KR102035088B1 - Storytelling-based multimedia unmanned remote 1: 1 customized education system - Google Patents

Abstract

The present invention relates to a storytelling-based multimedia unmanned remote one-to-one customized education system. More specifically, the present invention relates to a storytelling-based multimedia unmanned remote one-to-one customized education system capable of enabling a teacher to talk about a teaching plan creating process for teaching students, enabling the teacher to provide various teaching methods by converting the created teaching plan into unmanned learning type such as chatbot, and performing learning according to the inclination and level of each learner (student) without limitation of time and space since the students choose personalized learning from the teaching methods.

Description

Storytelling-based multimedia unmanned remote 1: 1 customized education system}

The present invention relates to a storytelling-based multimedia unattended remote 1: 1 customized education system, and more specifically, to enable the teacher to storytell the teaching process for teaching students, and to prepare the teaching material in the form of unattended learning such as chatbots. By transforming, teachers can easily provide a variety of teaching methods, and students can select learning that suits their personal tendency from various teaching methods, so that students can perform learning that suits each student's personality and level without limitation of time and space. The present invention relates to a telling-based multimedia unmanned remote 1: 1 customized education system.

A chatbot is an interactive messenger that enters a question as if it were chatting to a messenger, and the AI analyzed the question and gave the appropriate answer.

Messenger, which introduced the chatbot function, can check and process information in the chat window without launching a separate app, so it can be used as a platform for connecting various functions.

The chatbot was serviced at the level of providing a predetermined answer according to a pre-input algorithm. However, with the development of natural language analysis and processing technology along with big data processing technology, the chatbot is providing an optimal answer considering various variables.

In general, a chatbot processes a language input composed of text or voice of a chat into a form that can be processed on a computer by receiving a natural language processor (NLP) module, and processes the context of the processed natural language. After passing through the Context Analyzer that analyzes the data, Decision Maker classifies the answers according to the context and confirms the answers. Then, the verbal output comes out through the response generator that answers the user according to the confirmed answers. It becomes the structure that becomes.

In addition, recently, in the entry into the fourth industry, the problem distance such as the conventional overfitting in the machine learning, reinforcement learning, and deep learning in the artificial intelligence field has been solved, and thus, the development in the intelligent field has been remarkably progressed.

In addition, the interactive interface technology has also attracted attention, and natural language processing, conversation management methods, and related technologies have been developed.

Natural language processing techniques necessary for natural conversation processing with humans include morphological analysis and classifying objects, so that the elements necessary to infer the intention of the speaker can be classified in the sentence, and the objects are defined in advance. Technology has helped to answer the speaker's intentions and extract the necessary elements.

In recent years, many chatbot production technologies have emerged, and the interactive interface production technology required for its production has begun to emerge, and the technology for managing conversations is still at a simple level.

On the other hand, in the current information and information society, an educational paradigm that emphasizes the effective learning of learners rather than the teacher-centered knowledge transfer emphasized in the previous industrial society is becoming common.

Accordingly, the characteristics and educational needs of learners are emphasized as important elements of effective education.

In particular, with the rapid development of information technology, various multimedia educational media and educational methods are being developed, and at the same time, it is possible to provide students with various learning activities.

An example is unmanned distance education, which is being actively researched recently.

Conventional solutions for building an unattended distance learning environment required programming knowledge.

Conventional solutions were difficult to access for front-line teachers who did not know the program, and had difficulty in communicating with learners with the teaching experience of native teachers.

One of the central elements for unattended distance learning is the construction of systematic learning data for the curriculum.

In the learning contents, the essential elements for the learners and the contents of the curriculum should be included as components, which should be educational contents considering the characteristics of the learners.

These contents are backed by the storytelling of experienced teachers and, when reconstructed, can produce great educational synergies, but the technology for this is only a small step.

Thus, by enabling the teacher to storytell the process of drafting lesson plans for teaching students, and by converting the written lessons into unattended forms of learning, such as chatbots, teachers can easily provide a variety of teaching methods, and students can learn from individual teaching methods. By selecting the appropriate learning, we have devised a storytelling-based multimedia unmanned remote 1: 1 customized education system that can perform learning according to the inclination and level of individual learners (students) without any limitation of time and space.

(Previous Document 1) Republic of Korea Patent No. 10-1619748 (Previous Document 2) Republic of Korea Patent No. 10-1851786

Therefore, the present invention is to solve the above conventional problems,

A first object of the present invention is to enable the teacher to storytell the teaching process for teaching the student, and by converting the written teaching into unattended learning forms such as chatbots, the teacher easily provides a variety of teaching methods, the student By selecting learning according to individual tendency from teaching method, it is to provide storytelling based multimedia unmanned remote 1: 1 customized education system that can perform learning according to individual's tendency and level without restriction of time and space.

The second object of the present invention is to create an unattended learner (chatbot, etc.) for each teacher only by writing (story telling) a story that will occur in a real class without considering the production and operation principle of SW capable of unmanned learning such as a chatbot. It is intended to provide the convenience and efficiency of production that can be produced.

The third object of the present invention is to automatically analyze and machine-learn texts (storytelling results) of a plurality of similar unmanned learners, thereby teaching various subjects and learning contents corresponding to various responses of various learners and individual learners. The goal is to create and optimize unmanned learners.

In order to solve the above problems, storytelling based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention,

Using the template information provided from the storytelling-based customized education chatbot providing server 1000, implement a teaching plan view (web page or app environment view) suitable for the teacher terminal platform environment on the screen, and by teacher implemented through the implemented lesson plan view A teacher terminal 2000 for creating a contextual storytelling syllabus;

 Provides template information for creating a storytelling lesson plan to the teacher terminal (2000), reconstructs the storytelling lesson text provided from the teacher terminal into the chatbot form using machine learning, and reconstructs the storytelling lesson plan reconstructed in the chatbot form into the student terminal And a storytelling-based customized education chatbot providing server 1000 for providing feedback on the request of the student terminal through the service processing engine configured to be distributed to 3000;

Student terminal (3000) for displaying the storytelling text reconstructed in the form of chatbots distributed from the storytelling-based customized education chatbot providing server 1000, and providing request information to the storytelling-based customized education chatbot providing server 1000. It includes.

