JP2008191226A - Repetitive learning method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a repetitive learning program having a function suppressing addition of a new item and a hint giving function for learning for stepwise achieving voice expression, and to provide a method therefor. <P>SOLUTION: The "lowest review correct answer rate" is set for each review stage, so that in reviewing in each stage, when the "lowest review correct answer rate" is not achieved, a new item data is not added only once, thereby preventing an increase in number of item data for reviewing. Contents of hints of the stages are not fixed according to progress of learning, so that a hint can be given to an item of an incorrect answer even in a review stage having no hint. The amount of hints is increased for reviewing in a lower stage so as not to discourage a learner to learn continuously. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an iterative learning method and program for foreign language learning and the like.

Foreign language learning requires repeated learning of words (pronunciations, spellings, translations) and example sentences to establish memory. There are many educational programs that implement a rational iterative learning method on a computer as a means of supporting the establishment of such memories. For example, “A series that is not forgotten” (a program product provided by Kobeck Co., Ltd.) is also one of educational programs that realizes an iterative learning method on a computer (Non-patent Document 1).
The present inventor once developed this educational program “A series that can not be forgotten” (such as vocabulary) around 1996. This "No more forgotten series" is an iterative learning program that can memorize English meaning and spelling, and the content of the task is to memorize a Japanese translation of a certain number of English words.

  This iterative learning program “A series that is not forgotten” selects the meaning and spelling of English words displayed on a personal computer or the like from the choices, and gives correct / incorrect answers in question format. The learner can learn by simply pressing the number of an option so that the learner can learn like a game. In addition, when a spelling of an English word or the like is mistaken, a “word that becomes a hint” appears, and it is possible to obtain a hint from that and proceed. And, English words are translated and English spellings are written alternately, and the device is designed to prevent learners from getting bored. The wrong word appears several times (repeatedly learning), so it can be remembered naturally without difficulty.

Here, the iterative learning process flow in the iterative learning program “A series that is not forgotten” will be described below.
In the iterative learning process flow, the following steps 1) to 3) exist.
1) Step for preparing a learning plan 2) Step for determining the timing of iterative learning 3) Step for correcting the learning plan

First, the above 1) step of creating a learning plan is a step of calculating a “total number of reviews” from “the total number of items to be stored” and “the number of items to be learned at one time”, and learning is started. It is the first one executed before. Here, the expression “number of items” is used to encompass the number of English words and the number of English sentences for learning spelling and meaning.
For example, if the number of English words to be remembered is 500 words in a certain theme (for example, the first grade of junior high school) and the number to be remembered at a time is 10 words, the “total number of items to be stored” is 500. Thus, “the number of items to be learned at a time when newly appearing” is 10. The “total number of reviews” is calculated using these as input parameters. There are various ways of calculation.

For the “total number of reviews”, for example, simply “divide the total number of items to be stored” by “the number of items to be learned at one time” or “learn at a new time” The power when the power of 2 of the “number of items to be performed” exceeds the “total number of items to be stored” may be used.
In addition, the number of items to be stored in each review session is called “number of set words”.
More specifically, when the “total number of items to be stored” is 500, and the “number of items to be learned once at new start” is 10, the “total number of reviews” is “learned once at new start”. When the power of 2 when “the number of items” exceeds the “total number of items to be stored” is used, the “total number of reviews” is seven.
In addition, the “number of set words” in the first to seventh reviews is 10, 20, 40, 80, 160, 320, and 500, respectively.

Next, in the step 2) of determining the timing of iterative learning, the timing for reviewing the already learned items is determined.
For example, the timing for the Nth review is when the number of items reviewed in the (N-1) th is equal to or greater than the “number of set words”. The final review timing is when the total number of items to be stored is reached.

