KR102188553B1 - The System For Providing Korean Language Education System with Animation - Google Patents

Abstract

According to the present invention, a Korean language education system through animation-based storytelling includes: a video DB which receives and stores educational animations from at least one animation server; a text DB for receiving and storing text data for the educational animation; a learning material generation module for generating learning material including the educational animation and the text material; and a learning material providing module which provides the above learning materials to learners. According to the present invention, the Korean language education system through animation-based storytelling allows the learner to study Hangul more easily and interestingly through a storytelling method based on Korean language appearing in animation.

Description

The System For Providing Korean Language Education System with Animation}

The present invention relates to a Korean language education system through animation-based storytelling, and provides a Korean language education system through animation-based storytelling that enables learners to study Korean language more easily and more funly through Korean words appearing in animation.

According to the 2008 United Nations Development Program (UNDP), Korea's illiteracy rate is less than 1%. Hangul is usually learned from the age of 5 to 7 years old, but since it is the knowledge necessary for living and is usually the first learning to be performed, children who will continue to study as they grow up should be able to develop positive perceptions of learning. Therefore, the education of Hangeul is very important in human life, and as important as it is, a Hangul education system without reluctance is needed.

Moreover, because Hangeul has a very scientific structure, it is said that anyone learns the ability to read and write Hangul in a short time. This is limited to those with intelligence and learning ability of about 10 years or older, and usually speaks Hangul from the first grade of kindergarten or elementary school. It is common for learning children to devote more than two years to learning Korean.

There may be personal differences, but "ㄱ, ㄴ, ㄷ,...." Consonants such as "ㅏ, ㅓ, ㅜ,…." The vowel repetition memorization of the method, the whole letter education method such as "A, I, Da,...", and the memorization education through picture and letter matching (MATCHING) are difficult targets for children and foreigners to learn quickly. Various Hangul learning methods and learning devices are constantly being devised.

As a prior art for such a method of teaching Korean, Korean Patent Laid-Open No. 10-2016-0109377 discloses a method of teaching Korean and a learning device using the same. The above invention relates to a method and tools for teaching and learning Korean, and more particularly, to allow learners to easily understand and learn the meanings and appropriate methods of using various Korean words generated by directly combining consonants and vowels. It is designed so that not only Korean infants but also foreigners can easily learn the Hangul spelling method, and for this purpose, the contents of the basic alphabet defined in the current Hangul spelling have been re-established from a modern linguistic perspective and made into a document.

The present invention goes further from here, and attempts to disclose a Korean language education system through animation-based storytelling that enables learners to study Hangul more easily and more interestingly through a storytelling method based on Korean words appearing in animation.

The present invention was conceived to overcome the problems of the above technology, and a Korean language education system through animation-based storytelling that enables learners to study Hangul more easily and more interestingly through a storytelling method based on Korean words appearing in animation. Its main purpose is to provide.

Furthermore, it aims to enhance the learning effect by making the text included in the educational animation as a keyword so that it can be emphasized on the learning material.

In addition, it aims to further enhance the educational effect by automatically extracting text data from educational animations and providing them as learning materials.

In addition, it aims to increase the convenience and objectivity of data generation by automating the generation of representative keywords, which are the core contents of educational animation.

In addition, it is to improve the accuracy of analysis by presenting a new text analysis method in text data analysis.

In order to achieve the above object, the Korean language education system through animation-based storytelling of the present invention includes: an image DB for receiving and storing at least one educational animation; A text DB for receiving and storing text data for the educational animation; A learning material generation module for generating learning material including the educational animation and the text material; And a learning material providing module for providing the learning material to the learner.

In addition, the system further includes a text extraction module for extracting text information from the educational animation, wherein the text DB stores the text information extracted for the educational animation as the text data. .

In addition, the system further comprises a learning word generation module for generating a representative learning word from the text data, wherein the learning material generation module generates a learning image by synthesizing the representative learning word with the educational animation. It characterized in that it further comprises a learning image generation unit, and a learning data generation unit for generating learning materials including the learning image and the text data.

