JPH06289766A - Learning device - Google Patents

Abstract

PURPOSE:To determine the questions to be made so as to enhance a learning effect according to the past results of questioning or the results of the learner's answers. CONSTITUTION:A min. questioning frequency coefft. retrieving section 201 retrieves the min. value minival of the questioning frequency coefft. among questioning groups. A question to be made determining section 202 determines the question to be made in accordance with the min. value minival of the questioning frequency coefft. and the question to be made is made by a questioning section 203. The learner answers the question by operating an answering section 204. The result of the answer is decided in an answer result deciding section 205 and the questioning frequency coefft. is changed by a questioning frequency coefft. changing section 206 according to the result. The value of the questioning frequency coefft. to be changed complies with the characteristics of an index function having a negative inclination along a forgetting function.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a learning device for storing, for example, words in a foreign language, and more specifically, to improve a learning effect of a learner on a problem to be presented from a problem prepared in advance. Regarding technology for making decisions.

[0002]

2. Description of the Related Art In a device for making a learner learn a problem such as a foreign language word, the problem presenting order is conventionally a predetermined order or a random order. Either, or the problem may never be presented again when it is determined to be remembered.

Therefore, there is a problem that the level of the learner's memory is not reflected in the questions and the efficient learning cannot be performed. An object of the present invention is to make it possible to determine a question to be asked so that the learning effect is enhanced according to the past question result or the answer result of the learner.

[0004]

The present invention firstly has minimum question frequency coefficient searching means for searching the minimum value of question frequency coefficients representing the question frequencies of a plurality of questions.

Next, there is provided question setting means for determining a question based on the minimum value of the question frequency coefficient searched by the minimum question frequency coefficient searching means. Further, it has a question giving means for giving a question to the question determined by the question determining means.
This means is a means for displaying the question to be asked on the display, for example, or for notifying the learner by voice with a voice output device.

Next, the question giving means has an answering means for allowing the learner to answer the question given. This means is, for example, means for selecting a correct answer from a plurality of answers displayed on a question to be displayed on the display by using an operator.

Further, it has an answer result judging means for judging the answer result of the learner by the answer means. This means is, for example, a means for determining whether or not the answer selected by the learner using the operator is the correct answer for the plurality of answers given on the display.

In addition, according to the characteristic of an exponential function whose input includes a parameter indicating the number of times of learning or the elapsed time of the problem and a parameter indicating the correctness of the answer result of the problem judged by the answer result judging means, the slope is negative. It has a question frequency coefficient changing means for changing the question frequency coefficient of each question.

In the above-mentioned configuration of the invention, the question frequency coefficient changing means may be further provided with a changing means for changing the characteristic of the negative exponential function which follows when changing the question frequency coefficient in question.

Further, the question frequency coefficient changing means further includes, in the input, a parameter indicating the ratio of the correct answer to the question in the past and a parameter indicating the ratio of the incorrect answer to the question in the past. Can be configured to change the question frequency coefficient of the question according to the property of the negative exponential function.

Alternatively, the question frequency coefficient changing means changes the question frequency coefficient of the question according to the characteristic of the exponential function having a negative slope by further including a parameter indicating the response of the learner's answer by the answering means in the input. Can be configured to.

On the other hand, the minimum question frequency coefficient retrieving means can be configured to retrieve the minimum value of the question frequency coefficient corresponding to only the question corresponding to the designated question level among the plurality of questions.

The answering means can be configured to forcibly give a predetermined answer when the learner does not respond within a predetermined time after the questioning means has given the question. Further, the initial value of the question frequency coefficient can be configured to be determined according to each of the plurality of questions.

[0014]

The minimum question frequency coefficient searching means selects one question from the questions in the question group according to the question frequency coefficient.
The question frequency coefficient includes a parameter indicating the number of times of learning or elapsed time of the question and a parameter indicating the correctness of the answer result of the question determined by the answer result determination means by the question frequency coefficient changing means, and the slope is negative. Is modified according to the exponential property of. This property is in line with the fact that human memory for a given matter decreases exponentially over time according to a function called the forgetting function. This is the feature of the present invention.

Then, this characteristic is controlled by the parameter indicating the number of times of learning or the elapsed time of the question and the parameter indicating the correctness of the answer result of the question judged by the answer result judging means, for example, the previous question. Sometimes questions that have been answered correctly are less likely to be given after the next question, questions that have not been given in the past are more likely to be given, and questions that have been given the wrong answer during the previous question are more likely to be given after the next question. It is possible to control so that the question is easily given and iterative learning is performed, and it is possible to give the question according to the memory of the learner.

