KR20200109515A - Education contents generating method using big data - Google Patents

Abstract

The present invention provides an educational content generation method using big data, which uses a search keyword input by a user to perform a search in text content big data constructed as big data and provides educational content from a search result. According to one embodiment of the present invention, the educational content generation method using big data, which is executed by an educational content apparatus, comprises the following steps of: acquiring a search keyword input by a user; determining a major topic from topics tagged with each search keyword input by the user; extracting a first title keyword about the search keyword input by a first user from a title of the search keyword input by the first user related to a first major topic among the determined major topics; extracting a second title keyword for educational control from titles of text content big data constructed as big data among the determined major topics; calculating a topic similarity between the first major keyword and the second major keyword based on a similarity between the first topic keyword and the second topic keyword; and providing a search result about the search keyword input by the user as educational content based on the calculated topic similarity.

Description

Education contents generating method using big data}

The present invention relates to a method and apparatus for generating educational content using big data. In more detail, it relates to a method for generating educational content using big data that provides visualized educational content for a search keyword input by the user based on a topic keyword extracted from an educational content DB using big data. .

Current online education is not being used more actively. The fundamental cause for this is as follows. First, most of the online education-related tools provide results in a list format, so user convenience is poor. Specifically, an educator who is provided with the results in the form of a list must check the contents listed in the list in order to find the core contents related to the education. Since this is a time-consuming task, it is a factor that greatly reduces user convenience.

On the other hand, some online education-related tools perform clustering on searched contents to improve user convenience, and provide education contents as a result. For example, some education related tools, as shown in Figure 1

Likewise, the linked data indicated by the searched content is provided through a visualized network. However, the visualized network

Most of the education-related tools that provide educational content as educational content have a problem that the accuracy of educational content is poor because they grasp the similarity between educational topics based on the co-occurrent frequency of the topic keyword. .

Specifically, in the case of grasping the relationship between educational topics using the frequency of simultaneous occurrence, search terms included in the majority of contents and/or general-purpose words commonly used in the relevant field may be mistaken as core keywords. In addition, the visualized network provided based on this has a high probability of having a hub & spoke structure in which the search word and/or general language acts as a strong hub as shown in FIG. Value is low and quality educational content cannot be provided to educators.

KR 1020100068532 A KR 1020140050217 A KR 1020170034206 A

The technical problem to be solved by the present invention is to provide a driving method for a device that provides visualized educational content for a search keyword input by a user.

Another technical problem to be solved by the present invention is to provide a method capable of accurately calculating the similarity or relationship of a subject indicated by a search keyword input by a user.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

In order to solve the above technical problem, a method for generating educational content using big data according to an embodiment of the present invention includes a search input by a user in the method for generating educational content using big data performed by an educational content device. Acquiring a keyword, determining a main subject from among subject words tagged to each of the search keywords input by the user, from the title of a search keyword input by a first user associated with a first main subject among the determined main subject words , Extracting a first title keyword for the search keyword input by the first user, from the title of the textbook content big data constructed as big data among the determined main keywords, a second title keyword for the textbook content Extracting, based on the similarity between the first subject keyword and the second subject keyword, calculating a subject word similarity between the first subject word and the second subject word, and based on the calculated subject word similarity , Providing a search result for the search keyword input by the user.

In one embodiment, the determining of the main subject word includes calculating a frequency for each time period for each of the tagged subject words, for each of a plurality of time periods, based on the calculated frequency for each time period, the Including the step of excluding a general-purpose main word from among the tagged subject words and determining the subject of interest from among the remaining subject words from which the general-purpose main word is excluded, wherein the general-purpose main word is, over a plurality of time intervals, for each of the calculated time intervals It may be a key word whose frequency continuously exceeds a threshold value.

In one embodiment, the determining of the main subject word includes calculating a frequency for each time period for each of the tagged subject words for each of a plurality of time periods, and among the tagged subject words, in the plurality of time periods. It may include the step of determining the main subject word in which the calculated frequency of each time section is continuously increasing over time.

In one embodiment, the determining of the main subject word includes calculating a frequency for each time period for each of the tagged subject words for each of a plurality of time periods, and among the tagged subject words, in the plurality of time periods. It may include the step of determining the main subject word in which the calculated frequency of each time section is continuously decreasing.

In an embodiment, the calculating of the similarity of the subject word between the first subject word and the second subject word comprises generating a first keyword vector for the first subject word based on the frequency of the first subject keyword. Generating a second keyword vector for the second main keyword based on the frequency of the second subject keyword, and using a similarity between the first keyword vector and the second keyword vector, the main word It may include the step of calculating the similarity.

In one embodiment, the providing of the search result for the search keyword input by the user includes providing a relationship between the determined main keywords as a visualized network, wherein the first A node corresponds to the first main subject word, a second node constituting the visualized network corresponds to the second main subject word, and the weight of the trunk line between the first node and the second node is the first main subject word. It may be determined based on the similarity of the subject word between the subject word and the second subject word.

An educational content device according to another embodiment of the present invention for solving the above-described technical problem includes at least one processor, a network interface, a memory for loading a computer program executed by the processor, and input by a plurality of users. A storage for storing the searched keyword and the computer program, wherein the computer program includes an operation of determining a main subject from among subject words tagged to each of the search keywords input by the plurality of users, and a first among the determined main subject words Operation of extracting the first title keyword for the search keyword input by the first user from the title of the search keyword input by the first user related to the main subject word, textbook contents constructed with big data among the determined main subject words From the title of big data, an operation of extracting a second title keyword for the textbook content, based on the similarity between the first title keyword and the second title keyword, between the first and second main keywords An operation of calculating the similarity of the subject word of and an operation of providing a search result for the search keyword input by the user based on the calculated similarity of the subject word.