Furthermore, the system further includes a text analysis module that analyzes and processes the text data to generate a plurality of representative keywords, wherein the learning word generation module designates at least one of the plurality of representative keywords as a representative learning word. Characterized in that.

The Korean language education system through animation-based storytelling according to the present invention,

1) Providing a Korean language education system through animation-based storytelling that allows learners to learn Korean more easily and more fun through a storytelling method based on Korean words appearing in animation,

2) The text included in the educational animation is keywordized so that it can be emphasized on the learning materials, thereby enhancing the learning effect.

3) The educational effect was further improved by automatically extracting text data from educational animations and providing them as learning materials.

4) The convenience and objectivity of data generation was improved by automating the creation of representative keywords, which can be called the core contents of educational animation.

5) In generating representative keywords through text data analysis, there is an effect of increasing the accuracy of analysis by presenting a new text analysis method.

1 is a conceptual diagram showing a schematic configuration of a Korean language education system through animation-based storytelling of the present invention.
2 is a block diagram showing the configuration of a Korean language education system through animation-based storytelling of the present invention.
3 is a process diagram of a representative keyword calculation process of the Korean language education system through animation-based storytelling of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The accompanying drawings are not drawn to scale, and the same reference numerals in each drawing refer to the same elements.

1 is a conceptual diagram showing a schematic configuration of a Korean language education system 10 through animation-based storytelling of the present invention.

First, in the present invention, the learner (2) is a person who receives language, especially, Korean language education through the system 10 of the present invention. At this time, the age group of the learner (2) is not limited, but in general, infants and children, which are the first time to learn Korean, can be said to be the general learner (2). However, it is self-evident that all learners (2) can become learners regardless of age and gender when learning about Hangul.

The main server 1 of the present invention is a series of subjects (central control server) for implementing the system 10 of the present invention, and includes a server PC and a network communication network. In addition, the main server 1 has a central processing unit (CPU) and a program that can be executed in the central processing unit on a hardware basis having storage means such as memory and a hard disk, that is, software is installed to execute this software. A series of specific configurations of the software will be described later as structural units such as'module','part', and'part'.

This main server (1) performs a function to provide the learner (2) with a system (10) that allows the learner (2) to study Hangul more easily and more interestingly through the Korean words appearing in the animation of the present invention. Is done.

The configuration of such a'module' or'part' or'interface' or'part' is installed and stored in the storage means of the main server 1, software running through CPU and memory, or hardware such as FPGA or ASIC Means the work composition of. In this case, the configurations of'module','unit', and'interface' are not limited to hardware, and may be configured to be in an addressable storage medium or configured to reproduce one or more processors. As an example,'module' or'sub' or'interface' refers to components such as software components, object-oriented software components, class components and task components, processes, functions, and properties. S, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables.

The functions provided by these'module' or'sub' or'interface' may be combined into a smaller number of components and'sub' or'module', or additional components and'sub' or'module' Can be further separated.

In addition, the main server 1 refers to all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in a corresponding hardware device according to an embodiment. For example, a computing device as an example of a server may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and an application running on each device, but is not limited thereto.

The detailed configuration of the Korean language education system 10 through such animation-based storytelling will be described with the drawings as follows.

2 is a block diagram showing the configuration of the Korean language education system 10 through animation-based storytelling of the present invention, and FIG. 3 is a process diagram of representative keywords calculation of the Korean language education system through animation-based storytelling of the present invention. .

Referring to FIG. 3 along with FIG. 2, the Korean language education system 10 through animation-based storytelling of the present invention basically includes an image DB 100, a text DB 200, and a learning material generation module 300. , The learning data providing module 400 may be included, and further, as an additional configuration, a text extraction module 500, a learning word generation module 600, and a text analysis module 700 may be further included.

