Learning support system, method, and program

Abstract

Provided is a technique for assisting learning to correctly and deeply understand matters to be learned. 【Solution means】The learning support system includes a dialogue unit that conducts a dialogue with the learner so that the learner explains a predetermined learning matter, an accuracy determination unit that determines the accuracy, which is the accuracy of the explanation in the dialogue, and an understanding degree determination unit that determines the understanding degree, which is the depth of the learner's understanding of the learning matter.

Description

Technical Field

[0001] This disclosure relates to a technology for assisting learning.

Background Art

[0002] Active learning, in which learners actively and proactively engage in learning, is known to have a high learning effect. Active learning includes learning methods such as group discussions, personal experiences, and teaching others. In particular, the learning method of teaching oneself by teaching others through dialogue is said to have a high learning effect. However, it is not easy to implement the learning method of teaching others. Since it cannot be implemented by a single learner alone, it is necessary to secure others and provide a place for implementation. In addition, teaching others itself takes a certain amount of time, and others will be restricted during that time. Also, the act of teaching others is a rather difficult act for learners, and psychological factors such as not wanting to feel ashamed or not wanting to do it will work, which hinders implementation.

[0003] Patent Document 1 discloses a technology for assisting the implementation of learning through dialogue using a computer. The system disclosed in Patent Document 1 stores problem information regarding a predetermined problem statement of a problem presented to a learner and a plurality of correct answer component phrases included in the correct answer to the problem, presents the problem to the learner, and realizes dialogue with the learner by obtaining a response from the learner. Then, the correct answer component phrases included in the response from the learner are identified, and the correct answer component phrases not included in the response are further presented as hints. According to this system, a learner can perform learning through dialogue without hesitation alone with a computer as a partner.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the system of Patent Document 1, the degree of understanding of the learner is evaluated based on how many correct constituent phrases are included in the learner's utterance. However, even if the content of the learner's response is correct, it is not always the case that the learner deeply understands the knowledge and makes it their own. In this regard, according to the system of Patent Document 1, it is possible to evaluate the accuracy of the learner's understanding based on whether the correct constituent phrases are included in the learner's utterance, but it is difficult to evaluate the depth of the learner's understanding. One object included in the present disclosure is to provide a technique for supporting learning that correctly and deeply understands the matter to be learned.

Means for Solving the Problem

[0006] A learning support system according to one aspect of the present invention includes a dialogue unit that conducts a dialogue with the learner so that the learner explains a predetermined learning matter, an accuracy determination unit that determines the accuracy, which is the accuracy of the explanation in the dialogue, and an understanding degree determination unit that determines the understanding degree, which is the depth of the learner's understanding of the learning matter.

Effect of the Invention

[0007] According to one aspect included in the present disclosure, it becomes possible to support learning that correctly and deeply understands the matter to be learned.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Mode for Carrying Out the Invention

[0009] Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing the functional configuration of the learning support system according to the present embodiment.

[0010] The learning support system 10 includes a reading aloud unit 11, an interaction unit 12, an accuracy determination unit 13, a comprehensibility determination unit 16, a comprehensive determination unit 19, a motivation processing unit 20, and a feedback unit 23. The accuracy determination unit 13 includes a voice analysis unit 14 and a text evaluation unit 15. The comprehensibility determination unit 16 includes a paralanguage measurement unit 17 and a paralanguage evaluation unit 18. The motivation processing unit 20 includes a ranking unit 21 and a growth visualization unit 22.

[0011] The interaction unit 12 operates in cooperation with the generative artificial intelligence 81, interacts with the learner 80 as an avatar instructor, and prompts the learner 80 to explain predetermined learning items. The reading aloud unit 11 converts the speech of the avatar instructor generated by the interaction unit 12 into speech for reading aloud and outputs it to the learner 80. The voice analysis unit 14 converts the voice of the explanation regarding the learning item spoken by the learner 80 into text. The text is sent to the text evaluation unit 15. The text evaluation unit 15 calculates the accuracy of the explanation shown in the text. The accuracy is the correctness of the explanation. The accuracy is sent to the comprehensive determination unit 19. The paralanguage measurement unit 17 measures the paralanguage (e.g., speech rate and intonation) included in the learner 80's speech. The measurement result is sent to the paralanguage evaluation unit 18.

[0012] The paralanguage evaluation unit 18 calculates a predetermined paralanguage evaluation index and calculates the degree of understanding based on the paralanguage index. The degree of understanding is the depth of understanding of the learning items of the learner 80. The degree of understanding is sent to the comprehensive determination unit 19.