An educational content computer program according to another embodiment of the present invention for solving the above-described technical problem is combined with a computing device to obtain a search keyword input by a user, each of the search keywords input by the user Determining a main subject word from among the subject words tagged in, extracting a first subject keyword for the search keyword input by the first user from the title of the first search target content related to the first main subject among the determined main subject words The step of, extracting a second title keyword for the teaching material content from the title of the textbook content big data constructed with big data among the determined main keywords, and determining a similarity between the first title keyword and the second title keyword. As a basis, calculating a similarity of a subject word between the first subject word of interest and the second subject word, and providing a search result for a search keyword input by the user based on the calculated subject word similarity. It can be stored on a recording medium in order to do so.

According to the present invention, a network is provided in which a relationship between educational topics is visualized as educational content for a search keyword input by a given user. Through this, the user can intuitively check the education trend information, thereby improving information delivery of education content.

In addition, among the subject words tagged to the search keyword input by the user, a general-purpose subject word is excluded, and a subject word whose frequency is continuously increased may be determined as the main subject word. Accordingly, since the subject word, which is increasing the interest of the educator, is reflected in the visualized network, there is an effect that it is provided as high-quality educational content reflecting educational trend information in the field. In addition, since the general-purpose main word is prevented from acting as a strong hub node in the visualized network, there is an effect of providing accurate educational content.

Also, the similarity of the main subject words may be calculated based on the similarity of the subject keywords. Accordingly, major keywords with high similarity are concentrated, and the effect of a strong hub node is suppressed, thereby providing accurate and valuable educational content.

The effects of the present invention are not limited to the above-mentioned effects, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is an exemplary diagram of a network created based on a conventional co-occurrence frequency.
2 is a block diagram of an educational content system according to an embodiment of the present invention.
3 is a diagram for explaining exemplary keywords that can be referred to in some embodiments of the present invention.
4 is a block diagram showing an educational content device according to another embodiment of the present invention.
5 is a hardware configuration diagram of an educational content device according to another embodiment of the present invention.
6 is a flowchart of a method of generating educational content using big data according to another embodiment of the present invention.
FIG. 7 is a diagram for explaining the step S200 of determining a search target period shown in FIG. 6.
8 to 11 are diagrams for explaining the determining step (S400) of the main main words shown in FIG. 6.
12 to 15 are diagrams for explaining an operation S500 of calculating the similarity of the main word shown in FIG. 6.
16 to 18 are diagrams for explaining a search result providing step (S600) shown in FIG. 6.
19 and 20 are diagrams for explaining evaluation results of a network provided by a method for generating educational content using big data according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, and only these embodiments make the disclosure of the present invention complete, and are common knowledge in the technical field to which the present invention pertains. It is provided to completely inform the scope of the invention to those who have, and the invention is only defined by the scope of the claims. The same reference numerals refer to the same components throughout the specification.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used with meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically. The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings.

2 is a block diagram of an educational content system according to an embodiment of the present invention.

Referring to FIG. 2, the educational content system may be configured to include an educational content device 100, at least one big data DB 200a to 200n, and a user terminal 300. In particular, FIG. 2 shows an example of acquiring educational content from a plurality of big data DBs 200a to 200n. However, the configuration shown in FIG. 2 is only a preferred embodiment for achieving the object of the present invention, and of course, some components may be added or deleted as necessary. In addition, it should be noted that each of the components of the educational content system shown in FIG. 2 represents functional elements that are functionally divided, and at least one component may be implemented in a form in which they are integrated with each other in an actual physical environment. Hereinafter, each component of the content search system will be described.

In the educational content system, the educational content device 100 is a computing device that performs a search for a search keyword input by a user acquired from the big data DBs 200a to 200n and provides educational content.

Here, the computing device may be a notebook, a desktop, a laptop, a smart phone, etc., but is not limited thereto, and includes all kinds of devices equipped with computing means and communication means. can do. In addition, the search keyword input by the user may include textbooks, multimedia data, educational reports, and the like, and any type of content information may be used. That is, the scope of the present invention is not limited to the type of content.

In some embodiments, the search keyword input by the user may be content provided as a search result through a predetermined search tool.

According to an embodiment of the present invention, the educational content device 100 determines a main subject among the remaining subject words, excluding a general subject word from among subject words tagged to each of the acquired search keywords input by the user, and determines the similarity between the main subject words. Provides a visualized network as educational content. Here, the general-purpose keyword may refer to a daily term that appears in a search keyword input by a user with high frequency but has a low importance as an educational subject. According to the present embodiment, by removing the general-purpose main word before performing a search, it is possible to prevent the general-purpose main word from acting as a strong hub in a visualized network. Accordingly, more accurate educational content may be provided to the user. A more detailed description of the present embodiment will be described later with reference to the accompanying drawings in FIG. 7.

In an embodiment, when the search keyword input by the user is content in a language field provided as a search result through arbitrary educational information, the tagged subject word may be a language subject word as shown in FIG. 3. The language subject words have a hierarchical structure, which reflects the educational topics in the relevant field well, and consists of terms familiar to educators in the field. According to the present embodiment, by providing educational content based on keywords, there is an effect of increasing information delivery to educators in the relevant field.

In an embodiment, when the type of the search keyword input by the user is multimedia, the tagged subject word may be a title of multimedia data or a keyword defined by an author. For example, in the multimedia 410 as illustrated in FIG. 4, the tagged subject word may mean a keyword 413 separately included by the multimedia author in addition to the multimedia title 411.