The image DB 100 performs a function of receiving and storing at least one educational animation. In this case, the educational animation refers to an animation for Korean language education in which storytelling is focused based on animation. At this time, there are no restrictions on the method of producing the educational animation, the type of video, the length, etc., but most preferably, as an example, the heroes each having a mark for awakening on their body are By combining consonants and vowels to form words, it is possible to understand the principles of Hangul by awakening the combination of letters, which can be called characteristics of Hangeul, by correcting sweet words and profanity to brighten the world of darkness in the future. These educational animations may be stored in the image DB 100 by receiving pre-produced educational animations from a video production server that may be provided separately, or training by Korean education experts and video experts related to the system 10 of the present invention. It is also possible to produce an animation, and to receive it and store it in the image DB 100.

The text DB 200 performs a function of receiving and storing text data for the educational animation. At this time, the educational animation itself may contain some text as a subtitle of the video, but in order to further enhance the learning effect, the subtitle for the voice output on the educational animation video, or the consonant/which is the most important point in learning the educational animation. Supplementary explanations for vowels or words to be learned in the corresponding educational animation, or word cards for the corresponding words, considering that the learner (2) is generally an infant, may be text materials. Accordingly, all textual data related to the contents of the educational animation can be text data, and the input and storage thereof. Such text data may be stored in the image DB 100 by receiving text data previously produced for the corresponding educational animation from a data production server that may be provided separately, or a Korean education expert related to the system 10 of the present invention directly It is also possible to produce text data and input and store it in the text DB 200.

The learning material generation module 300 performs a function of generating learning materials obtained by integrating the educational animation and the text data. This can be implemented by simply providing educational animation and text data through a single platform. Or, if the educational animation does not contain separate subtitles at all, the synthesized image by synthesizing the text data into the educational animation. It is also possible as a method of creating learning materials that can be called animations. In other words, if the educational animation and text materials can be provided to the learner (2) together, there is no separate limitation on the method of generating learning materials.

The learning material providing module 400 performs a function of providing the generated learning material to the learner 2. At this time, as a method of providing learning materials, various methods such as providing through an Internet site, providing through an application, or transmitting and processing the learning material itself to the learner 2 can be implemented without limitation. However, in recent years, it is essential to learn video materials using tablet PCs or smartphones for early childhood learning. Most preferably, learners (2) can easily and conveniently receive learning materials anytime, anywhere through an application or website. It would be desirable to do it.

In addition, the system 10 of the present invention includes an additional function for automatically extracting, generating, and storing text data related to educational animation even if the administrator of the system 10 or a Korean language education expert does not directly generate the text data. I can. Therefore, for this, the system 10 of the present invention may further include a text extraction module 500 for extracting text information from the educational animation.

The text extraction module 500 has a function of extracting the text information included in the video included in the educational animation or the image for each frame of the corresponding video, or extracting text information from the voice data included in the educational animation through STT technology. Perform. This may refer to a conventional text extraction technique or text conversion of speech, so a detailed description thereof will be omitted.

Therefore, in the case where text information is automatically extracted and generated from the video/image/audio data included in the educational animation from the educational animation through the text extraction module 500, the text DB 200 converts the extracted text information into text. It will be saved as data. Therefore, it is possible to conveniently create and store text information without the need for additional text information input.

In addition, as linked to the learning material generation module 300, representative learning words, which can be called learning words that are the core of the educational animation, are generated from text materials related to the educational animation, and are automatically used as subtitles in the educational animation. Synthesis processing can increase the learning effect of educational animation.

First, for this, the system 10 may further include a learning word generation module 600 that designates a representative learning word from the text data, which is a representative learning of the most frequently exposed or repeated learning words in the text data. By designating it as a word, it is possible to prepare it for synthesis in an educational animation.

Furthermore, when the representative learning words are designated and generated in the learning word generation module 600 as described above, the educational animation is generated through the learning image generation unit 410, which is a detailed configuration that can be additionally included in the learning material generation module 300. By synthesizing the representative learning words in the form of subtitles, a learning image in which the representative learning words are synthesized in an educational animation can be generated, and furthermore, a learning image generated through the learning material generation unit 420, which is another detailed configuration, and It is possible to generate learning materials by integrating the text data.

