furthermore, as shown in this figure, this artificial intelligence includes a question information creation unit 31, a correct answer information creation unit 32, an estimation thinking process creation unit 33, a correct / incorrect judgment unit 34, a learning element extraction unit 35, and an achievement status judgment unit 36.

[0054] The question information creation unit 31 is a function that can create question information according to a pre-set learning prompt. Here, a learning prompt is a function that can create question information based on at least learning subject information, theme information, role information, and restriction information.

[0055] Here, "learning subject information" refers to information indicating the specific learning subject to be studied, such as English, Japanese language, mathematics (arithmetic), science, social studies (geography, history, civics), etc.

[0056] Thematic information refers to subjects related to the designated subject of study. For example, if the designated subject (subject information) is English, the topic would be English conversation; if the designated subject is mathematics, it would include topics such as factorization, trigonometric functions, differential and integral calculus, and properties of parallelograms.

[0057] Role information refers to the roles assigned to artificial intelligence. Examples include teachers corresponding to learning subjects such as English teachers and mathematics teachers, university professors, cram school instructors, learning consultants, educational consultants, private tutors, and correspondence course tutors.

[0058] Restriction information refers to information about the limitations imposed when creating question information. Examples include "only one question per subject," "the question must be within the curriculum guidelines for the subject," "the question must be at the level of a third-year junior high school student," and "it must be within the scope of basic instruction for the subject," but these are not required. If there is no restriction information, you can use something like "Null" as the restriction information.

[0059] The question information is information about questions created by artificial intelligence based on this information. This question information will be created based on at least the subject information, theme information, role information, and restriction information. Therefore, the created question information will reflect the theme information of the subject indicated by the subject information, the role according to the role information, and the restrictions imposed by the restriction information. For example, if the learning prompt shown in Figure 4 is input to the question information creation unit 31, the question information shown in Figure 5 can be generated.

[0060] The correct answer information generation unit 32 is a function that can create correct answer information, which is the correct answer to the question information. The created correct answer information is not limited to one; there may be multiple correct answers. Furthermore, the correct answer information does not need to be exactly the same; as long as it is created by artificial intelligence as the correct answer, it will be considered correct information even if there are differences in expression, etc. For example, expressions that include synonyms for the correct answer, simplified wording, approximate answers, etc. The correct answer information generation unit 32 may be pre-configured to output correct answer information for the question information.

[0061] The estimation thought process creation unit 33 is a function that can generate an estimation thought process consisting of multiple estimation thought steps that are presumed to be considered consciously or unconsciously from the question information to the correct answer information. An estimation thought step is an element that constitutes the estimation thought process (see Examples 1-3).

[0062] Here, "estimated thought process" is not limited to any process (procedure) that is not shown in the question information or the correct answer information, but is presumed to be a process that a person consciously or unconsciously considers when arriving at the correct answer information from the question information. Examples of "estimated thought processes" include "metacognition," "critical thinking," "intuitive thinking," "logical thinking," and "thinking based on thought utterances." Here, "metacognition" is described in detail, for example, in Reference 1: "Metacognition: An Overview" by John H. Flavell (1979), Reference 2: "Metacognition and Cognitive Neuropsychology: Monitoring and Control Processes" edited by GRD McClelland, GLP Williams, and PALG Delacour (2003), and Reference 3: "Handbook of Metacognition" edited by John Dunlosky and Janet Metcalfe (2009). Furthermore, "thought-aloud" refers to a technique of verbally explaining one's thoughts during the process of performing a specific task, and "thought-aloud-based thinking" refers to the process of thinking in a way that allows for thought-aloud. In addition, the concept of rubrics may be incorporated into these inferential thinking processes. Here, "rubric" refers to a scale of several levels that indicates the degree of success, and more specifically, it refers to an assessment tool for measuring the degree of learning achievement that has the following three characteristics.

