JP2017142315A - Learning order generation device, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a learning order generation device, a method, and a program that can generate a learning order for each topic.SOLUTION: A learning order generation device 1 includes: a reference information acquisition part 11 acquiring reference information defining learning orders of a plurality of items; a key word extraction part 12 extracting a key word from each item; and a combination processing part 13 dividing the items into groups based on the similarities of the key words and generating the learning order of a key word by a group-by-group basis.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a learning order generation apparatus, method, and program for generating a learning order of items.

  2. Description of the Related Art Conventionally, Internet technology has been used in e-learning and other educational fields. In addition, with the development of content distribution technology and curation services, video / audio content, Web pages, and the like have been utilized as learning materials. For this reason, it is imagined that the contents related to the teaching materials will increase without being organized by digitizing the learning materials such as textbooks and utilizing a large number of contents. In this case, it becomes difficult for the student to determine which content should be used in what order. In order to improve the efficiency and effectiveness of learning, a mechanism for learning necessary knowledge in order is required.

  Under such circumstances, Patent Document 1 proposes a technique for determining an order relationship between teaching materials based on similarity to a curriculum in which a lecture order is defined. Patent Document 2 proposes a technique for evaluating the consistency between sequentially arranged teaching materials according to the number of undefined nouns.

Japanese Patent Laying-Open No. 2015-4785 JP 2003-287998 A

  However, a plurality of items (topics) to be learned are mixed in the curriculum information in which a learning order such as a textbook or a learning guidance guide is described. For example, for a topic learned in the first year of junior high school, when learning more detailed contents in the second year of junior high school, the same topic is not consecutive in the teaching guidelines, so the learning order for each topic in the content group as teaching materials It was difficult to figure out.

  An object of this invention is to provide the learning order production | generation apparatus, method, and program which can produce | generate the learning order for every topic.

  The learning order generation device according to the present invention includes a reference information acquisition unit that acquires reference information in which a learning order of a plurality of items is defined, a keyword extraction unit that extracts a keyword from each of the plurality of items, and similarity of the keywords A combination processing unit that divides the plurality of items into groups based on the characteristics and generates the learning order of the keywords for each group.

  The combination processing unit may divide the plurality of items into groups based on similarity and inheritance of the keywords.

  The learning order generation device identifies a group based on a topic determination unit that determines a topic word of the group from a keyword for each group, and an input topic word, and determines a learning order of the keyword in the group. An output unit for outputting.

  The learning order generation device matches a keyword extracted from a plurality of input teaching material contents with a keyword for each group, and determines a learning order of the plurality of teaching material contents based on the learning order of the keywords And an output unit that outputs a learning order of the plurality of teaching material contents.

  The output unit may output a keyword that is not included in the plurality of teaching material contents among the keywords for each group.

  The learning order generation device matches a keyword extracted from the input teaching material content with a keyword for each group, and based on the learning order of the keyword among the keywords not included in the teaching material content, You may provide the matching part which extracts the keyword which should be learned before teaching material content, and the output part which outputs the keyword extracted by the said matching part.

  The learning order generation device matches a keyword extracted from the input teaching material content with a keyword for each group, and based on the learning order of the keyword among the keywords not included in the teaching material content, You may provide the matching part which extracts the keyword which should be learned after teaching material content, and the output part which outputs the keyword extracted by the said matching part.

  The learning order generation method according to the present invention includes a reference information acquisition step of acquiring reference information in which a learning order of a plurality of items is defined, a keyword extraction step of extracting a keyword from each of the plurality of items, and similarity of the keywords The computer executes a combination processing step of dividing the plurality of items into groups based on the sex and generating the learning order of the keywords for each group.

  The learning order generation program according to the present invention includes a reference information acquisition step of acquiring reference information in which a learning order of a plurality of items is defined, a keyword extraction step of extracting a keyword from each of the plurality of items, and similarity of the keywords The computer is caused to perform a combination processing step of dividing the plurality of items into groups based on the characteristics and generating the learning order of the keywords for each group.

  According to the present invention, a learning order for each topic can be generated.