According to an embodiment of the present invention, the educational content device 100 extracts a title keyword from a title of a search keyword input by a user associated with a main subject word. In addition, the educational content device 100 calculates the similarity of the main subject words based on the similarity between the subject keywords. That is, the similarity of the subject words is not calculated based on the frequency of simultaneous occurrence of the main subject words, but the similarity of the subject words between the main subject words is calculated based on the similarity of the subject keywords associated with the main subject words. This is to take advantage of the fact that the title of the content implies the core educational subject that the author wants to express and contains the latest keywords used in the relevant field. According to the present embodiment, similarity between main subject words may be more accurately calculated, and it may be prevented from acting as a strong hub node in a network in which universal subject words and search words are visualized. Accordingly, high quality and valuable educational content can be provided to the user. A detailed description of the present embodiment will also be described later with reference to the accompanying drawings in FIG. 11.

According to an embodiment of the present invention, the educational content device 100 may determine a subject of interest from a search keyword input by a user issued during a search target period, and provide the relationship between the determined main subject words in a visualized network form. have. The analysis target period may be set by a user or may be automatically set by the educational content device 100, which may vary according to embodiments. According to the present embodiment, information delivery may be improved by using a visualized network. In particular, when the search target period is set as the latest period, more valuable educational content can be provided because the latest educational trend is reflected in the visualized network. A more detailed description of the visualized network will be described later with reference to the accompanying drawings in FIG. 15.

In the educational content system, the big data DB 200a internally 200n is a data source that receives a search keyword input by a user. The big data DB (200a internal 200n) may be an externally located database, may mean a local storage of the educational content device 100, and may be implemented in any way as long as it can provide a search keyword entered by a user. It's okay.

In the educational content system, the user terminal 300 is a terminal used by a user to receive educational content for a search keyword input by the user.

Each component of the educational content system illustrated in FIG. 2 may communicate through a network. Here, the network is a local area network (LAN), a wide area network (WAN), and a mobile radio network.

communication network), Wibro (Wireless Broadband Internet), etc. can be implemented in all types of wired / wireless networks.

So far, an educational content system according to an embodiment of the present invention has been described with reference to FIGS. 2 to 3. Next, the configuration and operation of the educational content device 100 will be described with reference to FIGS. 4 and 5.

4 is a block diagram showing an educational content device 100 according to another embodiment of the present invention.

Referring to FIG. 4, the educational content device 100 includes a search target period determining unit 110, a main word extracting unit 130, a main main word determining unit 150, a similarity calculating unit 170, and a visualization unit 190. It can be configured to include. However, only components related to the embodiment of the present invention are shown in FIG. 4. Accordingly, those of ordinary skill in the art to which the present invention belongs may recognize that other general-purpose components may be further included in addition to the components illustrated in FIG. 4. In addition, it should be noted that each of the components of the educational content device illustrated in FIG. 4 represents functional elements that are functionally divided, and at least one component may be implemented in a form integrated with each other in an actual physical environment. Hereinafter, each component of the educational content device 100 will be described.

The search target period determining unit 110 determines a search target period based on the number of published contents for each time period. A detailed description of this will be described later with reference to FIG. 7.

The subject word extracting unit 130 extracts the subject word tagged to the search keyword input by the user issued during the search target period. The main subject word determination unit 150 removes the general-purpose subject word from the extracted subject words based on the frequency of each subject word, and determines the main subject word among the remaining subject words. A detailed description of this will be described later with reference to FIGS. 9 to 10C.

The similarity calculation unit 170 calculates the similarity between main subject words. Specifically, the similarity calculation unit 170 includes a first title keyword extracted from a title of a search keyword input by a first user associated with a first main keyword among the determined subject words of interest, and big data among the determined main keywords. Based on the similarity between the second title keyword extracted from the title of the textbook content big data constructed as, the similarity of the subject word between the first main subject and the second main subject is calculated. A detailed description of this will be described later with reference to FIGS. 12 to 15.

The visualization unit 190 visualizes the relationship between the main subject words as a network based on the calculated similarity of the subject words. A detailed description of this will be described later with reference to FIGS. 15 and 16.

Meanwhile, although not shown in FIG. 4, the educational content device 100 may be configured to further include an interface unit (not shown). The interface unit (not shown) provides a visualized network as educational content. In addition, the interface unit (not shown) receives various inputs from a user or a user terminal, and performs an operation triggered according to the various inputs. For example, the interface unit (not shown) may receive a selection input for a specific node constituting a network, and provide the frequency of a title keyword, related content, and the like in response to the selection input. A detailed description of this will also be described later with reference to FIGS. 16 to 18.

Each component of FIG. 4 may refer to software or hardware such as a Field Programmable Gate Array (FPGA) or an Application-Specific Integrated Circuit (ASIC). However, the components are not limited to software or hardware, and may be configured to be in an addressable storage medium, or may be configured to execute one or more processors. The functions provided in the above components may be implemented by more subdivided components, or may be implemented as one component that performs a specific function by combining a plurality of components.

Next, FIG. 5 is a hardware configuration diagram of an educational content device 100 according to another embodiment of the present invention.

Referring to FIG. 6, the educational content device 100 includes one or more processors 101, a bus 105, a network interface 107, and a memory 103 that loads a computer program executed by the processor 101. ), and a storage 109 for storing the educational content software 109a. However, only components related to an embodiment of the present invention are shown in FIG. 6. Accordingly, those of ordinary skill in the art to which the present invention belongs may recognize that other general-purpose components may be further included in addition to the components illustrated in FIG. 6.

The processor 101 controls the overall operation of each component of the educational content device 100. The processor 101 includes a CPU (Central Processing Unit), MPU (Micro Processor Unit), MCU (Micro Controller Unit), GPU (Graphic Processing Unit), or any type of processor well known in the technical field of the present invention. Can be. In addition, the processor 101 may perform an operation on at least one application or program for executing the method according to the embodiments of the present invention. The educational content device 100 may include one or more processors.