[0013] The comprehensive determination unit 19 calculates the proficiency based on the accuracy and the degree of understanding. The proficiency is an index that comprehensively evaluates how proficient the learner 80 is in the learning items, in other words, how well the learner can use the knowledge of the learning items. Information on the accuracy, the degree of understanding, and the proficiency is sent to the ranking unit 21 and the growth visualization unit 22 of the motivation processing unit 20, and to the feedback unit 23. The ranking unit 21 totals the degree of proficiency for each learner, ranks the learners, and presents them to the learner 80.

[0014] The growth visualization unit 22 records, in association with the learning items, the number of times of learning by teaching the learning items and the proficiency calculated in that session, and performs a visible display of the state of change in proficiency for each passing session.

[0015] The feedback unit 23 provides encouraging words to the learner in the dialogue based on at least one of the accuracy, the degree of understanding, and the proficiency. Also, the feedback unit 23 determines whether or not to teach a model explanation to the learner 80 based on one or more of the accuracy, the degree of understanding, and the proficiency, and if it is determined that a model explanation should be taught, provides the model explanation to the learner 80 in the dialogue. FIG. 2 is a block diagram showing the hardware configuration of the learning support system of the present embodiment. The learning support system 10 is composed of a server 30 and an information terminal 40. The server 30 and the information terminal 40 can communicate with each other via a communication network 82.

[0016] The server 30 includes a processor 31, a main memory 32, a storage device 33, a communication device 34, an input device 35, and a display device 36. The processor 31, the main memory 32, the storage device 33, the communication device 34, the input device 35, and the display device 36 are communicably connected to each other via a bus 37.

[0017] The processor 31 is composed of, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), a GPU (Graphics Processing Unit), an FPGA (Field-Programmable Gate Array), etc. By reading and executing various programs stored in the main memory 32 or the storage device 33, various functions of the learning support system 10 are realized.

[0018] The main memory 32 is a main storage device that stores programs and data, and is, for example, a Random Access Memory (RAM), a Read Only Memory (ROM), or a non-volatile semiconductor memory (Non-Volatile RAM (NVRAM)).

[0019] The storage device 33 is, for example, a Hard Disc Drive (HDD), a Solid State Drive (SSD), a storage system, an Integrated Circuit (IC) card, a Secure Digital (SD) memory card, a reading and writing device for recording media such as an optical recording medium (Compact Disc (CD), Digital Versatile Disc (DVD), etc.), or a storage area of a cloud server.

[0020] The cache, registers, main memory 32, and storage device 33 of the processor 31 may be collectively referred to as memory. The necessary programs and data are loaded into the memory and executed, thereby performing information processing in the learning support system 10. Also, various parameters such as threshold values may be stored in the memory. Programs, databases, datasets, etc. may be stored in the memory.

[0021] The communication device 34 is a wired or wireless communication interface that realizes communication with other devices such as an external server (not shown) and the information terminal 40 via the communication network 82. For example, the communication device 34 is a Network Interface Card (NIC), a wireless communication module, a Universal Serial Interface (USB) module, or a serial communication module.

[0022] The input device 35 is a device that accepts input from the administrator. The input device 35 is, for example, a keyboard, a mouse, a touch panel, a card reader, or a voice input device.

[0023] The display device 36 is a device that provides various information to the administrator. The display device 36 is, for example, a screen display device (such as a Liquid Crystal Display (LCD), a Head Mounted Display (HMD), etc.), a voice output device, a printing device, etc. The communication network 82 is a wired or wireless communication means such as a Local Area Network (LAN) or the Internet.

[0024] The information terminal 40 is a device that accesses the server 30 and provides the service provided by the server 30 to the learner 80 on the browser 43, and is, for example, a smartphone, a tablet terminal, or a personal computer. The information terminal 40 includes a speaker 41 and a microphone 42, and further includes a processor, a memory, an input device, a display device, and a communication device (not shown), and is a device that executes a software program by the processor. The microphone 42 can acquire the voice of the learner 80 in front of the information terminal 40. The speaker 41 can output voice to the learner 80. FIG. 3 is a flowchart of the dialogue learning process by the learning support system. The dialogue learning process is a process in which the learning support system 10 provides a dialogue learning service to the learner 80.