According to the present inventor storytelling-based multimedia unmanned remote 1: 1 customized education system,

By allowing teachers to tell the storytelling process for teaching students, and by converting written texts into unattended forms of learning, such as chatbots, teachers can easily provide a variety of teaching methods, and students can adapt their personality from various teaching methods. By choosing the learning, it is possible to perform the learning that suits the inclination and the level of the learner (student) individual without the limitation of time and space.

In addition, it is possible to create an unmanned learning machine (chatbot, etc.) for each teacher only by writing (story telling) the story that will occur in the actual class without considering the creation and operation principle of SW capable of unmanned learning such as chatbots. It will provide convenience and efficiency.

In addition, by automatically analyzing and machine learning the texts (storytelling results) of many similar unmanned learners, we can create unmanned learners that can teach various subjects and learning contents corresponding to various responses of various learners and individual learners. The effect will be optimized.

As a result, teachers can connect with students who follow their teaching methods and provide customized learning contents without the constraints of time and space outside the classroom, and students can learn and teach teachers' teaching methods and learning in accordance with their personal disposition, level, and satisfaction. By selecting and learning content, it provides a synergistic effect that can contribute to improving the teaching-learning quality of teachers-students through distance learning.

1 is an example of Quibble that is a quibbly answer to speech that is not recognized by an unmanned learner such as a chatbot.
2 is an overall configuration of a storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
3 is a block diagram of the learning planner 100 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
Figure 4 is a block diagram learning flow analysis unit 200 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
5 is a block diagram of the unmanned learner generation unit 300 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
6 is a block diagram of a natural language processing engine 400 of a storytelling-based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention.
7 is a block diagram of a service processing engine 500 of a storytelling-based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention.
8A is an exemplary view illustrating a storytelling unit 130 screen of a learning planner of a storytelling-based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention, and FIG. 8B is a preview screen example of a created storytelling. Degree.
9 is an exemplary view of a screen processed by the operation of the components of the learning flow analysis unit 200 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
10 is an exemplary view of a simulator screen generated by the unmanned learner generation unit 300 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.
11 is an exemplary screen output through the unmanned learner of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.

The following merely illustrates the principles of the invention. Therefore, those skilled in the art, although not explicitly described or illustrated herein, can embody the principles of the present invention and invent various devices that fall within the spirit and scope of the present invention.

In addition, all conditional terms and embodiments listed herein are in principle clearly intended to be understood only for the purpose of understanding the concept of the invention and are not to be limited to the specifically listed embodiments and states. do.

In describing the present invention, terms such as first and second may be used to describe various components, but the components may not be limited by the terms.

For example, without departing from the scope of the present invention, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component.

When a component is referred to as being connected or connected to another component, it may be understood that the component may be directly connected to or connected to the other component, but there may be other components in between. .

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention, and singular forms may include plural forms unless the context clearly indicates otherwise.

In this specification, the terms including or including are intended to designate that there exists a feature, a number, a step, an operation, a component, a part, or a combination thereof described in the specification, and one or more other features or numbers, It can be understood that it does not exclude in advance the possibility of the presence or addition of steps, actions, components, parts or combinations thereof.

Storytelling based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention,

Using the template information provided from the storytelling-based customized education chatbot providing server 1000, implement a teaching plan view for the teacher terminal platform environment on the screen, and create a storytelling lesson for each teacher and situation through the implemented lesson plan view. Teacher terminal (2000) to do,

 Provides template information for creating a storytelling lesson plan to the teacher terminal (2000), reconstructs the storytelling lesson text provided from the teacher terminal into the chatbot form using machine learning, and reconstructs the storytelling lesson plan reconstructed in the chatbot form into the student terminal (1000) and a storytelling-based customized education chatbot providing server (1000) for providing feedback on the request of the student terminal through the configured service processing engine,

The storytelling-based customized educational chatbot providing server 1000 displays the chatbot-type storytelling syllabus, and includes a student terminal 3000 for providing request information to the storytelling-based customized educational chatbot providing server 1000. Characterized in that the configuration.

In addition, the storytelling-based customized training chatbot providing server 1000,

Provides template information for creating a storytelling syllabus to the teacher terminal 2000, and using the storytelling syllabus provided from the teacher terminal 2000, dialogue and learning pictures, voices, and videos between students and teachers that may occur in a classroom situation. Learning lesson planner 100 for creating a storytelling lesson plan information including text information in a storytelling manner according to the learning flow;

A learning flow analysis unit for analyzing the storytelling instructional information created by the storytelling method according to each learning flow created by the learning syllabus creation unit 100 and storing it in the internal data processing repository 600 in a form that can be programmed. 200;

An unmanned learner generation unit 300 for reconfiguring the storytelling instructional information stored in the internal data processing storage 600 into a chatbot form using machine learning and simulating whether or not it operates properly;

When the analysis is requested by the learning flow analysis unit 200, the unmanned learning unit generation unit 300, and the service processing engine 500, at least one natural language analysis of morphological analysis, entity name analysis, intention analysis, expectation analysis, and emotion analysis is performed. A natural language processing engine 400 for performing natural language processing on the request of the student terminal and providing the natural language processing to the service processing engine 500;

A service processing engine 500 for providing or providing the storytelling syllabus information reconstructed in a chatbot form stored in the service storage 700 to the student terminal 3000 and providing feedback on the request of the student terminal;

An internal data processing repository 600 for storing information of conversations and learning pictures, voices, videos, texts, and storytelling instructional information between students and teachers, which may occur in the class situation;

And a service storage 700 for storing the storytelling instructional information reconstructed in a chatbot form generated by the unmanned learner generation unit 300 to service.