  More specifically, when the “total number of items to be stored” is 500, and the “number of items to be learned once at new start” is 10, the “total number of reviews” is “learned once at new start”. When the power of 2 when the number of items is greater than the total number of items to be stored is used, the number of newly learned words is 10 or more at the timing of the first review. The timing for the second review is when the number of words reviewed in the first review is 20 words or more, and the timing for the seventh (final) review was the sixth review. This is when the number of words reaches 500 words.

Next, in the step 3) of correcting the learning plan, the “number of set words” of each review is changed according to the correct answer rate to correspond to each learner. Establish a standard correct answer rate for each review time. If the correct answer rate for each review time falls below the set standard correct answer rate, lower the "set number of words", and conversely, if it exceeds the standard correct answer rate, set "number of set words" It is to raise.
Then, the “set word count” is lowered, and the “total review count” is increased when it overlaps with the lower “set word count”. Also, when the “set word count” is increased and overlapped with the upper “set word count”, the “total number of reviews” is reduced.

Series that can't be forgotten English ability-English word learning software (http://www.kobec.co.jp/eitan99.htm)

  In the above-described iterative learning program “No more forgotten series”, when the review process at each stage is completed, new items are learned each time, and items to be reviewed are added. However, in this case, even if the review correct answer rate in the initial stage is less than 50%, review items are added one after another. Therefore, if the correct answer rate continues to be less than 50%, the number of set words greatly exceeded. There is a problem that it is necessary to review the number of items, and it becomes impossible to take an appropriate review timing depending on the “number of set words”.

In the “Don't Forget Series”, the correct answer rate at the initial stage of the first and second review is maintained at 50% or higher. A function that can reduce the number of items is provided.
However, this measure is effective because the content of the “Since no more series” task is “Learn Japanese translation at the word level”. Depending on the task content, the selection format is inappropriate, and even if the number of new learning items is at least two, it may be difficult to maintain the initial correct review rate at 50%. For example, memorize conversational expressions in a language you don't know at all.

In addition, the above-mentioned iterative learning program “Now forgot series” has the problem of storing a Japanese translation of a certain number of words, but it stores a certain number of words and sentence speech expressions. It was not a thing. When storing such a speech expression, it may be difficult to store even if the number of items learned at one time is one. In such a case, it is desirable to memorize memorized speech expressions in stages as the review progresses.
In the iterative learning program, it is assumed that as the review stage progresses, memory stabilizes and becomes long-term memory. However, depending on the pace of learning, the stability of memory and the review stage may not increase proportionally. For this reason, the hint for storing the voice expression needs to be given an appropriate hint according to the achievement stage of the number of items to be learned.

  The present invention provides an iterative learning program and method having a function capable of suppressing the addition of new items and a function of giving hints for learning to achieve stepwise achievement of speech expression in order to respond to the above-mentioned problems and needs. An object of the present invention is to optimize the learning pace of the learner and to efficiently fix the memory in accordance with the memory ability and learning situation of each learner.

In order to achieve the above object, the iterative learning method of the present invention is a learning method for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner in a learner.
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
It is provided with. Thereby, the addition of a new data item can be optimized according to the memory ability and learning situation of each learner.

Moreover, the iterative learning method of the present invention is a learning method for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner in a learner.
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a hint level at each review stage;
3) The step of transitioning the hint level to a higher level or a lower level according to the correct / incorrect answer in the item data unit of the learner at each review stage;
It is provided with. Thereby, the difficulty level of a task can be optimized corresponding to the memory ability and learning situation of each learner.

Moreover, the iterative learning method of the present invention is a learning method for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner in a learner.
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) setting a hint level at each review stage;
4) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
5) a step in which the hint level transitions to a higher level or a lower level according to correct / incorrect answers in the item data unit of the learner at each review stage;
It is provided with.

  Here, the item data of the iterative learning method of the present invention is preferably a foreign language word or a foreign language sentence expressed in katakana, hiragana, or romaji for speech expression learning.

Further, the iterative learning program of the present invention is a learning program for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner for the learner.
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
It is a program which performs the iterative learning method provided with.