It is a figure which shows the function outline | summary of the learning order generation apparatus which concerns on embodiment. It is a figure which shows the structure of the learning order generation apparatus which concerns on embodiment. It is a figure which shows the example of extraction of the keyword which concerns on embodiment. It is a figure which shows the outline | summary of the similarity and inheritance of the keyword which concerns on embodiment. It is a figure which shows the example of a combination process of the item which concerns on embodiment. It is a figure which shows the 1st example of the service which concerns on embodiment. It is a figure which shows the 2nd example of the service which concerns on embodiment. It is a figure which shows the 3rd example of the service which concerns on embodiment.

Hereinafter, an example of an embodiment of the present invention will be described.
The learning order generation device 1 according to the present embodiment generates a learning order for each topic based on the similarity and inheritance of new keywords from the curriculum information such as a learning instruction guideline or a textbook describing the instruction order. Further, the learning order generation device 1 generates a learning order, and at the same time outputs keywords to be learned at each stage of the learning order so that the user understands the learning content.

FIG. 1 is a diagram illustrating an outline of functions of the learning order generation device 1 according to the present embodiment.
First, the learning order generation device 1 registers reference information that is the basis of educational materials such as a teaching guideline or a textbook in order to structure the learning order.
Subsequently, the learning order generation device 1 performs combination processing with order based on the order of each item of the registered reference information and the keywords described, and groups them in units of topics. In each group, keywords and their order are described.

  For example, in the instruction manual or textbook, “What is a cell” is item 1, “cell division” is item 2, “weather map” is item 3, “typhoon” is item 4, and “genetics” is item 5. Suppose that it is explained. The learning order generation device 1 extracts keywords from each item, calculates the degree of relevance of each item based on the similarity and inheritance of these keywords, and groups them.

  Since each grouped item has an order from item 1 to item 5, the learning order generation device 1 defines the order of the grouped keywords according to this order. For example, in a group (topic) related to “cells”, firstly “cells”, “cell walls”, “nuclei” and “cytoplasm”, and then “cell division”, “chromosome” and “replication” Next, keywords are arranged and structured in the order to be learned, such as “inheritance”, “meiosis”, and “character”.

  The educational service provider first presents the teaching material A having “cells” and “cell walls” by inquiring the learning order generation device 1 about the order of the teaching materials A and B held by itself. It is understood that it is desirable to present teaching material B having “cell division”, “inheritance”, and “meiosis”. In addition, “cytoplasm”, “nucleus”, and “chromosome” are also obtained as keywords that are leaked in the contents to be learned regarding “cells”.

FIG. 2 is a diagram illustrating a configuration of the learning order generation device 1 according to the present embodiment.
The learning order generation device 1 is a server device provided with various interfaces such as a communication unit or an input / output device, or an information processing device (computer) such as a PC (Personal Computer), in addition to a storage unit and a control unit. The learning order generation device 1 realizes various functions by causing the control unit to execute predetermined software (learning order generation program) stored in the storage unit.

  The control unit of the learning order generation device 1 includes a reference information acquisition unit 11, a keyword extraction unit 12, a combination processing unit 13, a topic determination unit 14, a matching unit 15, and an output unit 16.

The reference information acquisition unit 11 designates, for example, a teaching procedure in which a teaching order and its contents are described, or teaching material content numbered in the teaching order as reference information in which the learning order of a plurality of items is defined. , Acquire and accumulate specified reference information.
Here, as an example of the guideline, a learning guideline published by the Ministry of Education, Culture, Sports, Science and Technology is used, and as the teaching material content, a textbook itself developed by a textbook company is mentioned.
The keyword extraction unit 12 extracts keywords from each item of the reference information.

FIG. 3 is a diagram illustrating an example of keyword extraction according to the present embodiment.
In this example, the reference information acquisition unit 11 uses the “cells” content as the item 1, the “cell division” content as the item 2, the “weather map” content as the item 3, the “typhoon” content as the item 4, and the item 5 as the item 5. “Genetic” content is included.

  The keyword extraction unit 12 extracts a keyword group from the contents of the items 1 to 5 and stores it for each item. For example, “cell”, “cell wall”, “nucleus” and “cytoplasm” are extracted from item 1, and “cell division”, “nucleus”, “chromosome” and “replication” are extracted from item 2. .