In addition, the learning syllabus creation unit 100,

A basic template providing unit 110 for generating basic template information on a basic template of a learning flow (teaching plan) having a common flow for each education field and content for providing as a teacher terminal and providing the same as a teacher terminal;

A recommendation template providing unit 120 for generating recommendation template information for a recommendation template of a learning flow (teaching plan) having a common flow for each education field and content for providing as a teacher terminal and providing the same to a teacher terminal;

Using the storytelling syllabus provided from the teacher terminal 2000, the storytelling syllabus information including picture, voice, video, and text information for dialogue and learning between students and teachers, which can occur in the class situation, is prepared by storytelling according to the learning flow. Storytelling creation unit 130; characterized in that comprises a.

In addition, the learning flow analysis unit 200,

Outputs a list of conversations constituting a learning flow (teaching plan), instructs the linguistic information analysis unit to perform analysis on each storytelling unit data, and a story for receiving analysis result information analyzed by the linguistic information analysis unit. A telling navigation unit 210;

A language information analyzer 220 for performing analysis on each storytelling unit data according to the instruction of the storytelling navigation unit 210 and providing analysis result information to the storytelling navigation unit 210;

It provides annotation data, annotation system selection unit 230 for performing a modification to the selected annotation data; characterized in that it comprises a.

In addition, the unmanned learning unit generation unit 300,

Unmanned learning experiment unit (310) for extracting the storytelling instructional information that is actually distributed among the storytelling instructional information stored in the internal data processing storage 600, reconfiguring the chatbot form using machine learning, and simulating whether it works properly. );

The unmanned learner distribution unit 320 for storing the storytelling syllabus information reconstructed in a chatbot form in the service storage 700 when it operates properly, the test result of the unmanned learner experiment unit; characterized in that it comprises a .

In addition, the natural language processing engine 400,

Natural language analysis unit for performing any one of the natural language analysis of the morphological analysis, entity name analysis, intention analysis when the analysis request from the learning flow analysis unit 200, unmanned learning machine generation unit 300, service processing engine (500) ( 410);

An analysis data provider 420 for providing the analysis data analyzed in the natural language through the natural language analysis unit as one requested by the analysis of the learning flow analysis unit 200, the unmanned learning unit generation unit 300, and the service processing engine 500; Characterized in that it comprises a.

In addition, the service processing engine 500,

A speech selector for determining whether or not the response information obtained from the student terminal 3000 is a free speech that deviated from the learning flow (training plan) and selecting similar storytelling data within the same classification and providing the same to the student terminal in the case of free speech. 510);

And a service processor 520 for acquiring the learning data request signal from the student terminal 3000 and providing the response information to the student terminal.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the present invention storytelling-based multimedia unmanned remote 1: 1 customized education system.

Solution A term used for the description in the present invention refers to software or hardware used to solve a problem.

In addition, the term used in the present invention, Quibble refers to a murmuring response to speech that is not recognized by an unmanned learner such as a chatbot, as shown in FIG. 1, 'The manager likes sweet foods such as chocolate or cake. If you answer 'Tongyong?', The answer you didn't expect, you'll get a messy response instead of a matching answer to the unexpected answer, 'Tongyong?', Which is called Quibble. will be.

For example, the manager asks, 'Do you like sweet foods like chocolate or cake?' Without asking 'like / dislike', the answer the learner expects (the answer currently in the lesson), and not expecting it. If you answer 'Yong Yong?', The service processing engine is talking about whether you like sweet food for this answer ('Yong Yong?') That doesn't even exist in similar lessons. Let's answer again? ',' Will we focus again? ' The service processing engine makes a quizble that is not awkward in context, rather than a matched response to an unexpected answer, such as 'Tongyong?'.

In addition, storytelling refers to conveying what you want to tell in a story form through words, images, and sounds, and annotations give meaning to objects such as text, images, and videos. The action of giving the word or image the name 'fruit'.

And, machine learning means learning to increase performance by using a steady experience for a task, in the present invention, experience is data, performance is accurate feedback.

As the definition shows, the more quality data you have, the better the performance.

In addition, the learning flow (teaching plan) means all the text, audio, video, and image data that the teacher writes according to the flow of the learning contents and the utterance and the expected feedback for the learner to deliver to the learner in the virtual classroom environment. .

And, the conversation flow (speech (pair)) means that the teacher or learner in the learning flow and the relationship between the dialogue text data.

In this case, the present invention enables a teacher to tell the storytelling process for teaching a student, and by converting the written lesson into an unmanned learning form such as a chatbot, the teacher easily provides various teaching methods, and the student provides various teaching methods. By selecting the learning according to the personal tendency from the, it is possible to perform the learning according to the inclination and level of each learner (student) without the limitation of space and time.

2 is an overall configuration diagram of a storytelling based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention.

As shown in FIG. 2, the storytelling-based multimedia unattended remote 1: 1 customized education system of the present invention includes a storytelling-based customized education chatbot providing server 1000, a teacher terminal 2000, and a student terminal 3000. The teacher terminal 2000 and the student terminal 3000 may be configured in plural.

As shown in FIG. 2, the teacher terminal 2000 uses a template information provided from the storytelling-based customized education chatbot providing server 1000 to create a teaching plan for a teacher terminal platform environment (web page or app environment view). Is implemented on the screen, and the storytelling syllabus for each teacher and situation is prepared through the implemented teaching preparation view.

For example, as shown in FIG. 3, the storytelling-based customized education chatbot providing server 1000 provides template information for creating a storytelling lesson to the teacher terminal 2000, and the teacher terminal 2000. Using the provided template information, implement a teaching plan view (web page or app environment view), which is a template UI suitable for the teacher terminal platform environment, and through the implemented teaching plan view, the teacher creates a situational storytelling lesson. The completed contextual storytelling lesson plan is to be stored and managed in the storytelling-based customized education chatbot providing server 1000.

In addition, the storytelling-based customized education chatbot providing server 1000 obtains the storytelling instruction provided from the teacher terminal 2000, generates storytelling instructional information analyzed for each learning flow on the acquired storytelling instruction. In addition, the generated storytelling syllabus information is reconfigured in the form of a chatbot using machine learning, distributed to the student terminal 3000, and feedback on the request of the student terminal is provided through the configured service processing engine.