[0063] (1) Visualize the evaluation criteria and scales in a table. (2) Used to determine the degree to which learning objectives have been achieved. (3) This is a tool that serves as a judgment criterion for absolute evaluation, showing the evaluation "perspectives" and "scales" in a matrix table.

[0064] Specifically, for example, by providing the artificial intelligence with further prompts such as "Create an estimation thought process that takes metacognition into consideration," "Create an estimation thought process that takes critical thinking into consideration," "Create an estimation thought process that takes intuitive thinking into consideration," "Create an estimation thought process that takes logical thinking into consideration," or "Create an estimation thought process that takes into account thinking based on thought utterances," the artificial intelligence's estimation thought process creation unit 33 can create estimation thought processes that also take these into consideration. In addition, the estimation thought process creation unit 33 may also be configured to create estimation thought processes that take into consideration non-cognitive abilities such as "goal setting," "planning ability," and "self-regulation ability." By creating estimation thought processes that take such non-cognitive abilities into consideration, it is possible to more appropriately present to learners which items of the learning subjects they should study.

[0065] The correct / incorrect judgment unit 34 is a function that outputs a correct / incorrect judgment result indicating at which step of the estimation thinking process the learner made a mistake, or whether the learner arrived at the correct information through the estimation thinking process. The correct / incorrect judgment result may or may not be displayed on the output unit 11 of the learner terminal 10.

[0066] The learning element extraction unit 35 is a function that extracts learning elements recommended to learners from a list of learning elements included in a given learning subject, based on the correct / incorrect judgment result. Here, learning elements refer to a given learning unit, and examples include learning units categorized by the curriculum guidelines, learning units categorized by private cram schools, learning units categorized by institutions that create workbooks, learning programs provided by textbook publishing companies, open-source educational material libraries (for example, Khan Academy: an online learning platform covering a wide range of subjects such as mathematics, science, and history; Curriki: an online platform for educators to share and collaboratively create educational materials; OER Commons: a search platform for open educational resources (OER)), and digital resources for public education (for example, the curriculum guidelines database provided by the Ministry of Education, Culture, Sports, Science and Technology; the educational information sharing portal site provided by the National Institute for Educational Policy Research; and educational digital content provided by local governments).

[0067] The achievement status determination unit 36 ​​is a function that outputs an achievement status indicating the extent to which the learner has mastered the estimation thinking steps necessary to arrive at the correct answer from the question information, based on the estimation thinking process and the correct / incorrect judgment results. Specifically, for example, it may determine (output) that 80% of the estimation thinking steps necessary to arrive at the correct answer from the question information have been achieved (fulfilled). In this case, the output unit 11 of the learner terminal 10 may display "Achievement Rate 80%" as the achievement status, or display an illustration or other image illustrating the content. Displaying such achievement rates or illustrations can improve the learner's motivation to learn.

[0068] Next, the operation of the learning support system 1 of this embodiment will be described. As shown in Figure 6, when the learning support system 1 is activated, the learner or the person who instructs or supports the learner (teacher, parent, etc.) inputs a learning prompt into the input unit 12 of the learner terminal 10 (S1). This learning prompt is transmitted to the learning support server 30 via the network 20. The artificial intelligence installed in the learning support server 30 then uses the question information creation unit 31 to create question information based on the learning subject information, theme information, role information, and restriction information included in the learning prompt (S2: Step 0). The created question information is then transmitted to the learner terminal 10 via the network 20 and displayed on the output unit 11 of the learner terminal 10 (S3).

[0069] Subsequently, the learner who has seen (heard) the question information inputs the answer information to that question information into the input unit 12 (speaking, inputting handwritten information as an image, writing, etc.) (S4). This answer information is transmitted to the learning support server 30 via the network 20. Then, the artificial intelligence creates the correct answer information, which is the correct answer to the question information, using the correct answer information creation unit 32 (S5: first step). Next, the artificial intelligence creates an estimated thinking process, which consists of multiple estimated thinking steps that are estimated to have been considered consciously or unconsciously from the question information to the correct answer information, using the estimated thinking process creation unit 33 (S6: second step).