  Here, the reference information accumulated by the reference information acquisition unit 11 is not limited to a table format as shown in FIG. The keyword extraction unit 12 extracts a noun from a title or text of each item included in the reference information by a technique such as morphological analysis. In keyword extraction, adjustments that take into account the importance of each keyword are also possible based on indices such as TF-IDF. For example, it is possible to make an adjustment such that a keyword such as “experiment” frequently appearing in many items is not extracted because it is determined that the importance is low.

  The combination processing unit 13 divides the extracted keywords into new words that have newly appeared in each item and existing words that have already appeared in the previous item, and based on the similarity of the new words between the items. A plurality of items are divided into topic groups, and a keyword learning order is generated for each group. Further, the combination processing unit 13 improves the grouping accuracy by further taking into account the inheritance of keywords from new words to existing words between items.

FIG. 4 is a diagram showing an overview of keyword similarity and inheritance according to the present embodiment.
The item X and the item Z in the figure mean items that are in the context in the reference information. However, X and Z are not necessarily in a continuous context. W1 to W3 in the figure represent new words that appear for the first time in the item X, and W4 and W5 represent existing words that have already appeared before the item X and have reappeared in the item X. Similarly, W6 and W7 represent new words that appear in the item Z, and W2 represents existing words that appear in the item Z. That is, W2 newly appears in item X and is also used in item Z.

The similarity between new words is based on the similarity between new words of item X and item Z (between any of W1 to W3 and W6 or W7), for example, as in the following formula (1) Is calculated.

Here, the set X and the set Z are composed of only new words. Further, r (WX _m , WZ _n ) represents the similarity between the keywords, and a value obtained by normalizing the sum of the similarities with the number of keywords is defined as similarity R (X, Z) between the new words. .
As the similarity r for each keyword, various indexes such as context similarity or similarity based on the co-occurrence frequency may be used.

The inheritance of a new word is defined as a property that means how much a new word of a certain item exists as an existing word in a later item.
For example, in FIG. 4, among the new words W1, W2, and W3 of item X, W2 exists in item Z as an existing word. Assuming that the number of new words in item X is m as in equation (1) and the number of elements in the intersection of the existing words in item Z and the new words in item X, that is, the number of common words is s, the inheritance of new words The property T is represented by the following formula (2), for example.
T (X, Z) = s / m (2)

  The relevance level of each item is calculated by adding or multiplying the similarity obtained by Expression (1) and the inheritance obtained by Expression (2). Thereby, each item is grouped and the order between items is obtained simultaneously. In addition, the order of keywords can be obtained according to the order between items.

  The similarity and inheritance calculation formulas described above are merely examples, and the present invention is not limited to these. For example, the inheritance may be weighted according to the number of items for keywords inherited over a plurality of items. Thereby, the combination processing unit 13 can more appropriately determine the degree of association between items by multi-level inheritance.

  The topic determination unit 14 determines a topic word representing this group from the keyword group for each group. The topic word determination method is not limited. For example, the most frequently used keyword may be selected, or an algorithm such as a topic extraction model may be used.

FIG. 5 is a diagram showing an example of item combination processing according to the present embodiment.
The keyword group extracted for each item (for example, items 1 to 5) by the keyword extraction unit 12 is classified into a new word or an existing word by the combination processing unit 13, and the order of the items and the similarity of the new word And group based on inheritance.

In this example, items 1, 2, and 5 are grouped in order. The keyword “nuclear” of item 2 is a new word of item 1 and contributes to grouping by improving inheritance from item 1 to item 2.
From the groups of items 1, 2, and 5, “cell”, which is an important word representing the group, is selected as a topic word.

Items 3 and 4 have low relevance to items 1, 2, and 5, are grouped as different topics, and “atmospheric pressure” is selected as the topic word.
In this way, in the reference information, items 1 to 5 are defined as a continuous learning order, but items 1, 2, 5 and items 3, 4 sandwiched between them are different topics, respectively. Grouped.

When a plurality of teaching material contents are input from the outside, the matching unit 15 matches the keywords extracted from these teaching material contents with the keywords for each group and determines the learning order of the teaching material contents based on the learning order of the keywords. To do.
In addition, the matching unit 15 extracts keywords that are not included in the teaching material content as necessary words as a result of matching. At this time, the matching unit 15 separately extracts a keyword to be learned before the input teaching material content and a keyword to be learned later from the necessary words based on the learning order of the keywords.