In this case, since the representative learning words themselves are synthesized in the form of subtitles in the educational animation, it is possible to grasp the representative learning words to be representatively learned in the educational animation at a glance, and additional text materials are provided along with them. Detailed learning is also possible.

Further, the system 10 may further include a text analysis module 700 that analyzes and processes the text data to generate a plurality of representative keywords, wherein the learning word generation module 600 is At least one of a plurality of representative keywords may be designated as the representative learning word. In this case, the representative keywords are automatically analyzed and processed from the text data extracted from the educational animation, and the representative learning words can be designated through the selection of a language education expert or administrator of the system 10 or automatic selection on the system 10. There is.

At this time, the text analysis module 700 should perform a function of analyzing text data and extracting representative keywords, which may be the core of the content, and in this case, various text analysis methods may be used. In the case of extracting representative keywords within such text, uncertainty is inherent, so an efficient method should be used when extracting keywords. Accordingly, the text analysis module 700 of the present invention analyzes and processes the text data based on fuzzy theory, or in other words, fuzzy inference in this text data analysis to generate a representative keyword. At this time, in the case of fuzzy theory, a rule that uses approximate, subjective value and data that may be ambiguous and incomplete is created and data is analyzed through it, so that keywords can be inferred in a manner similar to the way humans think, so representative of the present invention It can be effectively applied in keyword generation.

At this time, the text analysis module 700 includes a keyword extraction unit 710, an appearance frequency determination unit 720, an importance calculation unit 730, a representative keyword designation unit 740, a similarity calculation unit 750, and a correction value calculation unit ( 760), it may be configured to include an additional keyword generation unit 770.

The keyword extracting unit 710 performs a function of extracting a plurality of basic keywords from text data related to the educational animation. In this case, the basic keyword generally means a noun included in the text data, but when learning a language Not only nouns, but also adjectives and verbs are essential to familiarize yourself with, so do not limit the parts of speech of basic keywords to nouns, but recognize nouns, adjectives, verbs, etc.

The frequency of appearance determination unit 720 performs a function of identifying the frequency of appearance of each of the basic keywords. In this case, the frequency of appearance refers to the number of times that the corresponding basic keyword appears in the corresponding text data if simplified. That is, if the basic keyword'apple' appears 3 times on text data, the frequency is 3. This basic concept of frequency is judged as the frequency of appearance. However, further subdividing and grasping the frequency to perform fuzzy inference based on the frequency. To this end, the appearance frequency determination unit 720 may be composed of an appearance frequency calculation part 721 and an overall frequency calculation part 722.

The appearance frequency calculation part 721 performs a function of identifying the appearance frequency of each of the plurality of basic keywords, and the above-described appearance frequency means the number of times the corresponding basic keyword appears on the text data.

The total frequency calculation part 722 is to grasp the total frequency, which is the frequency in which each of the basic keywords appears in the full text data stored in the text DB 200. Here, the total frequency is again explained. It means the number of times a corresponding basic keyword appears in the entire text data stored in the text DB 200 and processed.

That is, when a total of 3 text data of text No. 1 / text No. 2 / text data No. 3 are accumulated in the text DB 200 of the present invention, the basic keyword'apple' is No. 1 in text data No. Suppose that the basic keyword'apple' appears 3 times in text material 2 and the basic keyword'apple' appears 2 times in text material 3. At this time, the frequency of appearance for the basic keyword'apple' in the text data 2 is 3 and the total frequency is 6.

The importance calculation unit 730 performs fuzzy inference for a plurality of the basic keywords based on the number of appearances identified by the above-described frequency of appearance determination unit 720 and the total number of frequencies, thereby The function of calculating the numerical importance is performed. Furthermore, when calculating the importance, not only the appearance frequency but also the above-described total frequency can be used together. For this detailed configuration, the importance calculation unit 730 of the present invention includes an appearance normal calculation part 723, a full normal calculation part 724, an inverse normal calculation part 725, an input value calculation part 726, and an importance calculation part. It can be configured to include (727).