On the other hand, if this characteristic is changed by the changing means, the learner can adjust the questioning tendency. In addition, if the above-mentioned characteristics are controlled by a parameter indicating the ratio of the correct answer to the problem in the past and a parameter indicating the ratio of the incorrect answer to the problem in the past, Depending on how much the answer is correct or incorrect, the question frequency of the question after the next question can be changed.

Further, if the above-mentioned characteristics are controlled also by the response at the time of the answer of the learner, for example, the parameter indicating the touch response of the switch, the problem also arises depending on whether or not the learner was confident at the time of the answer. It is possible to change the question frequency after the next question.

On the other hand, when searching for the minimum value of the question frequency coefficient for a question, only the question corresponding to the specified question level among the plurality of questions is searched for the minimum value of the question frequency coefficient corresponding to them. Then, the learner can select a question according to his / her level.

Furthermore, if the learner does not respond within a predetermined time after the question is posed, a predetermined answer can be forcibly given, so that a time limit can be set for the answer to the posed question.

In addition, if the initial value of the question frequency coefficient is determined for each of a plurality of questions, the question tendency can be controlled according to the difficulty of the question.

[0021]

Embodiments of the present invention will be described below with reference to the drawings. First Embodiment First, a first embodiment of the present invention will be described. <Structure> The external appearance of the learning device in the first embodiment is shown in FIG.

In FIG. 1, first, the learner operates the operation unit 10.
The operation key 104 of No. 2 is operated to make the question appear on the display unit 101. In this case, for example, when an English word is to be presented, a question may be presented by voice from the speaker 103 in addition to the question being displayed on the display unit 102.

For this question, the learner operates the operation unit 102.
Operation key 104 to select an answer that seems to be correct from a predetermined number of answers displayed on display unit 101, and select YE
Press the S button 106. If the question is of the XX type, the YES button 106 or the NO button 105 is pressed.

As described above, the learner selects the YES button 106.
Alternatively, when the NO button 105 is pressed, the correct answer is displayed on the display unit 101. The question starts when the learner presses the question button 107, and when the learner's answer is given after that, the next question is automatically given. Alternatively, the learner can switch the question method switching switch 108 and press the question button 107 for each question so that the question can be given at any timing.

Next, FIG. 2 is a functional block diagram showing a functional configuration of the first embodiment according to the present invention. First, the minimum question frequency counting search unit 201 searches for the minimum value minival of the question frequency coefficient representing the question frequency of the question to be asked.

The question setting section 202 determines a question based on the minimum value minival of the question frequency coefficient, and the question section 203 issues the question. Next, the learner answers the question given by operating the answering unit 204 including the operation keys 104 and the NO button 105 or the YES button 106 in FIG. Then, the answer result of the learner is determined by the answer result determination unit 205, and the question frequency coefficient for determining future question questions is changed by the question frequency coefficient changing unit 206 according to the determination result. In this case, the question frequency coefficient of the incorrect question is reduced by the question frequency coefficient changing unit 206 to facilitate the question, and the learner can repeat the learning.

After that, the process returns to the process of the minimum question frequency coefficient search unit 201, and the question is repeated. Here, the question frequency coefficient m (i) representing the question frequency of the question of an arbitrary question number i for determining the question to be asked is defined by the following formula 1.

[0028]

## EQU00001 ## m (i) = exp (-a (i) (t (i) -K (i))) In Equation 1, t (i) is the number of past learnings of the problem of problem number i ( K (i) is a coefficient determined in accordance with whether or not the previous answer to the question of question number i was correct, or whether the question of question number i was given in the past. And a (i) is a coefficient.

The question frequency coefficient m (i) based on Equation 1 is
It is characterized by the fact that human memory for a given matter is defined exponentially (along the exp characteristic) over time according to a function called a forgetting function.

In the definition of Equation 1, when the learner answers the question of question number i correctly, for example, by setting t (i) = K (i), the question frequency is set. The value of the coefficient m (i) can be set to the maximum value of 1. If the question with question number i has not been given in the past, or if the learner has given an incorrect answer to the question with question number i, K (i) is set to a large negative number. As a result, the value of the question frequency coefficient m (i) can be set to a value close to 0. In this case, when the learner makes a wrong answer, the value of the question frequency coefficient m (i) can be set to a value closer to 0 by increasing the learning count t (i).