The memory 103 stores various types of data, commands and/or information. The memory 103 may load one or more programs 109a from the storage 109 to execute the content analysis method according to embodiments of the present invention. In Fig. 6, a RAM is shown as an example of the memory 103.

The bus 105 provides a communication function between components of the educational content device 100. The bus 105 may be implemented as various types of buses such as an address bus, a data bus, and a control bus.

The network interface 107 supports wired/wireless Internet communication of the educational content device 100. In addition, the network interface 107 may support various communication methods other than Internet communication. To this end, the network interface 107 may be configured to include a communication module well known in the art.

The storage 109 may non-temporarily store one or more programs 109a and a search keyword 109b input by a user. In FIG. 6, educational content software 109a is shown as an example of the one or more programs 109a.

The storage 109 is a nonvolatile memory such as a ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, etc., a hard disk, a removable disk, or well in the technical field to which the present invention belongs. It may be configured to include any known computer-readable recording medium.

Educational content software (109a) can execute the educational content generation method using big data according to an embodiment of the present invention. For example, the keyword analysis software 109a is loaded into the memory 103 and, by one or more processors 101, an operation of determining a main subject among subject words tagged to each of the plurality of search target textbooks, and the determined main subject words. An operation of extracting the first subject keyword for the search keyword input by the first user from the subject of the search keyword input by the first user associated with the first subject word of interest among the subject words, as big data among the determined main subject words An operation of extracting a second title keyword for a search keyword input by the second user from the title of the constructed textbook content big data, based on the first title keyword and the second title keyword, the first main An operation of calculating a similarity of a subject word between the subject word and the second subject word, and an operation of providing a search result for a search keyword input by the user based on the calculated similarity of the subject word may be executed.

So far, the configuration and operation of the educational content device 100 according to an embodiment of the present invention has been described with reference to FIGS. 4 and 5. Next, a method of generating educational content using big data according to another embodiment of the present invention will be described in detail with reference to FIGS. 6 to 18.

Hereinafter, each step of the method for generating educational content using big data according to an embodiment of the present invention may be performed by a computing device. For example, the computing device may be an educational content device 100. However, for convenience of explanation, the description of the operation subject of each step included in the method for generating educational content using big data may be omitted. In addition, each step of the method for generating educational content using big data may be implemented as an operation executed by the processor 101.

6 is a flowchart of a method of generating educational content using big data according to an embodiment of the present invention. However, this is only a preferred embodiment for achieving the object of the present invention, and of course, some steps may be added or deleted as necessary.

Referring to FIG. 6, in step S100, a plurality of contents are obtained. For example, the plurality of contents may be contents obtained as a search result through a predetermined search tool. For another example, the plurality of contents may be contents obtained through crawling from a preset data source. The method of acquiring the content can be any method.

In step S200, a search target period is determined based on the number of published contents for each time period. In addition, the search keyword input by the user may be determined as a content issued during the search target period among the plurality of acquired contents. However, according to embodiments, the search target period may be set by a user or may not be limited to a specific period. Hereinafter, a method of determining the search target period by the educational content device 100 will be further described with reference to FIG. 8.

7 shows an example of determining a search target period when the unit time period is set to "1 year".

Referring to FIG. 7, when the unit time interval is "1 year", the number of published contents per year is calculated based on the publication year of the acquired textbooks. Then, the first time interval in which the calculated number of published contents per year becomes equal to or greater than the preset threshold value 430 may be determined as the start interval of the search target period 450. 7 is shown as an example that 2010 is determined as the start time section of the search target period 450. The last time interval of the search target period 450 may be the most recent time interval, or may be the last time interval in which the number of published contents per year exceeds the threshold value 430. According to the present embodiment, a time section in the past in which the number of published contents per time section is less than a threshold value may be excluded from the search target period in consideration of the fact that the need for search is low. Accordingly, since the number of search keywords input by the user is reduced, computing cost required to perform a search can be saved.

According to an embodiment of the present invention, an average number of published contents per time section may be calculated, and a search target period may be determined based on this. For example, when the unit time section is set to "3 years", as shown in FIG. 7, after calculating the number of published contents by year, the average number of published contents may be calculated in units of three years. Then, a time interval corresponding to the first three years in which the average number of published contents becomes greater than or equal to the threshold value may be determined as a start time interval of the search target period. According to the present embodiment, the search target period may be more accurately determined in consideration of fluctuation of the frequency.

Further, according to an embodiment of the present invention, a plurality of search target periods may be determined. For example, the search target period 450 shown in FIG. 7 is determined as the total search target period, the first time period included in the total search target period becomes the first search target period, and the total search target period The second time period included in may be a second search target period. In this case, a network visualized for each search target period may be provided as educational content.

Referring back to FIG. 6, in step S300, a keyword tagged to each search keyword input by a user is extracted.

For example, the tagged subject word may be a keyword included in bibliographic information of a textbook when a search keyword input by a user is related to language education.

In step S400, the main subject words are determined among the tagged subject words. A detailed description of this step (S400) will be described later with reference to FIGS. 8 to 11.

In step S500, the similarity of the subject words between the main subject words is calculated. The degree of similarity between the main subject words is calculated based on the degree of similarity between subject keywords extracted from the subject of the search keyword input by the user associated with the main subject words. A detailed description of this step (S500) will be described later with reference to FIGS. 8 to 11.

In step S600, a search result for a search keyword input by the user is provided based on the similarity of the subject word. For example, the search result may be a visualized network. A detailed description of this step (S600) will be described later with reference to FIGS. 15 and 16.