The appearance normal calculation part 723 performs a function of performing normalization for the application of fuzzy inference rules in the appearance frequency, which is the most basic concept among the concepts of frequency. In the fuzzy inference rule, since a normalized value is required to generate an input value for fuzzy control, it can be referred to as a process of performing it. The appearance normalization frequency calculation through the appearance normalization calculation part 723 may be performed through Equation 1 below.

Equation 1,

는 기본키워드

에 대한 등장 정규화 빈도,

는 기본키워드

의 등장 빈도수,

는 기본키워드

의 전체 빈도수를 의미한다.here,

Is the default keyword

The appearance normalization frequency for,

Is the default keyword

The frequency of appearance of,

Is the default keyword

Means the total frequency of

The total normal calculation part 724 performs normalization on the total frequency, which is another concept among the concepts of frequency, to prepare for generating an input value for fuzzy calculation. In this case, the total normalization frequency may be calculated through the entire normalization calculation part 724 through Equation 2 below.

Equation 2,

는 기본키워드

에 대한 전체 정규화 빈도,

는 기본키워드

의 전체 빈도수,

은 상기 텍스트DB(200)에 저장된 텍스트자료의 총 개수를 의미한다.here,

Is the default keyword

The overall normalization frequency for,

Is the default keyword

The total frequency of,

Denotes the total number of text data stored in the text DB 200.

The inverse normalization calculation part 725 performs a function of calculating the inverse normalization frequency by performing normalization on the reciprocal of the total frequency, that is, the inverse normalization calculation part 725 ) To perform a preparatory work for fuzzy calculation, which will be described later. In this case, the calculation of the inverse normalization frequency through the inverse normalization calculation part 725 may be performed through Equation 2 below.

Equation 3,

는 기본키워드

에 대한 역 정규화 빈도,

는 기본키워드

의 전체 빈도수를 의미한다.here,

Is the default keyword

The inverse normalization frequency for,

Is the default keyword

Means the total frequency of

The emergence normal calculation part 723, the total normal calculation part 724, and the inverse normal calculation part 725 as described above perform a function of generating input variables for fuzzy inference by performing normalization for fuzzy calculation in different ways. do. At this time, instead of normalizing all variables in the same way, different frequencies or the number of texts that may be related to each variable are reflected in normalization, so that a more standardized value can be obtained.

The input value calculation part 726 performs a function of calculating an appearance normal input value, a total normal input value, and an inverse normal input value by fuzzifying the appearance normalization frequency, the total normalization frequency, and the inverse normalization frequency. The appearance normalization frequency, the total normalization frequency, and the inverse normalization frequency are fuzzy in a form suitable for fuzzy inference, and input values are calculated through them. In this case, there is no particular limitation on the method of fuzzing, but it is preferable to perform fuzzing using a triangular or quadrangular purge number, and to calculate an input value therethrough. This may refer to the conventional fuzzy inference, so a detailed description will be omitted.

The importance calculation part 727 applies a fuzzy inference rule to each of the appearance regular input value, the full normal input value, and the inverse normal input value, and performs a function of calculating the numerical importance for each of the basic keywords by de-fuzzying. do. At this time, the method of applying the fuzzy inference rule is not limited, but preferably, the appearing regular input value described above in the membership function of their previous units for each fuzzy rule generated according to the number of fuzzy determined in the input value calculation part 726 , After applying the entire normal input value and the inverse normal input value, the minimum values are taken. After that, the obtained values are classified into groups, and a fuzzy value is generated by taking the maximum value of the fuzzy value for each group. Thereafter, various conventional methods may be used for defuzzification, but most preferably, the importance of the corresponding basic term is calculated by applying defuzzification using the center of gravity method to the generated plurality of fuzzy values.

The representative keyword designation unit 740 of the text analysis module 700 performs a function of designating at least one representative keyword from among the plurality of basic keywords according to the high and low level of importance calculated through the importance calculation unit 730 described above. do. That is, a basic keyword of high importance is designated as the representative keyword, and in this case, the representative keyword may be one or more.