And, every time the question is given, the question frequency coefficient m
By selecting the question that minimizes (i), the question for which the correct answer was given at the previous question becomes difficult to be given after the next question, and the question which has not been given in the past becomes easy to be given. , It is possible to control the problem that the wrong answer was given at the previous question to make it easier to be given after the next question so that iterative learning is performed, and it is possible to give the question according to the learner's memory. .

In order to enable the above operation, in the first embodiment, the coefficient a (i), the learning number t (i), and the coefficient K (i) are stored for each problem. Control of each parameter and question control are performed by the operation described. <Operation> The operation of the first embodiment based on the above configuration will be described below with reference to the operation flowcharts of FIGS. Note that these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes a control program stored in a ROM or the like using a RAM as a work memory. Further, in the following description, a symbol indicating a variable may be described as a symbol indicating the value of the variable. For example, “the value of the variable MAXDATA that stores the total number of problems” is synonymous with “the total number of problems MAXDATA”.

First, steps S301 to S306 in FIG.
Then, the initialization process is performed. That is, step S301
After the problem number i is cleared in step S305, the problem number i is sequentially incremented in step S305,
The following process is executed by repeating the processes of steps S302 to S304 until it is determined in 306 that the problem number i has reached the total number MAXDATA of the problems.

First, a large negative number, for example, -10000 is substituted into the coefficient K (i) (step S302). This value has a role of setting the value of the question frequency coefficient m (i) to a value close to 0, and is a value corresponding to the memory level of the learner when the question of the question number i has not been given yet. .

Next, an appropriate constant P is substituted into the coefficient a (i) (step S303). Furthermore, the learning count t (i)
The value 0 is assigned to. As described above, the coefficients K (i) and a (i) corresponding to all the problems and the learning number t
Initialization of (i) is completed and the problem number i is the total number of problems MA
When XDATA is reached, the determination in step S306 becomes NO and step S307 is executed.

In step S307, until the questioning state is reached, that is, until a predetermined time elapses from the previous questioning or until the learner presses the questioning button 107, the determination is NO and the state is in the standby state. Step S30
If the determination in step 7 is YES, step S308 is executed.

In steps S308 to S311, after the problem number i is cleared in step S308, step S308 is executed.
At 310, while the problem number i is sequentially incremented,
Until it is determined in step S311 that the problem number i has reached the total number MAXDATA of problems, in step S309,
The question frequency coefficient m (i) for each question is recalculated according to the above-described equation (1). The values of the coefficients K (i), a (i), and the number of times of learning t (i) are calculated in step S301 described above.
Is initialized by the initialization process of S306, or
Since it is changed by the processing of steps S503 to S507 of FIG. 5 described later, in steps S308 to S311
The question frequency coefficient m for the new values of the above parameters
(I) is changed. In addition, this process is shown in FIG.
This corresponds to a part of the function of the question frequency coefficient changing unit 206.

Question frequency coefficient m corresponding to all problems
After the change of (i) is completed, the problem number i is the total number of problems MAXD
When ATA is reached, the determination in step S311 is NO, and step S401 in FIG. 4 is executed.

The processing of steps S401 to S408 in FIG. 4 corresponds to the function of the minimum question frequency coefficient search unit 201 in FIG. 2, and the question number mininu that minimizes the question frequency coefficient m (i).
This is the process of searching for m.

First, the value 0 corresponding to the first question number is assigned to the question number mininum of the question indicating that the value of the question frequency coefficient is the minimum (step S).
401), the question frequency coefficient m (0) corresponding to the first question is substituted as the initial value for the minimum question frequency coefficient minival (step S402).

Then, after 1 is assigned to the problem number i (step S403), the problem number i is the total number MAXDAT of the problems.
If the question number i is smaller than A and the decision in step S404 is YES, the question number i has the smallest value of the question frequency coefficient by repeating the processing in steps S404 to S408 while sequentially incrementing the question number i in step S408.
The mininum and its minimum value minival are calculated as follows.

First, the question frequency coefficient m of the question of question number i
It is determined whether the value of (i) is smaller than the minimum value minival up to the present (step S405). The judgment is N
If O, step S408 for the next question number.
If the determination is YES, the variable mininum
Is assigned to the question number i (step S406), and the value of the question frequency coefficient m (i) is assigned to the minimum value minival up to the present (step S407).