Further, the iterative learning program of the present invention is a learning program for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner for the learner.
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a hint level at each review stage;
3) The step of transitioning the hint level to a higher level or a lower level according to the correct / incorrect answer in the item data unit of the learner at each review stage;
It is a program which performs the iterative learning method provided with.

Further, the iterative learning program of the present invention is a learning program for extracting item data from an item data group composed of a plurality of item data and storing the item data in an interactive manner for the learner.
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) setting a hint level at each review stage;
4) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
5) a step in which the hint level transitions to a higher level or a lower level according to correct / incorrect answers in the item data unit of the learner at each review stage;
It is a program which performs the iterative learning method provided with.

  Here, the item data of the iterative learning program of the present invention is preferably a foreign language word or a foreign language sentence expressed in katakana, hiragana, or romaji for speech expression learning.

According to the iterative learning program and method of the present invention, the “minimum review correct answer rate” is set for each review stage. By passing the addition of data, there is an effect that an increase in the number of review item data can be suppressed.
In addition, since the content of hints was not fixed at the review stage according to the progress of learning, hints could be given to items that were incorrect even at the review stage without hints. There is an effect of avoiding the frustration of the learner's continued learning, such as increasing the amount as the review progresses to a lower level.
With the above effects, new data items are added and the difficulty level of tasks is optimized in accordance with the memory ability and learning status of individual learners.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, the scope of the present invention is not limited to the illustrated examples.

FIG. 1 shows a flowchart of a conventional iterative learning program. In traditional iterative learning programs, the focus has been on optimizing the timing of review. In order to optimize the timing of this review, the “number of set words” was increased or decreased. The increase / decrease of the “set word number” will be described below.
First, it is assumed that the increase / decrease width of each review stage is obtained by dividing the “number of set words” of each review stage by the number of review stages. For example, when the “total number of items to be stored” is 500 and the “number of items to be newly learned at one time” is 10, the increase / decrease width at each review stage is set as shown in Table 1 below (the minimum increase / decrease width) The unit is 10). However, the “number of set words” for the first and second times is fixed, and the increase / decrease width for the seventh time is calculated by setting “number of set words” to 640.

In the conventional iterative learning program, in order to optimize the timing of review, the correct answer rate at each review stage is evaluated, and if the correct answer rate is high, the “number of words” is increased by “increase / decrease”, and if the correct answer rate is low A process of reducing the number of set words by “increase / decrease” is performed. In the flowchart of FIG. 1, it is determined whether or not to increase or decrease the “number of set items” from the correct answer rate of the review in the process of step S103. Here, “number of setting items” is synonymous with “number of setting words”. Here, language learning is taken as an example, and the expression “number of set words” is used for explanation.
In the following, processing for determining whether to increase the “set word count” from the correct answer rate of the review and for determining whether to decrease the “set word count” will be described.

a) When determining whether or not to increase the “set word count”, the “predicted set word count” is set to (“set word count” + “increase”) multiplied by the correct answer rate. When “predicted set word count”> “set word count”, “set word count” is increased.
b) When determining whether or not to reduce the “number of set words”, the “number of predicted set words” is a value obtained by multiplying the “number of set words” by the correct answer rate. When “predicted set word count” <(“set word count” − “decrease”)), “set word count” is decreased. For example, in the above example, when the correct answer rate is 90% in the third review stage, the “predicted set word count” is increased to 45 words by multiplying (“set word count” 40 words + increased 10 words) by 90%. Become. Since “predicted set word count” is 45 words> “set word count” is 40 words, “set word count” is changed to 50 words. If the correct answer rate is 60%, the reduced “predicted set word count” is 24 words by multiplying the “set word count” 40 words by 60%. Since “predicted set word count” 24 words <(“set word count” 40 words−decrease 10 words), “set word count” is changed to 30 words.
In the conventional iterative learning program, the “number of set words” is changed every time the review is performed in this way, so that the review timing is optimal for the learner.