Conventionally, the situation in which a teacher transmits specific knowledge to a student in a classroom environment (real world) is limited in time and space, and there are various ways of teaching and learning the same knowledge.

Offline (real-world) education with space-time constraints is difficult to arbitrarily select the teaching method that suits the learner's inclination, and the teacher has a limitation in trying various teaching methods.

However, using the present invention, by automatically analyzing and machine learning the text (storytelling output) of a number of similar unmanned learners (e.g., chatbots), it is possible to respond to various reactions of various learners and individual learners. It is effective to create and optimize an unmanned learner that can teach various subjects and learning contents.

As a result, teachers can connect with students who follow their teaching methods and provide customized learning contents without the constraints of time and space outside the classroom, and students can learn and teach teachers' teaching methods and learning in accordance with their personal disposition, level, and satisfaction. By selecting and learning content, it provides a synergistic effect that can contribute to improving the teaching-learning quality of teachers-students through distance learning.

In this case, the student terminal 3000 displays a chatbot-type storytelling syllabus distributed by the storytelling-based customized education chatbot providing server 1000.

Therefore, through the student terminal, the student selects a specific storytelling plan in consideration of his or her personal inclination and level, and proceeds with the learning.

After that, if a student has a question, the request information including the question is provided to the storytelling-based customized education chatbot providing server (1000).

Next, with reference to Figure 2 will be described in detail with respect to the components and operation process of the storytelling-based customized education chatbot providing server 1000 according to an embodiment of the present invention.

As described above, the storytelling-based customized education chatbot providing server 1000 is a part of computer assistant language learning (CALL) that uses a computer to learn a language, particularly a language other than the native language, and is a natural field of artificial intelligence technology. By utilizing natural language processing (NLP), native teachers can provide unmanned learners, such as chatbots, which enable unattended learning (learning without teachers) of storytelling data that writes virtual class situations.

That is, as shown in Figure 2, the storytelling-based customized education chatbot providing server 1000 of the present invention is the learning syllabus creation unit 100, learning flow analysis unit 200, unmanned learning machine generation unit 300, natural language The processing engine 400, the service processing engine 500, the internal data processing storage 600, and the service storage 700 are configured to be included.

Specifically, the learning syllabus creation unit 100 provides template information for creating a storytelling syllabus to the teacher terminal 2000, and uses the storytelling syllabus provided from the teacher terminal 2000 in a class situation. Storytelling instructional information, including picture and voice, video, and text information for dialogue and learning between students and teachers, may be generated by storytelling according to the learning flow.

That is, through the learning syllabus creation unit 100, the teacher terminal 2000 provides the template information (data information to be displayed in the teaching plan view for creating a storytelling syllabus), and the teacher provides the teacher terminal 2000. It is to create multimedia as storytelling according to the learning flow (teaching) by teacher and situation by teacher and situation.

In addition, the learning flow analysis unit 200 analyzes the storytelling syllabus information prepared by the storytelling method according to each learning flow created by the learning syllabus creation unit 100, and stores the internal data processing in a form capable of program processing. Will be stored in.

That is, the learning flow analysis unit 200 checks and edits the automatically analyzed storytelling data (training plan) of the learning flow, respectively, and stores the data in an internal data processing repository in a machine-readable form.

And the unmanned learning machine generation unit 300 reconstructs the storytelling syllabus information stored in the internal data processing storage 600 by machine learning by the processing of the learning flow analysis unit in a chatbot form, and simulates whether it works properly. Will be.

For example, the automatically analyzed learning flow (teaching plan) reconstructs the stored storytelling data (teaching plan) into a chatbot or the like according to a series of processing algorithms such as machine learning or automata, which is expected from an unmanned learning machine (chatbot). Whether it works as it can be simulated in the unmanned learning unit generation unit 300, it can be distributed in the form of an unmanned learning unit such as the Web (Web) or App (Mobile App).

And, the natural language processing engine 400, when the analysis request from the learning flow analysis unit 200, unmanned learning unit generation unit 300, service processing engine 500, morphological analysis, entity name analysis, intention analysis, expectation analysis, emotion Natural language analysis is performed on at least one of the analyzes, and natural language processing is performed on the request of the student terminal to provide the service processing engine 500.

That is, when the analysis request from the learning flow analysis unit 200, the unmanned learning unit generation unit 300, the service processing engine 500, at least one of morphological analysis, entity name analysis, intention analysis, expectation analysis, emotional analysis Natural language analysis is performed, and natural language processing such as morphological analysis, individual name analysis, intention analysis, etc. of the request of the student terminal is performed.

At this time, the distributed unmanned learner (chatbot) obtains a user's request through a service processing engine and feeds back a natural language processing result.

In addition, the service processing engine 500 provides the storytelling instructional information reconstructed in the form of chatbots stored in the service storage to the student terminal 3000, or provides the service to the student terminal 3000. After performing the natural language processing through the natural language processing engine 400 for the student's request, the result of the request is to be fed back.

In addition, the internal data processing storage 600 stores information of conversations and learning pictures, voices, videos, texts, and storytelling instructional information between students and teachers, which may occur in a class situation, and stores a service storage 700. It is to store the storytelling syllabus information reconstructed in the form of chatbot generated by the unmanned learning unit generator to service.

In the prior art, in order to make an unmanned learner such as an educational chatbot, teachers needed programming knowledge to make an unmanned learner such as learning contents, class situations, and chatbots.

In addition, such a method is easy to produce a chatbot that performs a fixed workflow or a short response to perform a question and answer of goods, tasks, etc., but it is difficult to develop a chatbot that performs a variety of long conversation flow.

More specifically, in order to develop a chatbot in general, the entire scenario including the processing of input data (speech) according to the chat flow to be performed by the chatbot, feedback actions and conditions for each user's utterance, and the conversation flow should be written together.

As a result, the chat flow of the target chatbot and the response of the chatbot are significantly different than expected, or it takes a long time to develop the target chatbot.

In addition, due to such a burden, it is difficult to implement and utilize a chatbot having a variety of conversation flows on a large scale.