Subsequently, after the problem number i is incremented (step S408), the process returns to step S404. In this way, while the question number i is sequentially incremented, the minimum question frequency coefficient minival and its question number mininum are calculated, and the question number i is the total number of questions.
When it becomes equal to MAXDATA, the determination in step S404 becomes N.
When it becomes O, step S409 is executed.

The processing of step S409 corresponds to the function of the question setting section 202 of FIG. 2, and in this processing,
The question number mininum of the question determined to have the smallest question frequency coefficient is the variable q_nu that indicates the question number to be asked.
It is assigned to mber.

The processing of the next step S410 corresponds to the function of the questioning unit 203 of FIG. 2, and in this processing, the question S (q_numbe) in which the question number is equal to the value of the variable q_number.
r) is posed and is displayed on the display unit 101 of FIG. 1, or is uttered by voice from the speaker 103 of FIG.

In response to the question, the learner operates the operation key 104 of the operation unit 102 to select one of the predetermined number of answers displayed on the display unit 101 which seems to be correct and YES.
Press the button 106. Also, if the question is ○ × formula,
The learner pushes the YES button 106 or the NO button 105. These operation functions correspond to the functions of the answer unit 204 in FIG. Then, in step S501 of FIG. 5 following step S410 of FIG. 4, the determination is NO until such a response by the learner is made, and the learner waits for the response.

When the learner responds to the question, the determination in step S501 becomes YES and step S502 is executed. In step S502, it is determined whether the learner's answer is correct. This processing corresponds to the function of the answer result determination unit 205 in FIG.

Depending on whether the learner's answer is correct,
Steps S503 to S507 or Steps S505 to
Any processing group of S507 is executed. These processes correspond to some functions of the question frequency coefficient changing unit 206 in FIG.

When the learner's answer is correct and the determination in step S502 is YES, first in step S503,
A value obtained by adding 1 to the past learning number t (q_number) of the question is substituted into the coefficient K (q_number) corresponding to the question given this time, which has the question number of the value of the variable q_number. Next, in step S504, a value F (a) = F (a (is obtained by inputting the coefficient a (q_number) corresponding to the question given this time, into the predetermined function F. q_number)) is substituted. Then, in step S507, the learning number t (q_number) of the question given this time is incremented by one.

After the above processing, the process returns to step 307 in FIG. 3 and the next question is repeated. Thus, if the learner's answer to the given question is correct,
When the next question is decided, in the question that the question number is equal to the value of the variable q_number, t (q_numbe
As a result of r) = K (q_number), the value of the question frequency coefficient m (q_number) corresponding to the question becomes the maximum value 1 from the expression (1). Therefore, the question will be difficult to be asked after the next time. This indicates that the learner remembers the correct answer for the problem.

On the other hand, when the learner's answer is incorrect and the determination in step S502 in FIG. 5 is NO, first, in step S505, the question addressed this time having the question number of the value of the variable q_number is dealt with. Coefficient K (q_numbe
A large number-10000 is assigned to r). Next, in step S506, the constant P is substituted into the coefficient a (q_number) corresponding to the question given this time. Then, in step S507, the number of learning times t of the question given this time is t.
(Q_number) is incremented by 1.

After the above processing, the process returns to step 307 in FIG. 3 and the next question is repeated. In this way, if the learner's answer to the given question is incorrect, the question number will be the variable q_number when the next question is decided.
In that question that is equal to the value of, the coefficient K (q
_Number) and a (q_number) are substituted with initial values (see steps S302 and S303 in FIG. 3) when the question is not yet given, and the learning count t (q_nu
mber) is incremented, and the value of the question frequency coefficient m (q_number) corresponding to the problem is
It is smaller than the value when the problem has not been asked yet. Therefore, the question whose answer is incorrect becomes easy to be given after the next question. As for the problem in which the incorrect answer is repeated, the learning frequency t (q_number) increases, and as a result, the question frequency coefficient m corresponding to the problem is calculated from the equation (1).
The value of (q_number) is even closer to 0, which makes it easier to set questions. This results in the learner performing iterative learning.

Here, for the problem of an arbitrary problem number i that has not been asked in the past, the coefficients K (q_number) and a
The value of (q_number) remains the initial value, and according to the formula 1, the value of the question frequency coefficient m (i) becomes a value close to 0, though not so much as an incorrect question. Therefore, a problem that has not been asked in the past is relatively easy to be given after the next question. Second Embodiment Next, a second embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, their description will be omitted. In the second embodiment, an arbitrary question number i is used to determine the question to be asked.
The question frequency coefficient m (i) representing the question frequency of the question is
In place of the formula 1 in the case of the embodiment, the following formula 2 is defined.