[0070] Subsequently, the artificial intelligence, using the correct / incorrect judgment unit 34, outputs a correct / incorrect judgment result from the answer information, indicating at which step of the estimation thinking process the learner made a mistake, or whether the learner arrived at the correct answer information through the estimation thinking process (S7: third step).

[0071] Then, if the correct / incorrect judgment result indicates that the answer information was arrived at as correct information through the estimation thought process, the artificial intelligence uses the learning element extraction unit 35 to extract learning elements that the learner should learn next from the list of learning elements included in a predetermined learning subject (for example, learning elements from the list of learning elements included in a predetermined learning subject one grade higher) (S8: Steps 4 and 5).

[0072] On the other hand, if the correct / incorrect judgment result determines that the answer information was incorrect, the artificial intelligence uses the learning element extraction unit 35 to extract learning elements from the list of learning elements included in the designated learning subject, recommending that the learner relearn the incorrect part (S9). Needless to say, the learning elements recommended to the learner may be those that are deemed beneficial for the learner to relearn, regardless of age, grade level, etc. Then, these learning elements are displayed on the output unit 11 of the learner terminal 10, and the operation of the learning support system 1 ends.

[0073] As described above, by configuring the learning support system 1, it is possible to appropriately indicate to learners which items of a subject they should study. In particular, the learning support system 1 according to the present invention is groundbreaking because it not only recommends the optimal learning elements so that learners can relearn the parts they have gotten wrong, but also recommends the optimal learning elements that are considered to be the next things they should study.

[0074] (Embodiment 2) In the learning support system of Embodiment 1, the artificial intelligence generates question information by inputting a learning prompt into the learning support system, but the present invention is not limited to this. Question information may be pre-created by a person (for example, questions included in a problem set). Even if the learning support system is constructed in this way, the same effects as the learning support system of Embodiment 1 can be obtained.

[0075] <Example 1 (Mathematics)> The problem shown in Figure 7 was used as the question information, and the information shown in Figure 8 was used as the answer information for the learners (students A to C). When this information was input into the learning support system of this embodiment, the learning support system created the correct answer information using the correct answer information creation unit, and created the estimated thinking process (solution process) shown in Figure 9 using the estimated thinking process creation unit. In this embodiment, Google Gemini® was used as the artificial intelligence, and the estimated thinking process was created considering thinking based on the thought-aloud method.

[0076] Next, the learning support system, using its correct / incorrect judgment unit, outputted the correct / incorrect judgment results for each student, as shown in Figure 10, based on the answer information and estimated thought process. Subsequently, the learning support system, using its learning element extraction unit, extracted (outputted) recommended learning elements (learning tasks) for the learners, as shown in Figure 11. In this embodiment, to further clarify the learning elements, the learning element extraction unit also presents sample examples of learning tasks.

[0077] In this example, not only are the units to return to (learning elements to be relearned) indicated, but the units to move on to (learning elements to be learned next) are also presented.

[0078] From this, it was found that, according to the learning system of the present invention, even if the designated learning subject is mathematics, it is possible to show learners which items of the subject they should study.

[0079] <Example 2 (in the case of arithmetic)> The question information used was the problem shown in Figure 12, and the learner's answer information was the information shown in Figure 13. When this information was input into the learning support system of this embodiment, the learning support system created the correct answer information using the correct answer information creation unit, and created the estimated thinking process (solution process) shown in Figure 14 using the estimated thinking process creation unit. In this embodiment, Google Gemini® was used as the artificial intelligence, and the estimated thinking process was created considering thinking based on thought-aloud methods.