However, through the present invention based on storytelling, a teacher can concentrate on writing a learning flow through a virtual conversation with a student during class time, and simulates the storytelling data processed by a machine learning method to program or software. It provides the advantage of making unmanned learners such as chatbots without consideration.

In more detail, the teacher stores the template and storytelling data corresponding to the learning flow through the learning flow creation unit using a device such as a PC or a smartphone in the internal data processing storage 600. By using the learning flow analysis unit, the storytelling data is given meaning to store in the internal data processing storage 600 and the service storage 700.

At this time, in the process of assigning meaning, the natural language processing engine 400 undergoes a procedure of analyzing storytelling data into morphemes, entities, intentions, expectations, and emotions.

After the analysis and editing is finished, you can simulate the learning flow created so far in the unmanned learning unit generation unit 300.

If it works as expected, it distributes the unmanned learner, such as chatbots through the unmanned learner distribution unit 320 of the unmanned learner generation unit (300).

If the storytelling data does not work as expected due to missing or incorrect analysis, the storytelling unit 100 or the learning flow analysis unit 200 may modify the storytelling data or edit and re-analyze the analysis data. You can repeat the rating process.

The learner (student) may use the unmanned learner distributed through the unmanned learner distribution unit 320 of the unmanned learner generation unit 300.

If the learner leaves the learning flow by answering or answering something completely different from the storytelling data, he or she can continue the study using feedback of similar sentences in the same category stored in the service repository and use the main information of the spoken text. Even if you leave speech acts such as answering Quibble, you will be able to proceed naturally.

Through the configuration and operation as described above, a teacher's personal unmanned learning machine (chatbot, etc.) is simply written (story telling) the story to occur in a real class without considering the production and operation principle of SW capable of unmanned learning such as a chatbot. It is to provide an advantage that can be produced.

Next, with reference to Figures 3 to 7 will be described in detail the components and operation process of the storytelling-based customized education chatbot providing server 1000 according to an embodiment of the present invention.

3 is a block diagram of a learning planner 100 of a storytelling-based multimedia unattended remote 1: 1 customized education system according to an exemplary embodiment of the present invention.

As shown in FIG. 3, the basic template providing unit 100 includes a recommendation template providing unit 110, a template recommendation unit 120, and a storytelling creation unit 130.

Specifically, the basic template providing unit 110 generates basic template information for the basic template of the learning flow (teaching plan) having a common flow for each education field and content for providing to the teacher terminal and provides the same to the teacher terminal. .

The basic template information generated by the basic template providing unit 110 may include various stories. For example, the template information corresponding to Story 1 includes sub-categories such as introduction, greeting, topic selection, topic start, Korean practice, learning content evaluation, learning summary, conversation end, vocabulary practice, experience confirmation, and topic practice. The template information corresponding to Story 2 will include subcategories such as vocabulary practice and experience verification.

The recommendation template providing unit 120 generates recommendation template information for a recommendation template of a learning flow (teaching plan) having a common flow for each education field and content for providing as a teacher terminal, and provides it to the teacher terminal.

For example, if the teacher's educational field is Korean language education, suggest a recommendation template corresponding to Korean language education (e.g., an exemplary template written by another teacher) so that the teacher can write a storytelling lesson using the basic template or use the recommended template. This is to provide the convenience of choice for writing a storytelling lesson.

The storytelling preparation unit 130 uses storytelling teaching materials provided from the teacher terminal 2000 to provide storytelling teaching information including conversational and learning pictures, voice, video, and text information between students and teachers, which may occur in a classroom situation. According to the learning flow, the composition is written in a storytelling manner.

On the other hand, when creating a chat flow for a chatbot, the teacher needs to consider the outline for storytelling, the contents of learning (vocabulary, grammar), and the level of the learner, so that the chatbot can judge the development of the chat flow. It is also necessary to define changes in speech acts.

In this case, it is not only complicated for a teacher to compose such information together with storytelling, but also lack of consistency for computer processing.

Therefore, by configuring the learning syllabus creation unit 100, the teacher uses a template (teacher terminal) such as a PC or a smartphone to provide the template information (provided by the basic template providing unit 110 or the recommended template providing unit 120 ( Basic template information or recommendation template information) can be used to create a storytelling textbook of learning flows such as introduction of unmanned learners, greetings, topic selection, and topic progression.

That is, the teacher uses the template information (basic template information or recommendation template information) provided by the basic template provider 110 or the recommended template provider 120 through the teacher terminal to create a common flow for each education field and content (teaching plan). He can write his own storytelling lessons, or he can write storytelling lessons in the form of exemplary storytelling lesson plans written by other teachers.

The storytelling preparation unit 130 uses storytelling teaching materials provided from the teacher terminal 2000 to provide storytelling teaching information including conversational and learning pictures, voice, video, and text information between students and teachers, which may occur in a classroom situation. Write in storytelling style according to the learning flow.

That is, they set metadata such as the subject of learning (e.g. shopping, hospital, etc.), the main functions (e.g. apologize, explain, etc.) covered by the subject, the level of learners, the vocabulary and grammar to learn, Write storytelling instructional information, including text, images, voice, and video of incoming conversations.

When the creation of the storytelling instructional information is completed and stored, metadata information and storytelling instructional information data are stored in an internal data processing repository.

Traditionally, for storytelling, the entire outline of the lesson (the content of the lesson) had to be written in batches, or the teacher (or chatbot developer) had to program the appropriate learning flow, dialogue flow, and user feedback for each conversation.

However, it can be easily delivered by the learning syllabus creation unit 100 to accurately deliver the storytelling structure suitable for the individual class and machine-readable.

8A is an exemplary view of the storytelling creation unit 130 screen of the learning syllabus creation unit 100, and FIG. 8B is an exemplary view of a preview screen of the created storytelling.

As shown in Fig. 8A, for example, the metadata information includes the following classes (class 2, 3, 4, 5), detailed classes (upper, middle, lower), subject classification (category), function , Categories including learning objectives, vocabulary, grammar and expressions, next class, and class information written by teachers on each category.