[0025] Referring to FIG. 3, first, in step S101, the dialogue unit 12 starts a dialogue with the learner 80 by the avatar instructor. The dialogue is configured by preparing a certain degree of dialogue scenario of a branching type that branches according to the development of the conversation with the learner 80 in advance, and using the generative artificial intelligence 81 based on the scenario to generate the utterances of the avatar instructor who dialogues with the learner 80. The dialogue starts, for example, when the avatar instructor requests the learner 80 to explain a predetermined learning item. As an example, the avatar instructor makes an utterance such as "Mr. 〇〇, please explain the term xxx. If possible, please exemplify how to use the term or explain it in an easy-to-understand way using a metaphor." The dialogue proceeds in such a way that when the learner 80 responds to the request and explains the learning item.

[0026] The comprehension degree determination unit 16 continuously measures the paralanguage included in the utterance of the learner 80 while the dialogue is being conducted and the learner 80 is explaining the learning item. The paralanguage is, for example, the speaking speed, the number of fillers, the term selection, the number of silences, the terminal form, and the voice waveform frequency.

[0027] The speaking speed is an index based on the speed of speech. As an example, it is the number of speech units per unit time in speech. The speech units are, for example, words, moras, etc. The speaking speed is, for example, the number of moras per second, the number of words per minute, etc. The filler count is an index based on the number of fillers inserted during speech. For example, it is the number of fillers that appear per unit time in speech. Fillers are, for example, "um", "uh", "ah", etc. The term selection is whether technical terms are used correctly or not. The silence count is an index based on the number of silences that occur during speech. For example, it is the total number of silences in a series of explanatory speeches. Silence means that the state of no speech continues for a predetermined time or more. If the silent time continues for more than the threshold, it is judged as silence and the number is counted. The ending pattern is the pattern of the ending part of a continuous speech. As an example, it is judged whether the ending is clear or unclear. The audio waveform frequency is the waveform or frequency of the spoken voice. When the emotion rises, the frequency of the spoken voice becomes higher. Figure 4 is a conceptual diagram showing the state of dialogue learning using the learning support system. The state of the avatar teacher and the learner having a dialogue is shown. On the information terminal 40, the avatar teacher 61 and the score display area 62 are displayed.

[0028] The spoken voice 63 of the avatar teacher 61 is output from the speaker 41 of the information terminal 40. The spoken voice 64 of the learner 80 is acquired by the microphone 42 and taken into the information terminal 40. Thereby, the dialogue between the avatar teacher 61 and the learner 80 is realized. In the score display area 62, the accuracy of the explanation regarding the learning items of the learner 80 is displayed in real time.

[0029] Returning to FIG. 3, in step S102, the accuracy determination unit 13 measures the accuracy of the explanation from the speech of the learner 80 based on the accuracy evaluation method information.

[0030] Figure 5 is a diagram showing an example of the accuracy evaluation method information (point addition). The accuracy evaluation method information 71a is information for calculating the accuracy of the explanation of the learning items. The method for evaluating accuracy varies depending on the type of learning items to be evaluated. The evaluation method adopted is a point - adding method where a predetermined number of points are added if a predetermined explanation can be given. The types of learning items include phrase explanations, institutional purposes, basic matters (5W1H), basic matters (chronological order), institutional comparisons, case problems, and case law knowledge.

[0031] A phrase explanation represents the content of the phrases used in the law that defines the institution. The evaluation method for phrase explanations is as follows. (1) If the content of a technical term can be accurately explained, a maximum of 80 points will be added. (2) Furthermore, if an example of how to use the technical term is shown or a metaphor is used to explain it in an easy - to - understand way, a maximum of 20 points will be added.

[0032] The institutional purpose is the purpose for which the institution is defined. The evaluation method for institutional purposes is as follows. (1) If the institutional purpose can be accurately explained, a maximum of 80 points will be added. (2) Furthermore, if the institutional purpose is explained with specific examples, a maximum of 20 points will be added.

[0033] Basic matters (5W1H) are matters regarding "who", "whom", "when", "what", "how" related to the institution. The evaluation method for basic matters (5W1H) is as follows. (1) If the outline of the institution can be explained in the order of "who", "whom", "when", "what", "how", points will be added for each matter, and a maximum of 5×15 = 75 points will be added. (2) If they can be explained in connection with the institutional purpose, a maximum of 25 points will be added.