So far, a method of generating educational content using big data according to an embodiment of the present invention has been described with reference to FIGS. 6 and 7. Hereinafter, a method of determining a main subject word performed in step S400 will be described with reference to FIGS. 8 to 11.

8 is a detailed flowchart of step S400 shown in FIG. 6.

Referring to FIG. 8, in step S410, a frequency count for each time section is calculated for each tagged subject word.

In step S430, general-purpose main words are excluded from the tagged main words based on the calculated frequency for each time section. Specifically, as shown in FIG. 9, a main word whose frequency for each time period continuously exceeds a threshold value 470 over a plurality of time periods is determined as a general-purpose main word. In addition, among the tagged subject words, the general purpose subject words are excluded.

Depending on the embodiment, the educational content device 100 may determine a general-purpose main word using the average and/or standard deviation (or variance) of the frequency for each time section. For example, a main word whose average frequency for each time section is equal to or greater than a first threshold value and whose standard deviation or variance is equal to or less than a second threshold value may be determined as a general-purpose subject word.

In step S450, a main subject word is determined from among the remaining subject words from which the universal subject word is excluded based on the frequency of each time section.

In an embodiment, the main subject word may be determined as a subject word in which the frequency of each time period continuously increases over a plurality of time periods. For example, as shown in FIG. 10, a main subject with an increasing trend in the frequency of each time section may be a main subject. According to the present embodiment, a subject in which interests of educators in a number of related fields are reflected may be determined as a main subject. Accordingly, more valuable educational content can be provided to educators in related fields, and in particular, educational content reflecting the latest educational trends can be provided.

In an embodiment, the main subject word may be determined as a subject word in which the frequency of each time period continuously decreases over a plurality of time periods. For example, as illustrated in FIG. 11, a main subject with a decreasing trend in the frequency of each time section may be a main subject. According to the present embodiment, a subject that is distracting from educators in a number of related fields may be determined as a main subject. Accordingly, educational contents of various aspects can be provided.

So far, with reference to FIGS. 8 to 11, a method of determining a main subject word according to an embodiment of the present invention

Explained. According to the above-described method, a general-purpose subject word is excluded, and a subject in which a plurality of educators' interests are reflected may be determined as the subject of interest. Accordingly, valuable educational contents reflecting educational trends can be provided to users.

Hereinafter, a method of calculating the similarity of the main word performed in step S500 will be described with reference to FIGS. 12 to 15.

12 is a detailed flowchart of step S500 shown in FIG. 6. For ease of understanding, FIG. 12 shows a method of calculating the similarity between two subject words of interest (hereinafter, referred to as “first main subject words” and “second main subject words”, respectively).

Referring to FIG. 12, in step S510, a first title keyword for a first search target content is extracted from a title of a search keyword input by a first user associated with a first main subject word. 13, at least one natural language processing algorithm well known in the art may be used to extract the title keywords 520a to 520n from the title 510 of the content. The natural language processing algorithm It is okay if an algorithm is used.

In addition, the search keyword input by the first user may refer to content indexed by the first main subject word or content tagged with the first main subject word among search keywords input by all users. As shown in FIG. 12B, a plurality of contents may exist as contents indexed by the main subject word 530.

In step (S530), from the title of the textbook content big data built with big data, a second title keyword for the search keyword input by the second user is extracted.

In step S550, a first keyword vector for a first main subject word is generated based on the frequency of the first subject keyword. Specifically, the first keyword vector may be generated according to Equation 1 below. In Equation 1 below, Mi refers to the i-th main keyword, Tkj refers to the j-th title keyword, V refers to a vector, and freq(Tkj|Mi) is a search input by a user associated with the i-th subject of interest. Refers to the frequency of occurrence of the jth title keyword in the keyword.

Equation 1

With reference to FIG. 13, this step (S550) will be further described.

Referring to FIG. 13, in the titles 540a to 540n of search keywords input by a user associated with the main subject word 530, the frequency of each occurrence of the title keywords 550 is calculated. In more detail, the frequency 560a in which the title keyword 550a appears in the titles 540a to 540n of the search keyword input by the user is calculated, and the title keyword 550b is the title of the search keyword input by the user ( The frequency number 560b that appears in 540a to 540n is calculated. This process is performed for each of the title keywords 550.

Next, a keyword vector 570 having the frequency of each title keyword as a vector element is generated. That is, the vector element 570a of the keyword vector 570 is set to the frequency 560a of the title keyword 550a, and the vector element 570b of the keyword vector 570 is the frequency 560b of the title keyword 550b. Is set to In the same manner as above, in step S550, a first keyword vector having a k-dimension (where k is a natural number greater than or equal to 1) for the first main keyword may be generated.

Referring back to FIG. 12, in step S570, a second keyword vector for the second main subject word is generated based on the frequency of the second subject keyword. The second keyword vector may also be generated as a k-dimensional vector in the same manner as the first keyword vector.

For reference, until now, the value of the vector element constituting the keyword vector has been described as being limited to the frequency of each title keyword, but according to an embodiment of the present invention, the value of the vector element is TF-IDF (term frequency- inverse document frequency). Alternatively, the value of the vector element may be an IDF value, which may vary according to embodiments.

In addition, in the case of FIG. 12, the description is limited to the first and second major keywords, but as described above, this is only for convenience of understanding, and as shown in FIG. Is created.

Referring back to FIG. 12, in step S590, the similarity of the first keyword and the second keyword is calculated by using the similarity between the first keyword vector and the second keyword vector.

In an embodiment, a degree of similarity between the first keyword vector and the second keyword vector may be calculated as a cosine similarity as shown in Equation 2 and FIG. 14 below. The cosine similarity is well known in the art, and a description thereof will be omitted.