And, using the template information provided by the storytelling-based customized education chatbot providing server (1000), the teaching plan for the implementation of the teaching plan to fit the teacher terminal platform environment, for example, greetings, topic selection, topic start, topic progress , Korean practice, learning contents evaluation, learning contents summary, end of conversation, etc., and the teacher writes class information in each category, and at the bottom, selection buttons such as preview, temporary storage, and saving are provided. Done.

At this time, when the teacher selects the preview, as shown in FIG. 8B, the created class information is output on the screen so that the teacher can check.

In other words, the lesson information, grade, creation date, category, detailed grade, author, function, learning objectives, learning stage, introduction-precautions, greetings, development-topic selection, topic start, etc. .

4 is a block diagram of the learning flow analysis unit 200 of the storytelling based multimedia unattended remote 1: 1 customized education system according to an exemplary embodiment of the present invention.

As shown in FIG. 4, the learning flow analysis unit 200 includes a storytelling navigation unit 210, a language information analysis unit 220, and an annotation system selection unit 230.

Specifically, the storytelling navigation unit 210 outputs a conversation list constituting a learning flow (training plan), instructs the language information analysis unit to perform analysis on each storytelling unit data, and provides language information. The analysis result 220 receives the analysis result information. That is, the storytelling data drawn through the storytelling navigation unit is shown.

For example, as shown in the example screen on the left of Figure 9, Teacher: Where are you meeting ?, Student: Gathering at Guro-dong bowling alley, Teacher: Well done, Teacher: Think about it, Teacher: When are you meeting ?, Student: Every Friday It is to provide the original text of the syllabus.

The language information analysis unit 220 analyzes each storytelling unit data according to the instruction of the storytelling navigation unit 210 and provides the analysis result information to the storytelling navigation unit 210.

That is, the storytelling data analyzed by the natural language processing engine or the storytelling analysis data drawn from the internal data processing storage 600 may be modified or confirmed.

For example, as shown in the central example screen of FIG. 9, language analysis information, an object, and an intention category are provided, and the language analysis information includes data fields such as numbers, words, morphemes, tags, and the like. Data fields such as names and tags are included, and intents include data fields such as numbers, keywords, intents, expectations, and feedback.

The annotation system selection unit 230 provides annotation data and performs correction on the selected annotation data. When the analysis data is modified, the annotation system selection unit 230 utilizes data provided by the annotation system selection unit.

For example, as shown in the example screen on the right side of FIG. 9, subcategories including VC (designated), M (formula), I (independent), J (relational), and X (dependent) are included in the annotation system. Will be done.

On the other hand, conventionally, since natural language analysis is performed in order to predict feedback (sentences to be chatbot spoken) from learning data (prepared scenarios or large corpuses), an appropriate dialogue flow may be performed even if it does not correspond to the intention of the teacher (chatbot developer). It was difficult to control.

This process also made it difficult for the chatbot to utter the correct learning flow.

However, by configuring the learning flow analysis unit 200, the storytelling data is first drawn out from the internal data processing storage 600.

Subsequently, the extracted storytelling data is output to the storytelling navigation unit 210, and when the unit data is selected, an analysis data of the storytelling data is attempted to be extracted to the internal data processing storage 600.

If there is analysis data, it is withdrawn and output to the language information analysis unit 220. If not, the natural language processing engine 400 requests an analysis operation and outputs the analyzed result to the language information analysis unit.

Analytical information gives meaning to data such as text, images, and images such as morphological analysis, individual name analysis, intention analysis, expectation analysis, and emotion analysis.

For morphological analysis and individual name analysis, for example, the sentence `` today I want to eat an apple '' is morphologically analyzed in the form of `` today / silver / apple / ga / eat / go / want / da ''. An apple is given an object name of 'fruit' to the stem, and an object name of 'time' is given to the 'today' stem.

In addition, additional meanings or corrections may be made to the automatically analyzed data.

In the modification, the analyzed text may be directly modified or the annotation data (morpheme, object, intention, expectation, emotion, etc.) may be modified by selecting a desired annotation among various annotations presented in the annotation system selection unit 230.

After the learning flow analysis is finished, the analysis data is stored in the storage 600 for internal data processing.

Therefore, the teacher (writer) can transmit to the system to accurately reproduce the intended learning flow by allowing the utterance text created by the learning flow creator to output and edit the morphemes, objects, and intention analysis results.

In addition, by separating storytelling and linguistic analysis, it is possible to easily and accurately create a learning flow faithful to storytelling through editing of annotations without depending on the implementation as a chatbot.

5 is a block diagram of the unmanned learner generation unit 300 of the storytelling-based multimedia unmanned remote 1: 1 customized education system according to an embodiment of the present invention.

As shown in FIG. 5, the unmanned learner generation unit 300 includes an unmanned learner experiment unit 310 and an unmanned learner distribution unit 320.

In the prior art, the most difficult thing in creating an unmanned learner such as a chatbot is to consider and reflect an appropriate response to feedback of various users (learners).

In particular, much experience, training data, and expertise are required to enable consistent and flexible responses to deviations from centralized learning flows, unexpected user responses, and topics that are similar but somewhat different from the learning flow.

However, if the unmanned learning unit generation unit is configured, the above problems can be solved.

That is, the unmanned learner generation unit can distribute or simulate an unmanned learner such as a chatbot based on the previously generated storytelling data and analysis data.

Specifically, the unmanned learner experiment unit 310 extracts the storytelling instructional information to be actually distributed from the storytelling instructional information stored in the internal data processing repository 600, and reconfigures it into a chatbot form using machine learning. You will be simulating whether it works.

If it does not work as expected, return to the learning syllabus creation unit 100 to check the learning flow, or return to the learning flow analysis unit 200 may modify the analysis information.

After checking the normal operation is distributed to the unmanned learner through the unmanned learner distribution unit 320 is provided to the user (student).

For example, as shown in FIG. 10, when a question is asked, “When do you recruit members from the bowling club?” “When do you recruit members?” From April 30 to May 10 Recruit members by "If you provide the type of answer that the questioner expects, it will be analyzed as working.