[0034] Basic matters (chronological order) are multiple matters having a time sequence. The evaluation method for basic matters (chronological order) is as follows. (1) If multiple matters having a time sequence showing the outline of the institution can be explained in that order, points will be added for each matter, and a maximum of 75 points will be added. (2) If it can be explained in connection with the institutional purpose for those matters, up to 25 points will be added.

[0035] The comparison of systems is a comparison between two systems (the first system and the second system). The evaluation method of the system comparison is as follows. (1) If the differences can be explained by comparing the points to be compared between the first system and the second system, a predetermined number of points will be added for each point. If the differences can be explained for all points, up to 75 points will be added. (2) If the explanation of the difference includes an explanation of the reason based on the purpose of the system, a predetermined number of points will be added for each point. If the reason has been explained for all points, up to 25 points will be added.

[0036] The case problem is a problem of deriving a conclusion for a given case. The evaluation method of the case problem is as follows. (1) If it can be accurately explained what problem the case problem set by the avatar instructor asks, up to 25 points will be added. (2) If the approach to solving the case problem can be correctly explained, up to 25 points will be added. (3) If the conclusion of actually solving the case problem can be explained, up to 50 points will be added.

[0037] Case law knowledge is knowledge about case law. The evaluation method of case law knowledge is as follows. (1) If the background of what kind of case the questioned case law is can be explained, up to 25 points will be added. (2) If the issues of the case law can be explained, a predetermined number of points will be added for each issue. If all issues can be explained, up to 25 points will be added. (3) If the judgment made on the issue can be explained, an additional up to 50 points will be added.

[0038] Based on the above evaluation method, for each learning item, a plurality of words that should be included in the learner 80's explanation of the learning item, or the plurality of words and the order in which they appear, are preset in advance. By comparing the preset words with the words that appear in the learner 80's explanation of the learning item, it becomes possible to evaluate the learner 80's explanation of the learning item by grading. For example, when a plurality of words are set, a predetermined score can be added each time the word appears in the utterance. Also, for example, when a plurality of words and the order in which they appear are set, a predetermined score can be added each time the predetermined words appear in the predetermined order in the utterance.

[0039] Alternatively, a model explanation sentence may be created in advance such that a plurality of words appear or the words appear in a predetermined order, and grading may be performed based on the similarity (distance between vectors) between the sentence vector of the learner 80's explanation and the sentence vector representing the model explanation. Returning to FIG. 3, in step S103, the accuracy determination unit 13 updates the display in the score display area 62 based on the newly calculated accuracy.

[0040] FIG. 6 is a diagram showing an example of the display in the score display area. As shown in FIG. 6, in the score display area 62, the real-time accuracy value and its graph that are sequentially updated in synchronization with the dialogue are displayed. The entertainment value of being able to confirm that the score increases by showing correct understanding in the explanation can raise the motivation of the learner.

[0041] Returning to FIG. 3, if the explanation of the learning item has not ended in step S104, the process returns to step S102 to repeat the measurement of accuracy. If the explanation of the learning item has ended, in step S105, the accuracy determination unit 13 weights the accuracy with a weight corresponding to the learning item.

[0042] FIG. 7 is a diagram showing an example of accuracy evaluation method information (weighting). The accuracy evaluation method information 71b is information for adding weights to the accuracy of the explanations of learning items. The learning items have different levels of difficulty depending on their types. Therefore, weighting is performed according to the level of difficulty to reflect the level of difficulty in the evaluation of accuracy.

[0043] As described above, the types of learning items include word explanations, institutional purposes, basic matters (5W1H), basic matters (time series), institutional comparisons, case problems, and case law knowledge. Here, the value of the word explanation is multiplied by a weight of 1, the value of the institutional purpose is multiplied by a weight of 2, the value of the basic matter (5W1H) is multiplied by a weight of 4, the value of the basic matter (time series) is multiplied by a weight of 4, the value of the institutional comparison is multiplied by a weight of 6, the value of the case problem is multiplied by a weight of 8, and the value of the case law knowledge is multiplied by a weight of 10. Returning to FIG. 3, in step S106, the comprehension determination unit 16 calculates the comprehension level from the measured para - language information based on the comprehension evaluation method information. FIG. 8 is a diagram showing an example of comprehension evaluation method information.

[0044] In the comprehension evaluation method information 72, a method for calculating the comprehension level is defined for each para - language to be analyzed. Specifically, the method for calculating the index value of the comprehension level for each para - language is defined in FIG. 8. As described above, the para - languages to be measured include speech rate, number of fillers, term selection, number of silences, word - ending patterns, and audio waveform frequency.