Further, it is possible to further designate not only the first designated representative keyword, but also the representative keyword. In this case, when the representative keyword is additionally designated, some of the basic keywords having high similarity to the first designated representative keyword may be additionally designated as the representative keyword. To this end, the text analysis module 700 may further include a similarity calculation unit 750, a correction value calculation unit 760, and an additional keyword generation unit 770.

The similarity calculation unit 750 performs a function of calculating a numerical appearance frequency similarity between the designated representative keyword and the basic keyword. At this time, the reason for calculating the similarity of the appearance frequency is that basically, text data contains contents related to Hangeul education as it is data for Hangeul education. Therefore, each text material contains general, detailed, or partial content on the subject of Korean language education. Therefore, if the initial designated representative keyword is judged as the core content of the text data, there is a high probability that other basic keywords related to this will also appear in the text data, and the similarity in this frequency distribution can be used for calculating the representative keywords. There will be. Therefore, the similarity of the frequency distribution is calculated as an appearance frequency similarity, and the appearance frequency similarity is calculated based on Equation 4 below.

Equation 4,

는

번째 대표키워드와

번째 기본키워드 사이에서 산출된 등장빈도 유사도,

는 조정 상수,

는 상기 대표키워드지정부(740)에서 지정된 대표키워드의 수,

는

번째 대표키워드의 등장 빈도수,

는

번째 기본키워드의 등장 빈도수를 의미한다.here,

Is

1st representative keyword and

The similarity of the appearance frequency calculated between the first basic keywords,

Is the adjustment constant,

Is the number of representative keywords designated by the representative keyword designation unit 740,

Is

Frequency of appearance of the first representative keyword,

Is

It means the frequency of appearance of the first basic keyword.

Therefore, through Equation 4, the degree of association between the representative keyword and the basic keyword that is not divided into the representative keyword is calculated as a number, and the closer this value is to 1, the higher the similarity. That is, it means that the degree of relevance is high, and the closer to 0, the lower the degree of similarity, that is, the degree of relevance is also low. In the case of the above-described variable, the adjustment constant may be set by the administrator of the system 10, and the size of the number is not limited, but an integer value between 5 and 20 may be preferably set as the adjustment constant.

The correction value calculation unit 760 assigns a weight to each of the basic terms based on the calculated appearance frequency similarity and calculates a correction value by reflecting the weight. At this time, preferably, the higher the appearance frequency similarity, the higher the weight is calculated, and in the easiest way, the above-described inverse document frequency, that is, a value obtained by multiplying the reciprocal of the total frequency by the appearance frequency similarity, can be calculated as a weight. The correction value can be calculated by multiplying the weight by the above-described importance.

As described above, the additional keyword generator 770 performs a function of additionally designating a representative keyword among the basic keywords based on the height of the correction value in which the weight is reflected in the importance level.

Therefore, the representative keyword is first designated and additionally designated through the representative keyword designation unit 740 and the additional keyword generation unit 770 as described above, so that the representative keywords representing the corresponding text data can be representative learning words. Words can be provided to learners 2 as learning images.

As described so far, the configuration and operation of the Hangul education system through animation-based storytelling according to the present invention is expressed in the above description and drawings, but this is only an example, and the spirit of the present invention is described in the above description and drawings. It is not limited, of course, various changes and changes are possible within the scope of the technical spirit of the present invention.