[0080] Next, the learning support system, using its correct / incorrect judgment unit, outputted the learner's correct / incorrect judgment result, as shown in Figure 15, based on the answer information and estimated thought process. Subsequently, the learning support system, using its learning element extraction unit, extracted (output) the learning elements (learning tasks) recommended for the learner, as shown in Figure 16.

[0081] In this embodiment, not only is the place to return to (the learning element to be relearned) indicated, but the place to move forward to (the learning element to be learned next) is also shown.

[0082] From this, it was found that, according to the learning system of the present invention, even if the designated learning subject is mathematics, it is possible to show learners which items of the subject they should study.

[0083] In this embodiment, the learner inputs answer information for two types of question information, but the learning system according to the present invention may be configured to input answer information for even more types of question information. By inputting answer information for multiple types of question information, it is possible to more appropriately present to the learner which items of the subject they should study, to more appropriately grasp the learner's learning progress, and furthermore, to more appropriately present to the learner items of a higher learning stage in a given subject.

[0084] <Example 3 (in English)> The problem shown in Figure 17 was used as the question information, and the information shown in Figure 18 was used as the answer information for the learners (students A to C). When this information was input into the learning support system of this embodiment, the learning support system created the correct answer information using the correct answer information creation unit, and created the estimated thinking process (solution process) shown in Figure 19 using the estimated thinking process creation unit. Here, it was found that even with question information that does not have a clear correct answer, as in Examples 1 and 2, the artificial intelligence can create the correct answer information (example answer) using the estimated thinking process creation unit. In this embodiment, Google Gemini® was used as the artificial intelligence, and the estimated thinking process was created considering thinking based on thought-aloud methods.

[0085] Next, the learning support system, using its correct / incorrect judgment unit, outputted the correct / incorrect judgment results for each student, as shown in Figure 20, based on the answer information and estimated thinking process.

[0086] Subsequently, the learning support system used a learning element extraction unit to extract (output) recommended learning elements (learning tasks) for learners, as shown in Figures 21 and 22.

[0087] In this embodiment, the learning content to be reviewed includes not only English language use (e.g., grammar, vocabulary, etc.), but also learning elements from Japanese language, career education, technology and home economics, art, and social studies. In other words, this embodiment enables cross-curricular and longitudinal learning.

[0088] From this, it was found that, according to the learning system of the present invention, even if there are multiple predetermined learning subjects, it is possible to show learners which items of the learning subjects they should study.

[0089] (Embodiment 3) The learning support system 1 of Embodiment 1 was capable of indicating to each learner which items of the learning subject they should study, but the present invention is not limited to this. For example, as shown in Figure 23, the learning support system may be configured to have a learning support server 30A which further includes a common learning element extraction unit 37 that uses artificial intelligence to collect incorrect answer information for each learner and extracts learning elements that are commonly recommended for multiple learners from a list of learning elements based on the incorrect answer information of multiple learners.

[0090] The common learning element extraction unit 37 is a function that extracts learning elements that are commonly recommended for multiple learners. Specifically, for example, if the learning subject is English and the theme is English conversation on a predetermined topic, and the incorrect answer information of many learners includes the use of present and past tense verbs, the common learning element extraction unit 37 will extract "the use of present and past tense verbs" from the learning element list.

[0091] By configuring the learning support system to include such a learning support server 30A, it is possible to extract common learning elements that are recommended for multiple learners. If, for example, the teacher of the subject in question becomes aware of these common learning elements, they can then have those multiple learners focus on learning those common learning elements in actual lessons.

[0092] (Embodiment 4 (Feedforward)) In the embodiment described above, the learning support system only performed the task of extracting learning elements, but the present invention is not limited thereto. For example, as shown in Figure 24, the learning support system may be configured to further include a learning promotion information creation unit 38 that creates learning promotion information based on the extracted learning elements and predetermined learning promotion instruction information necessary for creating learning promotion information that artificial intelligence creates to promote the learner's learning with respect to those learning elements. The other components are the same as those of the learning support system according to Embodiment 1.