[0055]

## EQU00002 ## m (i) = exp (-a (i) (T-K (i))) In Expression 2, T represents the elapsed time from the start of learning, and K (i) and a (i) is the same as that in the equation 1 of the first embodiment described above.

In the definition of the equation (2), when the learner correctly answers the question of the question number i, for example, by setting T = K (i), the case of the above equation (1) Similar to, the value of the question frequency coefficient m (i) is almost the maximum value 1
Can be set to. If the question with question number i has not been given in the past, or if the learner has given an incorrect answer to the question with question number i, K (i) is set to a large negative number. As a result, the value of the question frequency coefficient m (i) can be set to a value close to 0.

On the other hand, the characteristic of the equation 2 is different from the characteristic of the equation 1 in the following points. That is, in the equation 1, in the problem in which the learner gave an incorrect answer, the learning count t (i)
Is increased, the value of the question frequency coefficient m (i) becomes close to 0, and the question frequency coefficient m (i) is set to be easy to be given after the next question. On the other hand, in Equation 2, the longer the elapsed time from the start of learning, the easier it is to set a question that has not been asked in the past or a question that the learner answered incorrectly. This corresponds to the fact that the learner's memory decreases exponentially over time.

In order to enable the above operation, in the second embodiment, the above-mentioned coefficient a (i) and coefficient K (i) are stored for each problem, and the elapsed time T is updated sequentially. By the operation described below, control of each parameter and question control are performed.

6 to 8 are operation flowcharts of the second embodiment. These operation flowcharts are
As in the case of the above embodiment, a microcomputer not particularly shown is realized as an operation of executing a control program stored in a ROM or the like, using the RAM as a work memory.

First, steps S601 to S60 in FIG.
3, steps S605 to S608, S610, and S611 are steps S301 to S301 of FIG. 3 in the first embodiment.
S303, S305 to S308, S310, and S31
The process is the same as the process No. 1. 6 differs from FIG. 3 in that
The point is that the elapsed time T is initialized to 0 in step S604, and the question frequency coefficient m (i) is changed in step S609 based on the above-described equation 2.

Next, steps S701 to S710 of FIG.
Are steps S401 to S401 of FIG. 4 in the first embodiment.
The process is exactly the same as 410. That is, the question frequency coefficient m
The problem that (i) is the minimum is searched and the problem is presented, which is the same in both the first and second embodiments.

Further, steps S801 and S80 of FIG.
2, S804, S805, and S806 are steps S501, S502, and S50 of FIG. 5 in the first embodiment.
This is the same processing as 4, S505, and S506. The difference between FIG. 8 and FIG. 5 is that, in step S803, when the learner's answer to the question given is correct,
The current elapsed time T is substituted into the coefficient K (q_number) corresponding to the question given this time, which has the question number of the value of the variable q_number, and step S5 of FIG.
The point that the process of 07 is not executed.

As a result, if the learner's answer to the question given is correct, the question number is equal to the value of the variable q_number when the question is decided next time, T≈K ( q_number),
From the equation (2), the question frequency coefficient m (q_
The value of (number) is almost the maximum value 1. Therefore, the problem is less likely to be given after the next time the question is given, as in the case of the first embodiment.

On the other hand, the longer the elapsed time T from the start of learning, the more frequently the question frequency coefficient m (i As a result of the value of) approaching 0, it becomes easier for questions to be asked.

Finally, FIG. 9 is an operation flowchart of the timer interrupt process for counting the elapsed time T. The process of step S901 of this operation flowchart is executed at regular time intervals based on an interrupt from a hardware timer (not shown), and the value of the elapsed time T, which is a variable cleared in step S604 of FIG. , +1 is incremented. Third Embodiment Next, a third embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, their explanation will be omitted. In the third embodiment, an arbitrary question number i is used to determine the question to be asked.
The question frequency coefficient m (i) representing the question frequency of the question is
This is defined by the same equation 1 as in the case of the embodiment.

The third embodiment is different from the first embodiment in that the coefficient a (i) in the equation (1) is changed when the learner's answer to the given question is correct. This is the point that the learner can freely change the function F. This enables the learner to adjust the questioning tendency.

The operation of the third embodiment for enabling the above operation is shown by the operation flowcharts of FIGS. Similar to the case of the first embodiment and the like, these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes the control program stored in the ROM or the like using the RAM as the work memory.