Equation 2

In an embodiment, the similarity between the first keyword vector and the second keyword vector may be calculated as Euclidean similarity. That is, the similarity between the first keyword vector and the second keyword vector may be calculated based on a Euclidean distance between points indicated by two vectors in a vector space model as illustrated in FIG. 14.

In addition, the similarity between the first keyword vector and the second keyword vector may be calculated in various ways.

So far, with reference to FIGS. 12 to 15, a method of calculating the similarity of a subject word according to an embodiment of the present invention has been described. According to the above-described method, the similarity between the main subject words is calculated based on the similarity of the related subject keywords of the main subject words. Since the title of the content implies a core educational topic that the author intends to express and includes the latest keywords in the field, according to this embodiment, the similarity between the main subject words can be accurately calculated through the similarity of the title keywords. . In addition, it can be prevented from acting as a strong hub in the network in which general-purpose keywords and search words are visualized.

Hereinafter, a method of providing a search result performed in step S600 will be described with reference to FIGS. 16 to 18.

According to an embodiment of the present invention, the educational content device 100 may provide a relationship between the main subject words in a visualized network based on the similarity of the main subject words. In the visualized network, each node corresponds to a main subject word, and a weight of an edge may be determined based on a similarity of main subject words between main subject words. Also, adjacent to each node

Corresponding main keywords can be displayed together.

An example of a network created according to an embodiment of the present invention is shown in FIG. 16. 16 is a network generated based on a search keyword input by the same user as the network shown in FIG. 1, but it can be seen that the network type is different. In particular, unlike the network illustrated in FIG. 1, it can be seen that there is no node operating as a strong hub.

In one embodiment, the visualized network may be configured based on a pathfinder network search. According to the present embodiment, educational content with higher accuracy than configuring a network based on a minimum spanning tree may be provided. This is because, when a network is configured based on the minimum spanning tree, important information may be lost due to the removal of high-important edges. The Pathfinder network search can be clearly understood by those of ordinary skill in the art, and a description thereof will be omitted.

Hereinafter, a method of providing various educational contents according to some embodiments of the present invention will be described. Hereinafter, for convenience of description, the first node is a node corresponding to the first main subject word, and the second node is set to be a node corresponding to the second main subject word.

In an embodiment, the color of each node constituting the network may be determined based on a category to which a corresponding main subject word belongs. For example, when a first main subject word belongs to a first category, a first node may be visualized with a first color, and if a second main subject word belongs to a second category, a second node may be visualized with a second color. In this case, a category to which each main subject word belongs may be predefined. For example, when the subject word is a subject related to language education, the subject word may be a term located in a lower layer of educational content, and the category may refer to a term located in a higher layer.

In an embodiment, at least one visual element of a length and a thickness of a corresponding trunk line may be determined based on a similarity (or weight of the trunk line) between the key words. For example, as the similarity of the main word between the first and second main words increases, the length of the trunk line between the first node and the second node becomes shorter or the thickness of the trunk line becomes thicker. According to the present embodiment, since the user can more intuitively grasp the similarity information between main subject words, there is an effect of improving information transmission.

In an embodiment, the size of each node constituting the network may be determined based on the number of neighboring nodes, the order of nodes, and/or weights of trunk lines. For example, the size of the first node constituting the network may be determined based on the number of first neighboring nodes adjacent to the first node and/or the number of second neighboring nodes adjacent to the first neighboring node. For another example, the size of the first node may be determined based on the number of first neighboring nodes adjacent to the first node and/or a weight of a trunk line with each neighboring node. According to the present exemplary embodiment, a size of a node having a trunk line with a large number of neighboring nodes or a high weight assigned to it may be visualized relatively large. Accordingly, a key subject word among the main subject words can be highlighted, and information transferability can be further improved.

In some embodiments, when a selection input for a specific node is received from a user, additional information may be provided to the user. Hereinafter, an embodiment in which the additional information is provided will be briefly described.

In one embodiment, in response to a selection input for a first node among nodes constituting the network, the title keyword extracted from the title of the search keyword input by the user associated with the first main keyword and the frequency information of each title keyword are Can be provided. That is, an element value of a keyword vector corresponding to the first main subject word may be provided for each title keyword. For example, as illustrated in FIG. 17, when a selection input for the node 610 is received, frequency information 630 of the title keyword may be provided. 16 illustrates that the frequency information is provided in the form of a graph, but the frequency information may be provided in any manner. According to the present embodiment, since keyword information for each main subject word is provided, it is possible to help a user to easily identify educational content.

In one embodiment, in response to a selection input for a first node among nodes constituting a network, a list of search keywords input by a user associated with the first main keyword may be provided. Alternatively, in response to a selection input for the title keyword provided in the above-described embodiment, a list of contents including the selected title keyword may be provided. According to the present embodiment, a search function based on a subject of interest or a keyword of interest of a user may be provided through a user-friendly interface. Accordingly, a user can quickly search for related content, and user convenience and satisfaction can be improved.

In one embodiment, when a specific main keyword is input from a user, a specific node corresponding to the specific main keyword among nodes constituting a network and a node located within a preset distance from the specific node among the nodes constituting the network in response to the input may be highlighted. have. Also, the same processing may be performed even when a selection input for a specific node is received from the user.

In the above-described embodiment, when a plurality of networks including educational content for different search target periods are simultaneously provided, highlight processing may be simultaneously performed on each network. This will be further described with reference to FIG. 18.