The unmanned learner distribution unit 320 stores the storytelling instructional information reconstructed in the chatbot form in the service storage 700 when the test result of the unmanned learning experiment unit is properly operated, and stores the stored storytelling instructional information in service processing. It is output (distributed) to students by the engine 500.

At this time, the unmanned learner distribution unit 320 is output (distributed) by including a conversation flow that is not included in the selected learning flow (teaching plan) by subject classification, annotation data learning, exception processing, and Quibble generation for each context.

In other words, the unmanned learner experiment unit 310 simulates in the form of an unmanned learner such as a chatbot to be actually distributed.

For example, as shown in FIG. 11, when the answer is 'yes', 'is sweet food good for health?', If the answer is 'not good', it is analyzed as a correct answer. It will be distributed through wealth.

In addition, the learner can freely attempt a conversation that does not fit the flow of storytelling data.

At this time, it is possible to continue the natural conversation while maintaining the learning flow through the speech selector 510 of the service processing engine.

If it did not proceed as expected, return to the learning flow creation unit 100 may modify the metadata or storytelling data, the learning flow analysis unit 200 may modify the meaning or text of the storytelling unit data.

The above process can be repeated until the desired simulation result is obtained.

If the simulation proceeds as expected, the unmanned learner distribution unit 320 stores the unattended learner such as chatbot in the service storage 700, and will be distributed through the service processing engine.

When the unmanned learner is distributed, the currently analyzed storytelling data is classified according to the subject classification, machine learning is performed on the data, and exceptions and contextual quizzles are generated and stored in the service repository.

At this time, the generated exception processing and quibble are used in the service processing engine.

6 is a block diagram of a natural language processing engine 400 of a storytelling based multimedia unattended remote 1: 1 customized education system according to an embodiment of the present invention.

As shown in FIG. 6, the natural language processing engine 400 includes a natural language analyzer 410 and an analysis data provider 420.

Specifically, the natural language analysis unit 410 is a learning flow analysis unit 200, the unmanned learning unit generation unit 300, the service processing engine 500 when the request for analysis, morphological analysis, entity name analysis, intention analysis, At least one natural language analysis of expectation analysis and emotion analysis is performed.

For example, for the morphological analysis and the entity name analysis, for the sentence 'today I want to eat an apple', the morphological analysis in the form of 'today / silver / apple / ga / eat / go / want / da' is used. It assigns an individual name of 'fruit' to the 'apple' stem and gives it an object name of 'time'.

The intention analysis described above is to analyze the intention to play the music of the shower for the sentence 'play the shower.'

The analysis data providing unit 420 is to provide the analysis data of the natural language analysis through the natural language analysis unit to any one of the analysis request of the learning flow analysis unit 200, the unmanned learning unit generation unit 300, the service processing engine 500. .

For example, the learning flow analysis unit analyzes the unit storytelling data in the form of morphemes, objects, intentions, expectations, and emotions through the analysis data provider and returns the analysis information through the analysis data provider.

The same is used for the learner utterance simulation of the unmanned learning unit generation unit, and the same is used for the learner utterance in the actual service by the service processing engine 500.

7 is a block diagram of a service processing engine 500 of a storytelling-based multimedia unattended remote 1: 1 customized education system according to an exemplary embodiment of the present invention.

As shown in FIG. 7, the service processing engine 500 includes a speech selector 510 and a service processing unit 520.

In detail, the speech selector 510 determines whether the response information obtained from the student terminal 3000 is a free speech that deviates from the learning flow (training plan), and selects similar storytelling data within the same classification when the speech is free speech. It will be provided to student terminals.

For example, as shown in the development category of FIG. 8B, a speech that is expected in the question 'Mr. Do you like exercise?' Is the answer of at least the analysis data of the speech, and the key word in the speech is 'Yes / no', 'likes / dislikes'.

If the learner's speech meets the minimum requirements, find the most similar data in the classification for this question.

If the learner's utterance means the same as the answer to the data found (comparison of analytic data of morphemes, entities, intentions, expectations, etc.), continue learning.

If the user's utterance is a utterance that does not satisfy the minimum condition, the user's utterance responds by using the key word, object, and the like.

For example, if the expected answer to the question 'Do you like exercise?' In the current learning flow is 'Yes / No', etc., the user says 'I like baseball'. Quibble response to continue the conversation.

The service processor 520 obtains a learning data request signal from the student terminal 3000 and provides response information to the student terminal.

That is, when a student selects a specific storytelling lesson, the service processor acquires a learning data request signal for the selected storytelling lesson, and extracts the requested storytelling lesson from the internal data processing repository 600 as a response information. It will be provided to the terminal.

As a result, when configured as described above, the teacher (chatbot developer) includes the unmanned learner response and processing automatically generated from the teaching materials other than the current teaching material (learning flow data) by the unmanned learning machine generation unit 300. Focus on storytelling and various exceptional situations.

As a result, the teacher (chatbot developer) can check whether the written learning flow is valid in exceptional circumstances and modify or supplement it to determine whether to create the final unmanned learning machine.

As a result, the teacher (chatbot developer) offers the advantage of easily predicting and determining the outcome of the final unmanned learner simply by modifying or changing the draft (learning flow data) without considering the technical processing and control of the final unmanned learner. Will be done.

In addition, the present invention is to create a learning flow according to the education field, content, etc. as a learning flow creation unit, to analyze the storytelling data using the learning flow (teacher) analysis unit, using the unmanned learning unit generation, such as simulating and chatbot unmanned Distribute the learner.

The distributed unmanned learner is provided to the learners through the service processing engine, and the natural language processing engine is used for the analysis of the learning flow and the user's speech analysis, thereby optimizing the teaching plan.

In addition, although the preferred embodiment of the present invention has been shown and described above, the present invention is not limited to the above-described specific embodiment, the technical field to which the invention belongs without departing from the spirit of the invention claimed in the claims. Of course, various modifications can be made by those skilled in the art, and these modifications should not be individually understood from the technical spirit or the prospect of the present invention.