[0093] Here, the predetermined learning promotion instruction information is pre-created information that includes at least one of the following: learning information (explanatory information on terms related to learning elements, etc.), learning procedures, learning content (information on problems (exercises) related to learning elements), exercise format information, learning task information, learning activity information, learning strategy information, material presentation information, evaluation criteria information, feedback information, learning design information, instructional strategy, learning support policy information, and learning form information. Examples of predetermined learning promotion instruction information include those shown in Figures 25 and 26. The location where the predetermined learning promotion instruction information exists (is stored) is not particularly limited; for example, it may be stored in the learning support server 30B, or it may be stored (displayed on a website) on another computer connected via the Internet, etc.

[0094] The learning promotion information creation unit 38 is not particularly limited as long as it can enable the artificial intelligence to create learning promotion information based on learning elements and predetermined learning promotion instruction information, for example, prompts given to the artificial intelligence. Examples of the learning promotion information creation unit 38 include those shown in Figures 27 and 28. Here, the "Homepage URL" in Figures 27 and 28 will contain the URL of an actual website.

[0095] Then, for example, using the learning support system of Embodiment 1, after extracting the learning elements shown in Figure 29, the learning promotion information creation unit 38 shown in Figure 27 can create learning promotion information as shown in Figures 30 to 33 based on the learning elements shown in Figure 29 and predetermined learning promotion instruction information shown in Figure 25 (Step 6). Note that Figure 31 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 30, Figure 32 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 31, and Figure 33 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 32.

[0096] Furthermore, for example, using the learning support system of Embodiment 1, after extracting the learning elements shown in Figure 34, the learning promotion information creation unit 38 shown in Figure 27 can create learning promotion information as shown in Figures 35 to 39 based on the learning elements shown in Figure 29 and predetermined learning promotion instruction information shown in Figure 26 (6th step). Note that Figure 36 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 35, Figure 37 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 36, Figure 38 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 37, and Figure 39 is an example of learning promotion information when the correct answer is entered to the question shown in Figure 38.

[0097] Thus, by further including the learning promotion information creation unit 38, the artificial intelligence can create appropriate learning promotion information for each learner based on learning elements and predetermined learning promotion instruction information. [Industrial applicability]

[0098] This invention can also be used in problems included in digital textbooks. [Explanation of symbols]

[0099] 1. Learning support system 10. Learner terminals 11 Output section 12 Input section 20 Networks 30, 30A, 30B Learning Support Server 31 Question Information Creation Department 32 Correct Answer Information Creation Department 33 Estimated Thought Process Creation Department 34 Correct / Incorrect Judgment Unit 35 Learning Element Extraction Unit 36. Achievement Status Judgment Department 37 Common Learning Element Extraction Unit 38 Learning Promotion Information Creation Department

Claims

1. A learning support system in which artificial intelligence assists a learner in learning a predetermined subject by inputting the learner's answer information in response to question information related to a predetermined subject, The aforementioned artificial intelligence, A correct answer information creation unit creates correct answer information, which is the correct answer to the aforementioned question information. An estimation thought process creation unit generates an estimation thought process consisting of multiple estimation thought steps that are presumed to be considered consciously or unconsciously from the aforementioned question information to the aforementioned correct answer information, A correct / incorrect judgment unit outputs a correct / incorrect judgment result that indicates, from the aforementioned answer information, which of the estimation thinking steps in the estimation thinking process the learner made a mistake, or how the learner arrived at the correct answer information through the estimation thinking process. The system includes a learning element extraction unit that extracts learning elements recommended to the learner from a list of learning elements included in a predetermined learning subject, based on the result of the correct / incorrect judgment. A learning support system characterized by the following features.

2. The artificial intelligence has a question information creation unit that creates the question information according to a pre-set learning prompt, The aforementioned learning prompt is, The learning subject information, which is the predetermined learning subject, Thematic information relating to the aforementioned prescribed learning subjects, Role information indicating the role of the aforementioned artificial intelligence, The restriction information necessary when creating the aforementioned question information, The learning support system according to feature 1.