[0045] Regarding the speaking speed, if the speaking speed is equal to or higher than the threshold value, the index value = A; if the speaking speed is lower than the threshold value, the index value = B (A and B are natural numbers where A > B). Regarding the number of fillers, if the number of fillers is equal to or less than the threshold value, the index value = C; if the number of fillers is more than the threshold value, the index value = D (C and D are natural numbers where C > D). Regarding the term selection, if the specialized knowledge is used correctly, the index value = E; if the specialized knowledge is not used correctly, the index value = F (E and F are natural numbers where E > F). Regarding the number of silences, if the number of silences is equal to or less than the threshold value, the index value = G; if the number of silences is more than the threshold value, the index value = H (G and H are natural numbers where G > H). Regarding the ending pattern, if the number of times the ending is unclear is less than the threshold value, the index value = I; if the number of times is equal to or more than the threshold value, the index value = J (I and J are natural numbers where I > J). Regarding the audio waveform frequency, the emotional value L (L is a natural number) of the learner measured based on the audio waveform is used as the index value. Here, as an example, the total value of the index values calculated by each para-language is regarded as the comprehensibility.

[0046] Returning to FIG. 3, in step S107, the comprehensive determination unit 19 calculates the proficiency indicating how proficient the learner is in the learning items based on the accuracy and the comprehensibility. The method for calculating the proficiency is not particularly limited, but as an example, the proficiency may be calculated by the following formula (1). This formula is based on the idea of taking into account the depth of understanding on the premise of having an accurate understanding.

[0047]

Number

[0048] Here, an example was shown in which proficiency considering the depth of understanding in addition to the correctness of understanding was calculated by adding the degree of understanding to the accuracy after weighting when the accuracy before weighting was equal to or greater than a certain value by multiplying the weights. However, the present invention is not limited to this. As another example, when the accuracy before weighting is equal to or greater than a certain value, proficiency considering the depth of understanding in addition to the correctness of understanding may be calculated by multiplying the degree of understanding by the accuracy after weighting. Further, for example, not only the accuracy but also the degree of understanding may be weighted according to the difficulty level.

[0049] Subsequently, in step S108, the ranking unit 21 determines the rank of the learner based on the proficiency and presents information regarding the rank to the learner. The ranking unit 21 always records the latest proficiency values of all learners and determines the rank of the learner among all learners based on the newly calculated proficiency of the learner.

[0050] Subsequently, in step S109, the growth visualization unit 22 records the number of times of learning performed so far and the proficiency calculated this time in association with the learning items learned this time, and performs a display that enables visual recognition of the state of change in proficiency each time the number of times increases. The learner 80 can confirm in comparison with others that their learning is progressing, and continuous learning is effectively supported.

[0051] FIG. 9 is a diagram showing a display example of a growth graph representing the state of change in proficiency. The growth graph 65 is, as an example, a line graph showing the proficiency with respect to the number of times of learning performed by teaching. The learner 80 can visually confirm that the proficiency improves each time the number of times of learning increases, and can maintain or improve motivation.

[0052] Returning to FIG. 3, in step S110, the feedback unit 23 provides a summary of learning such as encouragement, presentation of a model, and advice in accordance with the accuracy and depth of understanding of the learner 80 regarding the learning items and how proficient the learner is.

[0053] As an example, based on one or more of accuracy, comprehensibility, and proficiency, the feedback unit 23 determines whether to teach the learner 80 an exemplary explanation. If it is determined that an exemplary explanation should be taught to the learner 80, an exemplary explanation is provided to the learner 80 by means of the video and voice of the avatar instructor.

[0054] FIG. 10 is a diagram showing an example of the determination logic for whether to present an exemplary explanation. In this example, a threshold is set for the accuracy before weighting. If the accuracy before weighting is equal to or higher than the threshold, it is determined as "high", and if the accuracy before weighting is less than the threshold, it is determined as "low". Similarly, a threshold is set for comprehensibility. If the comprehensibility is equal to or higher than the threshold, it is determined as "high", and if the comprehensibility is less than the threshold, it is determined as "low". Then, as shown in the shaded area of FIG. 10, if either the accuracy before weighting or the comprehensibility is "low", it is determined that an exemplary explanation should be presented to the learner 80. On the other hand, as shown in the unshaded area of FIG. 10, if both the accuracy before weighting and the comprehensibility are "high", it is determined that there is no need to present an exemplary explanation to the learner 80. In that case, for example, encouraging words indicating that the learner understands well may be provided. When it is determined that it is desirable to further improve the accuracy of understanding or deepen the depth of understanding in the learning where the learner 80 learns by teaching, the learning effect of the learner 80 can be enhanced by the user experience of teaching the content and method of the exemplary explanation to the learner 80 during the dialogue.