The output unit 16 outputs various information based on the generated learning order and provides it to the user.
First, when a topic word is input, the output unit 16 specifies a group based on the input, and outputs a learning order of keywords in the specified group.

  Further, the output unit 16 outputs the learning order of the plurality of input teaching material contents as a processing result by the matching unit 15. At this time, the output unit 16 outputs a necessary word that is not included in the plurality of input teaching material contents among the keywords for each group.

  Furthermore, when the learning material content is input, the output unit 16 outputs a keyword to be learned before the input learning material content and a keyword to be learned after the necessary word extracted by the matching unit 15.

<Service example 1>
FIG. 6 is a diagram illustrating a first example of a service using the learning order generation device 1 according to the present embodiment.
In this service example, the learning order generation device 1 calculates how many items are covered by the teaching material contents based on the keywords of the teaching material contents imported from the outside, and the learning order of the teaching material contents and the necessary order. Output words.

  For example, the teaching material B has “cell division”, “genetics”, and “meiosis” as keywords, so that it can be seen that it corresponds to item 2 and item 3. At the same time, it can be seen that teaching material B has no description of the keyword “chromosome” in item 2.

  Further, the order of the teaching material A and the teaching material B is also determined by how many items are covered. Since it is understood that the learning material A covers item 1 and the learning material B covers more items 2, it is understood that learning should be performed in the order of the learning material A to the learning material B.

  Note that the degree of coverage is calculated based on the similarity of each keyword, but the index of similarity is not limited. Further, the type of external teaching material content is not limited as long as text is given, such as video content or audio content.

<Service example 2>
FIG. 7 is a diagram illustrating a second example of a service using the learning order generation device 1 according to the present embodiment.
In this service example, a learning material content holder acquires a review keyword to be learned before content as an educational material owned by the company, or a preparation keyword to be learned next.

The teaching material content holder who owns the teaching material B related to “genetics” wants to know the review keyword to be learned before learning the teaching material B. In this case, by using the learning order generated by the learning order generation device 1, it is possible to understand the necessity of learning the item 1 before the items 2 and 5 corresponding to the teaching material B.
Also, in the corresponding item 2, since it is necessary to learn about “chromosome” that is not in the teaching material B, it can be grasped as a review keyword to be learned in advance.

  In addition, the content holder who wants to know the keyword to prepare for the next learning can understand “cell division” and “chromosome” of the item 2 to be learned next because the teaching material A on “cell division” corresponds to the item 1. . Also, since it is understood that item 5 is next to item 1, “inheritance” and “meiosis” are not presented.

  Thus, since the learning order generation apparatus 1 performs matching based on keywords and their order, not matching of teaching material contents, it is possible to flexibly present preparation and review keywords.

<Service example 3>
FIG. 8 is a diagram illustrating a third example of a service using the learning order generation device 1 according to the present embodiment.
In this service example, an education service provider creates a learning application with an understanding level check using a learning path for each topic provided by the learning order generation device 1.

  The learning order generation device 1 responds with an ordered keyword group in response to a request from the education service provider. For example, from the group related to the topic word “cell”, 1: “cell” “cell wall” “nucleus” “cytoplasm” 2: “cell division” “chromosome” “replication” 3: “inheritance” “meiosis” “ Data of the learning path “trait” is returned to the education service provider.

Based on the acquired learning path, the education service provider, for example, provides a test related to “nuclear” between item 1 and item 2 so that an application that cannot proceed to item 2 unless it understands “nuclear”. Can be created.
Here, in item 1, weighting is performed such that “nucleus” is most important by an index of importance such as TF-IDF, thereby automating test design.

According to the present embodiment, the learning order generation device 1 uses the similarity between new words and the inheritance of new words to group the reference information in each topic to generate a learning order. it can. As a result, the learning order generation device 1 can realize an efficient and effective learning environment, and can contribute to the development of education using Internet technology.
In other words, the education infrastructure development provider can automatically acquire the learning order of each item and keyword by registering the learning materials or instruction points possessed by the company or other companies in the learning order generation device 1, and the learning path for each topic. Can be created.