A) Setting of minimum review correct answer rate In the conventional iterative learning program, when the review process at each stage is completed, new items are learned each time and review items are added. However, there is a problem in that review items are added one after another even if the review correct answer rate in the initial stage is less than 50%.

  Accordingly, in the iterative learning program of the present invention, “minimum review correct answer rate” L (i) (indicating i: i-th review) is set at each review stage to optimize the addition of new items. If less than the “minimum correct correct answer rate” in review at each stage, the increase in the number of review items is suppressed by passing the learning of new items once. For example, when reviewing for the third time or more, if the “minimum correct answer rate” at each stage is the correct answer rate when the “number of set words” can be lowered, the introduction of new items will be moderately suppressed when there are many reviews It is to be done.

  Next, how to determine the “minimum correct answer rate” is described. The correct answer rate required at each review stage is set to increase as the stage goes up. The learner gradually settles the memory by clearing the required correct answer rate. The review timing is taken by changing the “number of set items” so that this gradual fixing of the memory goes smoothly. Here, reducing the “number of set items” means that it is difficult for the learner to temporarily store the correct answer rate required for the program. If new items are suppressed here, the learner can perform the necessary review earlier, and the subsequent learning becomes smooth. Therefore, the “minimum correct answer rate” is basically the correct answer rate at which the “number of set items” is reduced at each review stage. However, in the initial review where the “number of set items” is fixed, a “minimum review correct answer rate” is determined separately.

B) Stepwise achievement of learning task As described above, depending on the task content, it may be difficult to memorize even if the number of items to be learned at one time is one. In such a case, as the review progresses, it is known that storing speech expressions to be memorized step by step improves the storage efficiency. For example, when you are given the task of being able to speak the Korean word “Odiso Wassoyo” meaning “Where are you from?” .
Since the new expression voice expression is not used, the hint level is subdivided and hints are given so as to be displayed / hidden for each voice expression character unit. Depending on the progress of learning, in consideration of getting used to speech expression, a hint is given to display / hide in syllable units.

(Concrete example)
・ New learning (hint) "Odiso Wasso"
・ First review (hint) "Odiso wah"
・ Second review (hint) "Odiso"
・ The third review (hint) "Odi"
・ After the 4th review (no hint)

In the iterative program, it is assumed that as the review stage progresses, the memory stabilizes and becomes long-term memory. However, depending on the pace of learning, the stability of memory and the review stage may not increase proportionally.
Therefore, in the iterative learning program of the present invention, the hint contents are not fixed at the review stage, and hints are given to items that are incorrect even at the review stage without hints. Increase the amount of hints as you go to the lower level review.

(Concrete example)
-Hint is issued regardless of correct answer or incorrect answer for new learning and first to third time.
・ 4th review (no hint) correct answer ・ 5th review (without hint) incorrect answer
This item data is stored in the “4th review” array.
-4th review (Hint) "Odi" Incorrect answer This item data is stored in the "3rd review" array.
-Third review (Hint) "Odiso" Correct answer This item data is stored in the "4th review" array.
-4th review (Hint) "Odi" Correct answer This item data is stored in the "5th review" array.
-5th review (no hint) Correct answer This item data is stored in the "6th review" array.

  Here, staging by the amount of hints is called a hint level. Also, learning in this way is called stepwise achievement of a learning task. The level of hints to be given at which stage of review is determined in advance for each learning item, and the data is filed.

FIG. 2 is a flowchart of the iterative learning program according to the first embodiment. Hereinafter, the flowchart of the first embodiment will be described with reference to the flowchart of the conventional iterative learning program shown in FIG.
The iterative learning program according to the first embodiment first calculates the total number of review times from the total number of item data to be learned and the number of item data newly generated at one time as in the conventional iterative learning program (step S201). Information at the time of previous learning is read from the data file (step S202). Information at the time of the previous learning includes an array in which learning data numbers for the N-th review are stored, an array in which “number of set items” data for the N-th review is stored, and the number of new items already learned This is an array in which hint levels of learning data numbers are stored.