3. The estimation thought process creation unit creates the estimation thought process by considering at least one of metacognition, critical thinking, logical thinking, intuitive thinking, and thinking based on thought utterances. A learning support system according to feature 1 or 2.

4. The aforementioned artificial intelligence, The system includes an achievement status determination unit that outputs an achievement status indicating the extent to which the learner has mastered the estimation thinking steps necessary to arrive at the correct answer information from the question information, based on the estimation thinking process and the correct / incorrect judgment result. The learning support system according to feature 1.

5. The aforementioned learners consist of multiple learners, The artificial intelligence further includes a common learning element extraction unit that collects the correct / incorrect judgment results for each learner and, based on the correct / incorrect judgment results of the multiple learners, extracts common learning elements from the learning element list that are commonly recommended for the multiple learners. The learning support system according to feature 1.

6. The aforementioned subject information is in English. The aforementioned topic information is a topic for English conversation. The aforementioned role information is that of an English teacher. The aforementioned restriction information is that there is only one question. The learning support system according to feature 2.

7. The artificial intelligence further includes a learning promotion information creation unit that creates learning promotion information based on the learning elements and predetermined learning promotion instruction information necessary for creating learning promotion information that the artificial intelligence creates to promote the learner's learning with respect to the learning elements. The learning support system according to feature 1.

8. The learning instruction sheet includes at least one of the following: descriptive information on terms related to the learning element and information on problems related to the learning element. The learning support system according to feature 7.

9. A learning support method in which artificial intelligence assists a learner in learning a predetermined subject by inputting the learner's answer information in response to question information related to a predetermined subject, The aforementioned artificial intelligence, The first step is to create correct answer information, which is the correct answer to the aforementioned question information, A second step involves generating an estimation thinking process consisting of multiple estimation thinking steps that are presumed to be considered consciously or unconsciously from the aforementioned question information to the aforementioned correct answer information, A third step involves outputting a correct / incorrect judgment result from the aforementioned answer information, indicating at which of the estimation thinking steps the learner made a mistake in the estimation thinking process, or how they arrived at the correct answer information through the estimation thinking process. A fourth step is to extract the learning elements recommended to the learner from the list of learning elements included in the predetermined learning subject, based on the result of the correct / incorrect judgment. A learning support method characterized by the following features.

10. The artificial intelligence has a first step of creating the question information according to a pre-set learning prompt, The aforementioned question information is generated by artificial intelligence according to pre-set learning prompts. The aforementioned learning prompt is, The learning subject information, which is the predetermined learning subject, Thematic information relating to the aforementioned prescribed learning subjects, Role information indicating the role of the aforementioned artificial intelligence, The restriction information necessary when creating the aforementioned question information, The learning support method according to feature 9.

11. The second step involves creating the inference thinking process by considering at least one of metacognition, critical thinking, logical thinking, intuitive thinking, and thinking based on thought utterances. The learning support method according to feature 9 or 10.

12. The aforementioned artificial intelligence, A fifth step, based on the estimation thinking process and the correct / incorrect judgment result, outputs an achievement status indicating the extent to which the learner has mastered the estimation thinking steps necessary to arrive at the correct answer information from the question information. The learning support method according to feature 9.

13. The learning support method according to claim 9, characterized in that the first to third steps are repeated 10 to 30 times.

14. The artificial intelligence includes a sixth step of creating learning promotion information based on the learning elements and learning promotion instruction information necessary for creating learning promotion information that the artificial intelligence creates to promote the learner's learning with respect to the learning elements. The learning support method according to feature 9.

15. The learning instruction sheet includes at least one of the following: descriptive information on terms related to the learning element and information on problems related to the learning element. The learning support method according to feature 14.

Images