First, steps S1001 to S1 in FIG.
The processing of 006 and S1009 to S1013 is the same as the processing of steps S301 to S306 and S307 to S311 of FIG. 3 in the first embodiment. Figure 10
Is different from FIG. 3 in that the determination in step S1009 is NO.
In the waiting state for the question, the learner operates, for example, the question method switching switch 108 in FIG. 1 to determine YES in step S1007, and as a result, the coefficient a (i) for changing the coefficient a (i) in step S1008. This is that the characteristic of the function F (a) = F (a (i)) can be changed.

Next, steps S1101 to S1 in FIG.
110 is step S40 of FIG. 4 in the first embodiment.
The process is exactly the same as 1 to S410. That is, the point that the question having the smallest question frequency coefficient m (i) is searched and the question is given is the same in both the first and third embodiments.

Further, steps S1201 to S1 of FIG.
Reference numeral 207 denotes step S50 of FIG. 5 in the first embodiment.
The process is exactly the same as 1 to S507. However, step S
The coefficient a (q_number) changed in 1204 is
It is changed by the function F (a) that can be changed in step S1008 of FIG. 10 described above. As a result, FIG.
The characteristic of the question frequency coefficient m (i) changed in step S1011 of step S1011, that is, the question tendency of the question can be changed. Fourth Embodiment Next, a fourth embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, their description will be omitted. In the fourth embodiment, an arbitrary question number i is used to determine the question to be asked.
The question frequency coefficient m (i) representing the question frequency of the question is
This is defined by the same equation 1 as in the case of the embodiment.

The fourth embodiment is different from the first embodiment in that the coefficient a (i) in the equation 1 is changed when the learner's answer to the question is correct. This is that the input parameter of the function F includes not only the value of a (i) itself but also the number of correct answers and the number of incorrect answers of the past problem. This makes it possible to change the question frequency of the question after the next question is given, depending on how much the question has been correct or incorrect in the past.

The operation of the fourth embodiment for enabling the above operation is shown by the operation flowcharts of FIGS. Similar to the case of the first embodiment and the like, these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes the control program stored in the ROM or the like using the RAM as the work memory.

First, steps S1301 to S1 in FIG.
The processes of 304 and S1307 to S1313 are the same as the processes of steps S301 to S304 and S305 to S311 of FIG. 3 in the first embodiment. FIG.
Is different from FIG. 3 in that steps S1305 and S130 are
6 is that the past incorrect answer number d (i) and correct answer number c (i) of each problem are both initialized to 0.

Next, steps S1401 to S1 in FIG.
410 is step S40 of FIG. 4 in the first embodiment.
The process is exactly the same as 1 to S410. That is, the point that the question having the smallest question frequency coefficient m (i) is searched and the question is given is the same in both the first and fourth embodiments.

Further, steps S1501 to S1 of FIG.
503, S1506, S1508, and S1509,
Steps S501 to S50 of FIG. 5 in the first embodiment.
This is the same processing as 3, S505, S506, and S507. 15 is different from FIG. 5 in the following points. That is, first, if the learner's answer to the question given is correct, then in step S1504, the past correct answer number c (q_number) corresponding to the question given this time, which has the question number of the value of the variable q_number, is +1. To be done. In step S1505, the coefficient a corresponding to the question given this time
In (q_number), the three parameters of the coefficient itself a (q_number), the past correct answer number c (q_number) of the problem, and the past incorrect answer number d (q_number) of the problem are input to a predetermined function F. Value F (a, c, d) obtained by
= F (a (q_number), c (q_number), d (q_
number)) is substituted. Next, if the learner's answer to the question given is incorrect, step S15.
At 07, the number of incorrect answers d (q_numbe in the past corresponding to the question given this time, which has the question number of the value of the variable q_number
r) is incremented by 1.

As a result of the above processing, step S1 in FIG.
The characteristic of the question frequency coefficient m (i) changed in 311 or the question tendency of the question can be changed according to the past correct and incorrect answers of each question. Fifth Embodiment Next, a fifth embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, their explanation will be omitted. In the fifth embodiment, when a learner selects a mode called a level mode, when a question having a minimum question frequency coefficient m (i) is searched for, a learner selects from only questions of a preset level. This is different from the case of the first embodiment.

Therefore, in the fifth embodiment, the coefficient a (i) for defining the equation 1 and the learning number t are set for each problem.
In addition to (i) and the coefficient K (i), a level value L (i) that defines the level of each problem is stored, and control of each parameter and question control are performed by the operation described below.