  Further, the learning order generation device 1 can not only generate the learning order of items, but also clarify the keywords for each item. Therefore, the user can not only systematically manage the increasing teaching material contents but also create an efficient and effective learning opportunity.

  More specifically, by using the learning order generation device 1, the education service provider can match a plurality of teaching material contents with a learning path for each topic and provide it to students in the order to learn. Furthermore, it is possible to present optimal content after understanding the level of understanding of the students.

  Moreover, since the learning order generation device 1 can also present keywords to be learned in each item of the learning order, the user can grasp what is learned in each item and is not covered by the given teaching material content Further, it is possible to grasp the contents to be learned before or after the teaching material contents. As a result, it is possible to improve the learning efficiency using the teaching material content.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. Further, the effects described in the present embodiment are merely a list of the most preferable effects resulting from the present invention, and the effects of the present invention are not limited to those described in the present embodiment.

  Further, the present invention may be realized by recording a program for realizing the function of the learning order generation device on a computer-readable recording medium, causing the computer system to read and execute the program recorded on the recording medium. Good.

  The “computer system” here includes an OS and hardware such as peripheral devices. The “computer-readable recording medium” refers to a storage device such as a portable medium such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, or a hard disk built in a computer system.

  In addition, “computer-readable recording medium” means that a program is dynamically stored for a short time, like a communication line when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. It is also possible to include one that holds a program for a certain time, such as a volatile memory inside a computer system that becomes a server or client in that case. Further, the program may be for realizing a part of the above-described functions, and may be capable of realizing the above-described functions in combination with a program already recorded in the computer system. .

DESCRIPTION OF SYMBOLS 1 Learning order generation apparatus 11 Reference | standard information acquisition part 12 Keyword extraction part 13 Combination processing part 14 Topic determination part 15 Matching part 16 Output part

Claims (9)

A reference information acquisition unit that acquires reference information in which a learning order of a plurality of items is defined;
A keyword extraction unit for extracting a keyword from each of the plurality of items;
A learning order generation device comprising: a combination processing unit that divides the plurality of items into groups based on similarity of the keywords and generates the learning order of the keywords for each group.   The learning order generation device according to claim 1, wherein the combination processing unit divides the plurality of items into groups based on similarity and inheritance of the keywords. A topic determination unit that determines a topic word of the group from the keyword for each group;
The learning order generation device according to claim 1, further comprising: an output unit that identifies the group based on the input topic word and outputs the learning order of the keyword in the group. A matching unit that matches keywords extracted from a plurality of input teaching material contents with a keyword for each group, and determines a learning order of the plurality of teaching material contents based on a learning order of the keywords;
The learning order generation apparatus according to claim 1, further comprising: an output unit that outputs a learning order of the plurality of teaching material contents.   The learning order generation apparatus according to claim 4, wherein the output unit outputs a keyword that is not included in the plurality of teaching material contents among the keywords for each group. The keyword extracted from the input teaching material content is matched with the keyword for each group, and the keywords not included in the teaching material content are learned before the teaching material content based on the learning order of the keywords. A matching unit that extracts keywords that should be
The learning order generation apparatus according to claim 1, further comprising: an output unit that outputs the keyword extracted by the matching unit. The keyword extracted from the input teaching material content is matched with the keyword for each group, and among the keywords not included in the teaching material content, learning should be performed after the teaching material content based on the learning order of the keywords. A matching unit that extracts keywords;
The learning order generation apparatus according to claim 1, further comprising: an output unit that outputs the keyword extracted by the matching unit. A reference information acquisition step for acquiring reference information in which a learning order of a plurality of items is defined;
A keyword extraction step of extracting a keyword from each of the plurality of items;
A learning order generation method in which a computer executes the combination processing step of dividing the plurality of items into groups based on the similarity of the keywords and generating the learning order of the keywords for each group. A reference information acquisition step for acquiring reference information in which a learning order of a plurality of items is defined;
A keyword extraction step of extracting a keyword from each of the plurality of items;
A learning order generation program for causing a computer to execute a combination processing step of dividing the plurality of items into groups based on the similarity of the keywords and generating the learning order of the keywords for each group.

Images