Next, before reviewing, the above-mentioned minimum review correct answer rate in each review stage is set (step S203), and the above-described hint level in each review stage is set (step S204).
Then, the learner receives the assignment and the hint display, and inputs an answer (step S205). Specifically, a learner displays assignments and hints using a screen of a personal computer or the like, and inputs a reply using a keyboard or a mouse.

  From the answer result, the hint level is shifted to the upper and lower levels according to the correct / incorrect answer of the learner's item data unit at each review stage (step S206). Depending on the correct / incorrect answer for each item data unit, the hint level is raised or lowered for each item data unit. More specifically, as shown in the flowchart of the review part for transitioning the hint level in FIG. 3, the hint level is transitioned by making a comparison determination with a variable parameter indicating the upper and lower limit values of the hint level.

Further, it is determined whether or not the correct answer rate of the review is lower than the “minimum correct answer rate” (step S207). When the correct answer rate is lower than the “minimum review correct answer rate”, the number of item data to be newly added is not added (step S208). If the correct answer rate is equal to or higher than the “minimum review correct answer rate”, the number of newly created item data is added, and the number of new items already learned is updated (step S209).
FIG. 4 shows a flowchart of a correct answer rate evaluation part for setting a flag for suppressing a new item. As described above, the “minimum review correct answer rate” is basically a correct answer rate in which the “number of set items” is reduced at each review stage.

  There is a so-called dispersion effect as a theoretical background for establishing memory in iterative learning. In the experiment of the dispersion effect in memory, when the specific learning item is repeatedly presented several times and the memory performance is examined in the experiment of the dispersion effect in memory, the learning item is presented twice or three times continuously ( Compared to (concentration), memory performance is better when a certain interval is placed between the learning items to be repeated with a certain interval (distribution).

  In order to increase the dispersion effect in iterative learning, it is necessary to optimize the interval between repeated learning items. It is considered that the memory can be smoothly fixed by appropriately taking this interval. The state immediately before the establishment of memory is a state where it seems that it can be remembered but cannot be remembered. If the memory is confirmed in this state, the memory is likely to be fixed.

  For example, when you can recall a name that you have heard a little more, it is easier to remember if you say that name. Optimizing the timing of review in iterative learning is to search for an interval at which such a state occurs. In the iterative learning program and method of the present invention, such a state is also generated by raising or lowering the difficulty level of the task content. That is, by gradually increasing the hint level and learning while reducing the amount of hints, the memory is fixed.

  In order to prove this, the average correct answer rate of the review was compared using the learning item which is different for the learner by alternately using the conventional iterative learning program and the iterative learning program of the present invention. For the seven learners a to f, using the conventional iterative learning program (Monitor A) and the iterative learning program (Monitor B) of the present invention, the correct answer rate is calculated at the end of each learning (15 minutes). I took the data. Table 2 below shows a comparison table of average correct answer rate (review).

  FIG. 5 is a graph showing the temporal change in the average correct answer rate (review) of seven learners, and FIG. 6 is a graph showing the average progress of the seven learners. In FIG. 5, the horizontal axis indicates the number of reviews (one time is 15 minutes), and the vertical axis indicates the average correct answer rate (%) of the review. In FIG. 6, the horizontal axis indicates how many reviews (one time is 15 minutes), and the vertical axis indicates the number of learned items.

  From the difference between the correct answer rate and the progress status, it can be understood that the seven learners can learn more efficiently by using the iterative learning program (monitor B) of the present invention. Monitor B has a large amount of hints in the initial stage, so the correct answer rate is high and the progress is naturally good. However, even if learning progresses and the amount of hints decreases, the average correct answer rate hardly decreases. For this reason, it is considered that the learner inferred that even if the hint disappears on the monitor B, it can be remembered. In fact, some learners said that Monitor A was easy to remember in the first and second learning, but as the number of times progressed, they began to select Monitor B.