[0055] In addition, based on at least one of accuracy, comprehensibility, and proficiency, the feedback unit 23 provides encouraging words to the learner 80 in the dialogue, so that the learner 80 can receive encouragement according to the progress of learning, and continuous learning is effectively supported.

[0056] The embodiments of the present invention described above are examples for explaining the present invention, and are not intended to limit the scope of the present invention only to those embodiments. A person skilled in the art can implement the present invention in various other modes without departing from the gist of the present invention.

[0057] In addition, the present embodiment includes each of the following matters. However, the matters included in the present embodiment are not limited to only those shown below.

[0058] (Matter 1) A dialogue unit that conducts a dialogue with the learner so that the learner explains a predetermined learning matter, An accuracy determination unit that determines the accuracy, which is the accuracy of the explanation in the dialogue, An understanding degree determination unit that determines the understanding degree, which is the depth of the learner's understanding of the learning matter, A learning support system having the above. According to this, in the learning method of learning by teaching oneself, since the accuracy and depth of understanding are determined from the learner's utterance, learning support that evaluates not only whether the explanation is accurate but also the depth of understanding becomes possible.

[0059] (Matter 2) In the learning support system according to Matter 1, The understanding degree determination unit calculates a paralinguistic index, which is one or more paralinguistic indices, from the utterance by the learner in the dialogue, and calculates the understanding degree based on the paralinguistic index. Thereby, the depth of the learner's understanding of the learning matter appearing in the paralinguistic in the learner's utterance can be accurately measured.

[0060] (Matter 3) In the learning support system according to Matter 2, The paralinguistic index includes one or more of an utterance speed, which is an index based on the speed of utterance, a filler count, which is an index based on the number of fillers inserted during the utterance, a silence count, which is an index based on the number of silences occurring during the utterance, and an utterance-final aspect, which is the aspect of the end of a continuous utterance.

[0061] (Matter 4) In the learning support system according to Matter 2, The accuracy determination unit converts the voice spoken by the learner into text, and calculates the accuracy by scoring based on the comparison between the words appearing in the text and the words predetermined for the learning items. Thereby, by evaluating the text obtained by converting the voice in which the learner explained the learning item, it is possible to accurately measure how accurately the learner understands the learning item.

[0062] (Item 5) In the learning support system according to Item 4, The learning items include a statement explanation representing the content of the terms used in the law that defines the system, a system purpose that is the purpose defined by the system, basic matters that are a plurality of basic matters related to the system, a system comparison that is a comparison between a first system and a second system, an example problem that is a problem for deriving a conclusion for a given example, and case knowledge that is knowledge about case law. The accuracy determination unit, as the accuracy, Regarding the statement explanation, points are added if the content of the statement can be correctly explained, points are added if the usage of the statement is exemplified, or further points are added if a metaphor is used in the explanation of the content of the statement. Regarding the system purpose, points are added if the purpose of the system can be correctly explained, and further points are added if the purpose of the system is explained by a specific example. Regarding the basic matters, points are added for each matter if a plurality of the matters can be explained in a predetermined order, and further points are added if the explanation is linked to the purpose of the system. Regarding the system comparison, points are added for each argument if the differences regarding a plurality of arguments to be compared between the first system and the second system can be explained, and further points are added if the explanation of the differences includes an explanation of the reason based on the purpose of the system. Regarding the example problem, points are added if the question asked by the given problem can be correctly explained, further points are added if the problem and the solution method can be correctly explained, and further points are added if the conclusion of solving the problem can be explained. Regarding the case knowledge, points are added if the background of what kind of case the case is can be explained, more points are added if the issue of the case can be explained, and still more points are added if the judgment issued regarding the issue can be explained.

[0063] (Item 6) In the learning support system described in Item 5, The accuracy determination unit updates the accuracy in parallel with the dialogue and presents the updated accuracy to the learner. According to this, the motivation of the learner can be increased by the entertainment nature of the user experience of the dialogue itself, which allows the learner to confirm that the score increases by showing correct understanding in the explanation.