FIG. 18 shows a first network 650 for a first search target period and a second network 670 for a second search target period that is different from the first search target period. Shows an example of input. In addition, FIG. 17 shows a case where the preset distance is set to "2". Referring to FIG. 18, when the specific main keyword is input from a user, a node 653 located within a preset distance from a node 651 corresponding to the specific main control of interest in a first network 650 and a node 651 Highlight processing for can be performed. In this case, according to an exemplary embodiment, the node 651 corresponding to the specific main keyword may perform highlight processing with a different color, size, and color density than the node 653. At the same time, the second network 670 may perform highlight processing on the node 671 corresponding to the specific subject of interest and the node 673 located within a preset distance from the node 671. According to the present embodiment, when a specific main subject word is input, the user intuitively provides the trend of the input main subject through the location occupied by the node corresponding to the input main subject in each network, the number of neighboring nodes, and the number of links Can spot change.

So far, a method of providing educational content according to an embodiment of the present invention has been described with reference to FIGS. 16 to 18. Finally, with reference to Figs. 19 and 20, an evaluation result of a network provided according to an embodiment of the present invention will be briefly described.

19 illustrates a plurality of clusters 710 to 760 designated by an expert in the technical field related to a search keyword input by a user. Referring to FIG. 19, it can be seen that the network generated according to the embodiment of the present invention does not have a strong hub node and includes a plurality of sub-networks in a radiated form. In addition, it can be confirmed that each sub-network corresponds to the clusters 710 to 760 designated by the expert. This can be understood as indicating that, when the similarity of the main subject words is calculated based on the similarity of the subject keyword, the subject words with high relevance are actually concentrated and the effect of the strong hub node is suppressed. In addition, it can be understood that it has been confirmed that the method of generating educational content using big data according to an embodiment of the present invention provides high-quality educational content with high accuracy.

So far, a method of generating educational content using big data according to an embodiment of the present invention has been described with reference to FIGS. 6 to 20. According to the above-described method, a network is provided which visualizes the relationship between topics included in the search keyword input by the user as educational content for the search keyword input by a given user. Through this, the user can intuitively check the education trend information, and information delivery of search information can be improved.

The concept of the present invention described so far with reference to FIGS. 6 to 20 may be implemented as computer-readable codes on a computer-readable medium. The computer-readable recording medium is, for example, a removable recording medium (CD, DVD, Blu-ray disk, USB storage device, removable hard disk) or a fixed recording medium (ROM, RAM, computer-equipped hard disk). I can. The computer program recorded in the computer-readable recording medium may be transmitted to another computing device through a network such as the Internet and installed in the other computing device, thereby being used in the other computing device.

Although the operations are illustrated in a specific order in the drawings, it should not be understood that the operations must be executed in the specific order shown or in a sequential order, or all illustrated operations must be executed to obtain a desired result. In certain situations, multitasking and parallel processing may be advantageous. Moreover, the separation of the various components in the above-described embodiments should not be understood as necessitating such separation, and the program components and systems described may generally be integrated together into a single software product or packaged into multiple software products. It should be understood that there is.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those of ordinary skill in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. I can understand. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting.

Claims (20)