In the second embodiment, hint levels in Korean language learning software using the iterative learning program of the present invention will be described. The number of hint level steps is 5 (including no hint), and if there is no incorrect answer, the hint level of the learning item changes as follows.
1) New learning: Hint level 1
2) First review: hint level 2
3) Second review: hint level 3
4) Third review: hint level 4
5) 4th review: hint level 5 (no hint)
6) From the fifth review onwards, no hint will be given unless it is incorrect.

If there is an incorrect answer, for example, if the answer is incorrect at the first review (hint level 2), the first review will transition from hint level 2 to hint level 1. When the correct answer is given at the first review (hint level 1), the second review will transition from hint level 1 to hint level 2.
Also, for example, if the answer is incorrect at the seventh review (hint level 5), the sixth review shifts from hint level 5 to hint level 4. If the answer is incorrect at the sixth review (hint level 4), transition to the hint level 3 at the fifth review.

  When the average correct answer rate and the average hint level at each review stage of the seven learners a to f were examined, it was as shown in Table 3 below.

  From Table 3 above, at the fifth review stage, the hint level is 5.0 and the hint disappears, but the average correct answer rate increases. This is because the hint level is raised or lowered in accordance with the correct answer and the incorrect answer in the initial stage, so that the memory in the initial stage advances smoothly and the hint is gradually unnecessary as the learning progresses.

  The iterative learning program and method of the present invention can be used in an educational software program for foreign language learning.

Flowchart diagram of a conventional iterative learning program The flowchart figure of the iterative learning program of Example 1 Flowchart diagram of the review part that changes the hint level Flowchart of correct answer rate evaluation part that sets a flag to suppress new items A graph showing the change over time in the average correct answer rate (review) of learners Graph showing average progress of learners

Claims (8)

In a learning method for extracting item data from an item data group composed of a plurality of item data and allowing the learner to memorize item data interactively,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
An iterative learning method characterized by comprising: In a learning method for extracting item data from an item data group composed of a plurality of item data and allowing the learner to memorize item data interactively,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a hint level at each review stage;
3) The step of transitioning the hint level to a higher level or a lower level according to the correct / incorrect answer in the item data unit of the learner at each review stage;
An iterative learning method characterized by comprising: In a learning method for extracting item data from an item data group composed of a plurality of item data and allowing the learner to memorize item data interactively,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) setting a hint level at each review stage;
4) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
5) a step in which the hint level transitions to a higher level or a lower level according to correct / incorrect answers in the item data unit of the learner at each review stage;
An iterative learning method characterized by comprising:   4. The iterative learning method according to claim 1, wherein the item data is a foreign language word or a foreign language sentence expressed in katakana, hiragana, or romaji. 5. A learning program that extracts item data from an item data group composed of a plurality of item data and causes the learner to store the item data in an interactive manner,
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
A program for executing an iterative learning method comprising: A learning program that extracts item data from an item data group composed of a plurality of item data and causes the learner to store the item data in an interactive manner,
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a hint level at each review stage;
3) The step of transitioning the hint level to a higher level or a lower level according to the correct / incorrect answer in the item data unit of the learner at each review stage;
A program for executing an iterative learning method comprising: A learning program that extracts item data from an item data group composed of a plurality of item data and causes the learner to store the item data in an interactive manner,
On the computer,
1) calculating the total number of reviews from the total number of item data to be stored and the number of new item data to be stored;
2) setting a minimum review correct answer rate at each review stage;
3) setting a hint level at each review stage;
4) comparing the learner's review correct answer rate and the minimum review correct answer rate in each review stage, and determining whether or not to add a new item data number;
5) a step in which the hint level transitions to a higher level or a lower level according to correct / incorrect answers in the item data unit of the learner at each review stage;
A program for executing an iterative learning method comprising:   The program for executing the iterative learning method according to any one of claims 1 to 3, wherein the item data is a foreign language word or a foreign language sentence expressed in katakana, hiragana, or romaji.