The operation of the fifth embodiment for enabling the above operation is shown by the operation flowcharts of FIGS. Similar to the case of the first embodiment and the like, these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes the control program stored in the ROM or the like using the RAM as the work memory.

First, steps S1601 to S1 in FIG.
The processing of 606 and S1608 to S1612 is the same as the processing of steps S301 to S306 and S307 to S311 of FIG. 3 in the first embodiment. FIG.
Is different from FIG. 3 in that the processing for each mode is executed in step S1607.

Now, when the learner selects a mode called the normal mode by using, for example, the question method switching switch 108 in FIG. 1, the same question operation as in the case of the first embodiment is executed. That is, step S1 of FIG.
The processing of 608 to S1612 is the same as in FIG. 3 in the first embodiment.
17 is exactly the same as the processing of steps S307 to S311 of FIG. 17, and the processing of steps S1701 to S710 of FIG. 17 is steps S401 to S410 of FIG. 4 in the first embodiment.
The processing is exactly the same as that of step S1801 in FIG.
The processing of steps S1807 to S1807 is exactly the same as the processing of steps S501 to S507 of FIG. 5 in the first embodiment.

On the other hand, when the learner selects the mode called the level mode by using, for example, the question method switching switch 108 in FIG. 1, step S1901 in FIG.
The subsequent processing is executed.

First, when the learner inputs a level value with, for example, the operation key 104 of FIG. 1 in the level input standby state in which the processing of step S1607 of FIG. 16 and the processing of step S1901 of FIG. 19 are repeated, step S
The determination result in 1901 is YES, and step S1902 is executed.

In step S1902, the level value set in the variable L is substituted. Next, step S1903.
After the problem number i and the corresponding level problem number j are cleared in step S1907, the problem number i is sequentially incremented until step S1908 determines that the problem number i has reached the total number MAXDATA of problems. The following processing is executed by repeating the processing of S1904 to S1908.

First, it is determined whether the level value L (i) of the problem number i is equal to the set level value L (step S).
1904). If L (i) = L, then the next problem number i
The process of step S1907 for is executed.

If L (i) = L, then array N
The problem number i is substituted into (j), and the correspondence level problem number j
Is incremented and step S1907 is executed. When the question number i reaches the total number MAXDATA of questions by repeating the above-described processing, the determination in step S1308 is NO, and step S1909 is executed. By the processing up to this point, the question number corresponding to the set level value L is obtained in the array N (j).

Next, steps S1909 to S1 of FIG.
The basic function of the processing of 912 is the same as the function of the processing of steps S307 to S311 of FIG. 3 in the first embodiment. However, in FIG. 19, the question frequency coefficient is changed not for an arbitrary question number i but only for the question number N (i) whose level value is equal to the set level value. Different from the case. In this case, step S191
The total number of problems judged in 2 is j, not MAXDATA.

Steps S2001 to S2010 of FIG.
The basic function of the processing of (1) is the same as the function of the processing of steps S401 to S410 of FIG. 4 in the first embodiment. However, in FIG. 20, the search for the minimum question frequency coefficient in step S2005 is not for the question frequency coefficient m (i) of the question of an arbitrary question number i, but for the question number N (whose level value is equal to the set level value). 4 is different from the case of FIG. 4 in that it is performed on the question frequency coefficient m (N (i)) of the problem of i). In this case, in step S2006, the variable mi
The question number assigned to ninum is N (i) instead of i. Also, in step S2007, the variable minival
The question frequency coefficient substituted into m (N) is not m (i).
(I)).

Steps S2101 to S2107 of FIG.
Are steps S501 to S501 of FIG. 5 in the first embodiment.
It is exactly the same as 507. However, in FIG. 21, as step S2108, the process returns to step S1607 of FIG. 16 or step S190 of FIG. 19 depending on whether the current mode is the normal mode or the level mode.
Whether to return to the process of 9 is branched. Sixth Embodiment Next, a sixth embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, their description will be omitted. In the sixth embodiment, an arbitrary question number i is used to determine the question to be asked.
The question frequency coefficient m (i) representing the question frequency of the question is
This is defined by the same equation 1 as in the case of the embodiment. Further, as in the case of the fourth embodiment, when the learner's answer to the question given is correct, the input parameter of the function F for changing the coefficient a (i) in the equation 1 is Not only the value of a (i) itself, but also the number of correct answers and the number of incorrect answers of the problem in the past are included. The sixth embodiment is different from the fourth embodiment in that if the learner's answer to the question given is correct, the function F for changing the coefficient a (i) in the formula 1 is changed. This is that the input parameter further includes information on the strength of touch (touch response) when the learner presses the YES button 106 or the NO button 105 in FIG. Generally, the learner tends to push the answer button strongly when he / she is confident in the answer and weakly push the answer button when he / she is not confident. Therefore, by reflecting the strength of touch in the function F, not only how well the question was correct or incorrect in the past, but also whether the learner was confident at the time of this answer, It is possible to change the question frequency after the question is given.