100: lesson plan writing department
200: learning flow analysis unit
300: unmanned learning machine generation unit
400: Natural Language Processing Engine
500: service processing engine
600: Storage for internal data processing
700: service storage
1000: Storytelling-based customized educational chatbot server
2000: Teacher's terminal
3000: Student terminal

Claims (7)

In storytelling based multimedia unmanned remote 1: 1 customized education system,
Using the template information provided from the storytelling-based customized education chatbot providing server 1000, implement a teaching plan view for the teacher terminal platform environment on the screen, and create a storytelling lesson for each teacher and situation through the implemented lesson plan view. Teacher terminal (2000) to do,
Provides template information for creating a storytelling lesson plan to the teacher terminal (2000), reconstructs the storytelling lesson text provided from the teacher terminal into the chatbot form using machine learning, and reconstructs the storytelling lesson reconstructed in the chatbot form into the student terminal (1000) and a storytelling-based customized educational chatbot providing server (1000) for providing feedback on the request of the student terminal through the configured service processing engine,
The storytelling-based customized educational chatbot providing server 1000 displays the chatbot-type storytelling syllabus, and includes a student terminal 3000 for providing request information to the storytelling-based customized educational chatbot providing server 1000. Configured by

The storytelling-based customized education chatbot providing server 1000,
Provides template information for creating a storytelling syllabus to the teacher terminal 2000, and using the storytelling syllabus provided from the teacher terminal 2000, dialogue and learning pictures, voices, and videos between students and teachers that may occur in a classroom situation. Learning lesson planner 100 for creating a storytelling lesson plan information including text information in a storytelling manner according to the learning flow;
A learning flow analysis unit for analyzing the storytelling instructional information created by the storytelling method according to each learning flow created by the learning syllabus creation unit 100 and storing it in the internal data processing repository 600 in a form that can be programmed. 200;
An unmanned learner generation unit 300 for reconfiguring the storytelling instructional information stored in the internal data processing storage 600 into a chatbot form using machine learning and simulating whether or not it operates properly;
When the learning flow analysis unit 200, the unmanned learning unit generation unit 300, the service processing engine 500, at least one natural language analysis of morphological analysis, entity name analysis, intention analysis is performed, the request of the student terminal A natural language processing engine 400 for performing natural language processing on the matters and providing the natural language processing to the service processing engine 500;
A service processing engine 500 for enhancing or servicing the storytelling syllabus information reconstructed in a chatbot form stored in the service storage to the student terminal 3000 and providing feedback on the request of the student terminal;
An internal data processing repository 600 for storing information of conversations and learning pictures, voices, videos, texts, and storytelling instructional information between students and teachers, which may occur in the class situation;
Storytelling-based multimedia unattended remote 1: 1 customized, characterized in that it comprises a; storage for service 700 for reconstructing the storytelling instructional information reconstructed in the chatbot form generated by the unmanned learning unit generation unit; Education system.
delete The method of claim 1,
The learning syllabus creation unit 100,
A basic template providing unit 110 for generating basic template information on a basic template of a learning flow (teaching plan) having a common flow for each education field and content for providing as a teacher terminal and providing the same as a teacher terminal;
A recommendation template providing unit 120 for generating recommendation template information for a recommendation template of a learning flow (teaching plan) having a common flow for each education field and content for providing as a teacher terminal and providing the same to a teacher terminal;
Using the storytelling syllabus provided from the teacher terminal 2000, the storytelling syllabus information including picture, voice, video, and text information for dialogue and learning between students and teachers, which can occur in the class situation, is prepared by storytelling according to the learning flow. Storytelling creation unit 130; the storytelling-based multimedia unmanned remote 1: 1 customized education system, characterized in that comprises a.
The method of claim 1,
Learning flow analysis unit 200,
Outputs a list of conversations constituting a learning flow (teaching plan), instructs the linguistic information analysis unit to perform analysis on each storytelling unit data, and a story for receiving analysis result information analyzed by the linguistic information analysis unit A telling navigation unit 210;
A language information analyzer 220 for performing analysis on each storytelling unit data according to the instruction of the storytelling navigation unit 210 and providing analysis result information to the storytelling navigation unit 210;
Storytelling-based multimedia unattended remote 1: 1 customized education system, characterized in that it comprises a; providing an annotation data, an annotation system selection unit 230 for performing a modification to the selected annotation data.
The method of claim 1,
Unmanned learning machine generation unit 300,
Unmanned learning experiment unit (310) for extracting the storytelling instructional information that is actually distributed among the storytelling instructional information stored in the internal data processing storage 600, reconfiguring the chatbot form using machine learning, and simulating whether it works properly. );
Experimental results of the unmanned learning unit experimental unit, unattended learner distribution unit 320 for storing the storytelling syllabus information reconstructed in the form of chatbots in the service storage 700, if it operates properly; Storytelling based multimedia unattended remote 1: 1 customized education system.
The method of claim 1,
The natural language processing engine 400,
Natural language analysis unit for performing any one of the natural language analysis of the morphological analysis, entity name analysis, intention analysis when the analysis request from the learning flow analysis unit 200, unmanned learning machine generation unit 300, service processing engine (500) ( 410);
An analysis data provider 420 for providing the analysis data analyzed in the natural language through the natural language analysis unit as one requested by the analysis of the learning flow analysis unit 200, the unmanned learning unit generation unit 300, and the service processing engine 500; Storytelling based multimedia unattended remote 1: 1 customized education system, characterized in that the configuration, including.
The method of claim 1,
The service processing engine 500,
A speech selector for determining whether or not the response information obtained from the student terminal 3000 is a free speech that deviated from the learning flow (training plan) and selecting similar storytelling data within the same classification and providing the same to the student terminal in the case of free speech. 510);
Storytelling-based multimedia unattended remote 1: 1 personalized, characterized in that comprises ;; a service processor 520 for obtaining a learning data request signal from the student terminal 3000, and providing the response information to the student terminal; Education system.

Images