In the educational content generation method using big data performed by an educational content device,
Acquiring a search keyword input by a user;
Determining a main subject word among subject words tagged to each of the search keywords input by the user;
Extracting a first title keyword for the first search target content from a title of a search keyword input by a first user associated with a first main keyword among the determined main subject words;
Extracting a second title keyword for the teaching material content big data from the title of the teaching material content big data constructed with big data among the determined main subject words;
Calculating a similarity of a subject word between the first subject word and the second subject word of interest based on a degree of similarity between the first subject keyword and the second subject keyword; And
Including the step of providing a search result for the search keyword input by the user based on the calculated similarity of the main word,
The step of calculating the similarity of the main word between the first main subject and the second main subject,
Generating a first keyword vector for the first main subject word based on the frequency of the first subject keyword;
Generating a second keyword vector for the second main subject word based on the frequency of the second subject keyword; And
Calculating the similarity of the main word by using the similarity between the first keyword vector and the second keyword vector
Characterized in that it comprises a,
A method of creating educational content using big data.
The method of claim 1,
Further comprising the step of determining a search target period using the number of issued contents per time period for the search keyword input by the user,
The step of determining the main subject word,
From among the search keywords input by the user, determining the subject of interest based on a search keyword input by the user issued during the search target period,
A method of creating educational content using big data.
The method of claim 2,
The step of determining the search target period,
For each of a plurality of time intervals, calculating the number of issued contents for each time interval for the search keyword input by the user; And
It characterized in that it comprises the step of determining, from among the plurality of time intervals, a time interval in which the calculated number of published contents for each time interval is equal to or greater than a threshold value as a start interval of the analysis target period,
A method of creating educational content using big data.
The method of claim 1,
The step of determining the main subject word,
For each of a plurality of time intervals, calculating a frequency for each time interval for each of the tagged main words;
Excluding general-purpose main words from the tagged main words based on the calculated frequency for each time section; And
Including the step of determining the main subject from among the remaining subject words from which the universal subject word is excluded,
The general term is,
Over a plurality of time intervals, the calculated frequency for each time interval is a key word that continuously exceeds a threshold value,
A method of creating educational content using big data.
The method of claim 1,
The step of determining the main subject word,
For each of a plurality of time intervals, calculating a frequency for each time interval for each of the tagged main words; And
It characterized in that it comprises the step of determining, among the tagged subject words, the main subject words in which the calculated frequency of each time period continuously increases over the plurality of time periods,
A method of creating educational content using big data.
The method of claim 1,
The step of determining the main subject word,
For each of a plurality of time intervals, calculating a frequency for each time interval for each of the tagged main words; And
Among the tagged subject words, determining a main subject word in which the calculated frequency of each time period continuously decreases over the plurality of time periods as the main subject word,
A method of creating educational content using big data.
The method of claim 1,
The first title keyword includes a 1-1 title keyword and a 1-2 title keyword,
The first keyword vector includes a first vector element and a second vector element,
The first vector element,
The 1-1 title keyword is determined based on a frequency number appearing in the title of the search keyword input by the first user,
The second vector element,
Characterized in that the 1-2 title keyword is determined based on the frequency of appearing in the title of the search keyword input by the first user,
A method of creating educational content using big data.
The method of claim 1,
The degree of similarity between the first keyword vector and the second keyword vector is
Characterized in that it is calculated based on cosine similarity,
A method of creating educational content using big data.
The method of claim 1,
Providing a search result for the search keyword input by the user,
Including the step of providing the relationship between the determined main subject words in a visualized network,
The first node constituting the visualized network corresponds to the first main keyword,
The second node constituting the visualized network corresponds to the second main keyword,
The weight of the trunk line between the first node and the second node is determined based on a similarity of a main word between the first main subject and the second main subject,
A method of creating educational content using big data.
The method of claim 9,
The visualized network,
Characterized in that visualized through a pathfinder network search,
A method of creating educational content using big data.
The method of claim 9,
The first main keyword is classified into a first category among the predefined categories,
When the second main keyword is classified as a second category among the predefined categories,
The first node is visualized with a first color, and the second node is visualized with a second color,
A method of creating educational content using big data.
The method of claim 9,
At least one visual element of the length of the trunk line and the thickness of the trunk line between the first node and the second node,
Characterized in that it is determined based on the similarity of the main word between the first main subject and the second main subject,
A method of creating educational content using big data.
The method of claim 9,
The size of the first node is,
Characterized in that it is determined based on the number of first neighboring nodes adjacent to the first node and the number of second neighboring nodes adjacent to the first neighboring node,
A method of creating educational content using big data.
The method of claim 9,
Receiving a selection input for the first node from a user; And
In response to the selection input of the first node, the first search associated with the first title keyword and the first main keyword
It characterized in that it further comprises the step of providing information on the frequency of occurrence of the first title keyword in the title of the target content,
A method of creating educational content using big data.
The method of claim 14,
The first title keyword includes a 1-1 title keyword and a 1-2 title keyword,
Receiving a selection input for the first-1 title keyword from the user; And
In response to a selection input of the 1-1 title keyword, providing a list of contents including the 1-1 title keyword among search keywords input by the first user. ,
A method of creating educational content using big data.
The method of claim 9,
Receiving a selection input for the first node from a user; And
In response to receiving the selection input, further comprising the step of providing a list of search keywords input by the first user associated with the first main subject word,
A method of creating educational content using big data.
The method of claim 9,
The search keyword entered by the user,
A search keyword entered by a user issued during a first search target period and a search keyword input by a user issued during a second search target period different from the first search target period,
Providing the determined relationship between the main subject words in a visualized network,
Providing, to a first network, a relationship between main subject words determined from a search keyword input by a user issued during the first search target period; And
It characterized in that it comprises the step of providing a relationship between the main subject words determined from the search keyword entered by the user issued during the second search target period to a second network,
A method of creating educational content using big data.
The method of claim 17,
Receiving a third main key word from a user;
Highlighting nodes located within a preset distance from the 3-1 node and the 3-1 node corresponding to the third main key word in the first network; And
In the second network, it characterized in that it comprises the step of highlighting a node 3-2 corresponding to the third main keyword and a node located within a preset distance from the node 3-2,
A method of creating educational content using big data.
One or more processors;
Network interface;
A memory for loading a computer program executed by the processor; And
Including a storage for storing the search keyword and the computer program input by the user,
The computer program,
An operation of determining a main subject word among subject words tagged to each of the search keywords input by the user;
An operation of extracting a first title keyword for the first search target content from a title of a search keyword input by a first user associated with a first main keyword among the determined main subject words;
In the title of the textbook content big data built with big data among the determined main keywords, the second search target statement
An operation of extracting a second title keyword for the second term;
An operation of calculating a similarity of a subject word between the first subject word and the second subject word of interest based on a degree of similarity between the first subject keyword and the second subject keyword; And
Including an operation of providing a search result for a search keyword input by the user based on the calculated similarity of the subject word,

The operation of calculating the similarity of the subject word between the first subject word and the second subject word is,
An operation of generating a first keyword vector for the first main keyword based on the frequency of the first subject keyword;
An operation of generating a second keyword vector for the second main keyword based on the frequency of the second subject keyword; And
And an operation for calculating the similarity of the main word by using the similarity between the first keyword vector and the second keyword vector,
A method of creating educational content using big data.
Combined with a computing device,
Obtaining a search keyword input by a user;
Determining a main subject word among subject words tagged to each of the search keywords input by the user;
Extracting a first title keyword for the first search target content from a title of a search keyword inputted by a first user associated with a first main keyword among the determined main subject words;
In the title of the textbook content big data built with big data among the determined main keywords, the second search target statement
Extracting a second title keyword for heun;
Calculating a similarity of a subject word between the first subject word and the second subject word of interest based on a degree of similarity between the first subject keyword and the second subject keyword; And
When executing the step of providing a search result for the search keyword input by the user based on the calculated similarity of the main word
Keyed, the step of calculating the similarity of the main word between the first main subject and the second main subject,
Generating a first keyword vector for the first main keyword based on the frequency of the first subject keyword;
Generating a second keyword vector for the second main keyword based on the frequency of the second subject keyword; And
A computer program stored in a computer-readable recording medium comprising the step of calculating the similarity of the main word by using the similarity between the first keyword vector and the second keyword vector.

Images