1: main server 2: learner
10: System 100: Video DB
200: Text DB 300: Learning material generation module
400: learning material providing module 410: learning image generation unit
420: learning material generation unit 500: text extraction module
600: learning word generation module 700: text analysis module
710: keyword extraction unit 720: appearance frequency identification unit
721: appearance frequency calculation part 722: total frequency calculation part
723: Regular appearance calculation part 724: Overall regular calculation part
725: reverse normal calculation part 726: input value calculation part
727: importance calculation part 730: importance calculation part
740: Representative keyword designation government 750: Similarity calculation department
760: correction value calculation unit 770: additional keyword generation unit

Claims (12)

As a Korean language education system through animation-based storytelling,
An image DB for receiving and storing at least one educational animation;
A text DB for receiving and storing text data for the educational animation;
A text analysis module for generating a plurality of representative keywords by analyzing and processing the text data based on the fuzzy theory;
A learning word generation module for designating at least one of the plurality of representative keywords as a representative learning word;
A learning material generation module including a learning image generation unit that generates a learning image by synthesizing the representative learning words with the educational animation, and a learning material generation unit that generates a learning material including the learning image and the text data;
Including; a learning material providing module for providing the learning material to the learner,
The text analysis module,
A keyword extraction unit that extracts a plurality of basic keywords from the text data, an appearance frequency determination unit that determines the appearance frequency of each of the basic keywords, and fuzzy inferences for a plurality of basic keywords based on the appearance frequency, respectively An importance calculation unit that calculates numerical importance for each of the basic keywords by performing, and a representative keyword designation unit that designates at least one or more representative keywords from among a plurality of the basic keywords according to the level of the importance, and the designated representative A similarity calculation unit that calculates a numerical appearance frequency similarity between a keyword and the basic keyword, a correction value calculation unit that calculates a correction value for each of the basic keywords based on the appearance frequency similarity, and the correction value It includes an additional keyword generation unit for additionally designating the representative keyword among the basic keywords,
The appearance frequency determination unit,
An appearance frequency calculation part for identifying the frequency of appearance of each of the plurality of basic keywords, and an overall frequency calculation part for determining the total frequency, which is the frequency of each of the basic keywords appearing in the entire text data stored in the text DB,
The importance calculation unit,
An appearance normal calculation part for calculating the appearance normalization frequency by performing normalization on the number of appearance frequencies, an overall normal calculation part for calculating the overall normalization frequency by performing normalization on the entire frequency number, and a normalization for the reciprocal of the total frequency number An inverse normalization calculation part that calculates the inverse normalization frequency by performing a function, and an input that calculates the appearance normal input value, the total normal input value, and the inverse normal input value by fuzzifying the appearance normalization frequency, the total normalization frequency, and the inverse normalization frequency. Includes a value calculation part and an importance calculation part that calculates numerical importance for each of the basic keywords by applying a fuzzy inference rule to each of the appearance regular input value, full normal input value, and inverse normal input value and defuzzying and,
The similarity calculation unit,
A Korean language education system through animation-based storytelling, characterized in that the similarity of appearance frequency is calculated based on the following equation.
Equation,

(here,

Is

1st representative keyword and

Similarity of appearance frequency calculated between the first basic keywords,

Is the adjustment constant,

Is the number of representative keywords designated by the representative keyword designation government,

Is

Frequency of appearance of the first representative keyword,

Is

Frequency of appearance of the first basic keyword) The method of claim 1,
The system,
Further comprising a; text extraction module for extracting text information from the educational animation,
The text DB,
A Korean language education system through animation-based storytelling, characterized in that storing the text information extracted for the educational animation as the text data. The method of claim 1,
The appearance normalization frequency is,
A Korean language education system through animation-based storytelling, characterized in that it is calculated through Equation 1 below.
Equation 1,

(here,

Is the default keyword

The appearance normalization frequency for,

Is the default keyword

The frequency of appearance of,

Is the default keyword

Total frequency of) The method of claim 1,
The total normalization frequency is,
A Korean language education system through animation-based storytelling, characterized in that it is calculated through Equation 2 below.
Equation 2,

(here,

Is the default keyword

The overall normalization frequency for,

Is the default keyword

The total frequency of,

Is the total number of text data stored in the text DB) The method of claim 1,
The inverse normalization frequency is,
A Korean language education system through animation-based storytelling, characterized in that it is calculated through Equation 3 below.
Equation 3,

(here,

Is the default keyword

The inverse normalization frequency for,

Is the default keyword

Total frequency of) delete delete delete delete delete delete delete

Images