[0064] (Item 7) In the learning support system described in Item 4, It further has an overall determination unit that calculates a proficiency level indicating how proficient the learner is in the learning item based on the accuracy and the comprehension level.

[0065] (Item 8) In the learning support system described in Item 7, The accuracy determination unit further weights the pre-weighted accuracy calculated based on the comparison between the words appearing in the text obtained by converting the voice spoken by the learner and the words predetermined for the learning item with a weight corresponding to the difficulty level of the learning item, thereby calculating the post-weighted accuracy. If the pre-weighted accuracy is equal to or higher than a predetermined threshold, the overall determination unit calculates the proficiency level by taking the comprehension level into account in the post-weighted accuracy. If the pre-weighted accuracy is less than the threshold, the post-weighted accuracy is taken as the proficiency level. According to this, it is possible to effectively support learning with an appropriate proficiency level by taking into account the depth of understanding appearing in paralanguage only when the correctness of understanding appearing in the text of the voice has reached a certain level or more.

[0066] (Item 9) In the learning support system described in item 7, Based on one or more of the accuracy, the comprehension, and the proficiency, it is determined whether to teach the learner an exemplary explanation. If it is determined that an exemplary explanation should be taught to the learner, the learning support system further includes a feedback unit that provides the exemplary explanation to the learner in the dialogue. Thereby, when it is desirable to increase the accuracy of understanding or deepen the depth of understanding in the learning in which the learner learns by teaching others, the learning effect can be further enhanced by the user experience of teaching the content and method of the exemplary explanation to the learner in the dialogue.

[0067] (Item 10) In the learning support system described in item 7, Corresponding to the learning item, the number of times of learning by teaching about the learning item and the proficiency calculated in that session are recorded, and a growth visualization unit is further provided that performs a visible display of the state of change of the proficiency each time the session is repeated. According to this, by enabling the learner to visually confirm that their learning is progressing, continuous learning can be effectively supported.

[0068] (Item 11) In the learning support system described in item 7, The ranking of the learner is determined by the proficiency, and the learning support system further includes a ranking unit that presents information regarding the ranking to the learner. According to this, by enabling the learner to confirm in comparison with others that their learning is progressing, continuous learning can be effectively supported.

[0069] (Item 12) In the learning support system described in item 7, Based on at least one of the accuracy, the comprehension, and the proficiency, the learning support system further includes a feedback unit that provides encouraging words to the learner in the dialogue. According to this, by enabling the learner to receive encouragement according to the progress of learning, continuous learning can be effectively supported.

[0070] (Item 13) A computer conducts a dialogue with the learner so that the learner explains a predetermined learning item, determines the accuracy, which is the accuracy of the explanation in the dialogue, and determines the degree of understanding, which is the depth of the learner's understanding of the learning item. A learning support method for performing the above.

[0071] (Item 14) Cause a computer to conduct a dialogue with the learner so that the learner explains a predetermined learning item, determine the accuracy, which is the accuracy of the explanation in the dialogue, and determine the degree of understanding, which is the depth of the learner's understanding of the learning item. A learning support program for causing the above to be executed.

Explanation of Signs

[0072] 10…Learning support system, 12…Dialogue unit, 13…Accuracy determination unit, 14…Voice analysis unit, 15…Text evaluation unit, 16…Degree of understanding determination unit, 17…Paralanguage measurement unit, 18…Paralanguage evaluation unit, 19…Comprehensive determination unit, 20…Motivation processing unit, 21…Ranking unit, 22…Growth visualization unit, 23…Feedback unit, 30…Server, 31…Processor, 32…Main memory, 33…Storage device, 34…Communication device, 35…Input device, 36…Display device, 37…Bus, 40…Information terminal, 41…Speaker, 42…Microphone, 43…Browser, 61…Avatar instructor, 62…Score display area, 63…Spoken voice, 64…Spoken voice, 65…Growth graph, 71a…Accuracy evaluation method information, 71b…Accuracy evaluation method information, 72…Degree of understanding evaluation method information, 80…Learner, 81…Generative artificial intelligence, 82…Communication network

Claims

1. a dialogue unit for dialogue with the learner so that the learner explains a predetermined learning matter; an accuracy determination unit that converts the speech of the learner into text and calculates the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; a comprehension level determination unit that calculates a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculates a comprehension level, which is a depth of the learner's understanding of the learning subject, based on the paralinguistic index; a comprehensive assessment unit that calculates a level of mastery indicating how well the learner has mastered the learning subject based on the accuracy and the level of understanding; having the accuracy determination unit further weights the unweighted accuracy calculated based on a comparison between words appearing in a text converted from the speech of the learner and words predetermined for the learning item with a weight corresponding to the difficulty level of the learning item to calculate a weighted accuracy; The comprehensive determination unit calculates the proficiency level by adding the degree of understanding to the accuracy after weighting if the accuracy before weighting is equal to or greater than a predetermined threshold, and calculates the proficiency level by using the accuracy after weighting if the accuracy before weighting is less than the threshold. Learning support system.

2. a dialogue unit for dialogue with the learner so that the learner explains a predetermined learning matter; an accuracy determination unit that converts the speech of the learner into text and calculates the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; a comprehension level determination unit that calculates a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculates a comprehension level, which is a depth of the learner's understanding of the learning subject, based on the paralinguistic index; a comprehensive assessment unit that calculates a level of mastery indicating how well the learner has mastered the learning subject based on the accuracy and the level of understanding; a feedback unit that determines whether or not an exemplary explanation should be given to the learner based on one or more of the accuracy, the comprehension, and the proficiency, and provides the exemplary explanation to the learner in the dialogue when it is determined that an exemplary explanation should be given to the learner; Learning support system.

3. The computer Conducting a dialogue with the learner so that the learner explains a predetermined learning matter; converting the speech of the learner into text, and calculating the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; Calculating a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculating a level of understanding, which is a depth of understanding of the learner about the learning subject, based on the paralinguistic index; calculating a mastery level indicating how well the learner has mastered the learning subject based on the accuracy level and the understanding level; In a learning support method for carrying out the above, The computer, a weighting step for calculating a weighted accuracy by weighting the unweighted accuracy calculated based on a comparison between words appearing in a text converted from the speech of the learner and words predetermined for the learning item with a weight corresponding to the difficulty level of the learning item; If the accuracy before weighting is equal to or greater than a predetermined threshold, the accuracy after weighting is added to the understanding level to calculate the proficiency level, and if the accuracy before weighting is less than the threshold, the accuracy after weighting is set to the proficiency level. Learning support methods.

4. The computer Conducting a dialogue with the learner so that the learner explains a predetermined learning matter; converting the speech of the learner into text, and calculating the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; Calculating a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculating a level of understanding, which is a depth of understanding of the learner about the learning subject, based on the paralinguistic index; calculating a level of mastery indicating how well the learner has mastered the learning subject based on the level of accuracy and the level of understanding; determining whether or not an exemplary explanation should be given to the learner based on one or more of the accuracy, the comprehension, and the proficiency, and when determining that an exemplary explanation should be given to the learner, providing the exemplary explanation to the learner in the dialogue; A learning support method that puts this into practice.

5. On the computer, Conducting a dialogue with the learner so that the learner explains a predetermined learning matter; converting the speech of the learner into text, and calculating the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; Calculating a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculating a level of understanding, which is a depth of understanding of the learner about the learning subject, based on the paralinguistic index; calculating a mastery level indicating how well the learner has mastered the learning subject based on the accuracy level and the understanding level; In this learning support program, The computer includes: a weighting step for calculating a weighted accuracy by weighting the unweighted accuracy calculated based on a comparison between words appearing in a text converted from the speech of the learner and words predetermined for the learning item with a weight corresponding to the difficulty level of the learning item; If the accuracy before weighting is equal to or greater than a predetermined threshold, the accuracy after weighting is added to the understanding level to calculate the proficiency level, and if the accuracy before weighting is less than the threshold, the accuracy after weighting is set to the proficiency level. A learning support program that helps students put this into practice.

6. On the computer, Conducting a dialogue with the learner so that the learner explains a predetermined learning matter; converting the speech of the learner into text, and calculating the accuracy of the explanation in the dialogue by adding points based on a comparison between words appearing in the text and words predetermined for the learning matter; Calculating a paralinguistic index, which is an index of one or more paralanguages, from the utterances by the learner in the dialogue, and calculating a level of understanding, which is a depth of understanding of the learner about the learning subject, based on the paralinguistic index; calculating a level of mastery indicating how well the learner has mastered the learning subject based on the level of accuracy and the level of understanding; determining whether or not an exemplary explanation should be given to the learner based on one or more of the accuracy, the comprehension, and the proficiency, and when determining that an exemplary explanation should be given to the learner, providing the exemplary explanation to the learner in the dialogue; A learning support program that helps students put this into practice.

Images