The operation of the sixth embodiment for enabling the above operation is shown by the operation flowcharts of FIGS. Similar to the case of the first embodiment and the like, these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes the control program stored in the ROM or the like using the RAM as the work memory.

First, steps S2201 to S2 of FIG.
The processing of 213 is the same as the processing of steps S13301 to S1313 of FIG. 13 in the fourth embodiment.
The processing of steps S2301 to S2310 of FIG. 23 is also the same as that of steps S1401 to S1 of FIG. 14 in the fourth embodiment.
The process is the same as the process of 410.

Then, steps S2401 and S in FIG.
The processing of 2403 to S2405 and S2407 to S2410 is performed by S1501 and S15 of FIG. 15 in the fourth embodiment.
The processing is the same as the processing of 02 to S1504 and S1506 to S1509. 24 differs from FIG. 15 in the following points. That is, first, the learner gives an answer and the step S
If the determination in 2401 is YES, step S2
In 402, the touch response data indicating the touch of these buttons, which is output from these buttons when the learner presses the YES button 106 or the NO button 105 in FIG. 1, is substituted into the variable TR. Then, in step S2406, the coefficient a (q_numbe) corresponding to the question given this time is added to the coefficient a (q_numbe).
r), the number of correct answers c (q_number) in the past of the problem, the number of incorrect answers d (q_number) in the past of the problem, and a value obtained by inputting three parameters of the touch response data TR into a predetermined function F. F (a, c, d, TR) = F
(A (q_number), c (q_number), d (q_numb
er) and TR) are substituted.

As a result of the above processing, step S1 in FIG.
In addition to the characteristics of the question frequency coefficient m (i) changed in 311, that is, the question tendency of the question, in addition to the past correct and incorrect answers of each question, whether the learner was confident at this answer It can be changed accordingly. Seventh Embodiment Finally, a seventh embodiment of the present invention will be described.

Since the external appearance and the functional configuration of this embodiment are the same as those shown in FIGS. 1 and 2 in the first embodiment, the description thereof will be omitted. In the seventh embodiment, when the learner selects a mode called the auto mode, after the question is posed, even if the learner does not answer, a predetermined time elapses and the answer of NO is automatically and forcibly obtained. This is different from the case of the first embodiment in that a time limit can be set for the answer to the question given by doing.

The operation of the seventh embodiment for enabling the above operation is shown by the operation flowcharts of FIGS. 25 to 30. Similar to the case of the first embodiment and the like, these operation flowcharts are realized as an operation in which a microcomputer (not shown) executes the control program stored in the ROM or the like using the RAM as the work memory.

First, steps S2501 to S2 in FIG.
The processing of 506 and S2509 to S2513 is the same as the processing of steps S301 to S306 and S307 to S311 of FIG. 3 in the first embodiment. Figure 25
3 is different from FIG. 3 in that processing for each mode is executed in step S2507, and in the normal mode, in order to stop the timer interrupt operation of FIG.
That is the point.

Now, when the learner selects a mode called the normal mode by using, for example, the question method switching switch 108 of FIG. 1, the same question operation as in the case of the first embodiment is executed. That is, step S2 of FIG.
The processing of steps 509 to S2513 is shown in FIG.
26 is exactly the same as the processing of steps S307 to S311 of FIG. 26, and the processing of steps S2601 to S2610 of FIG. 26 is the same as steps S401 to S41 of FIG. 4 in the first embodiment.
It is exactly the same as the processing of 0, and step S2611
Is YES, and steps S2701 to S2701 in FIG.
The processing of S2707 is exactly the same as the processing of steps S501 to S507 of FIG. 5 in the first embodiment.

On the other hand, when the learner selects a mode called the auto mode by using, for example, the question method switching switch 108 in FIG. 1, step S2801 in FIG.
The subsequent processing is executed.

First, after the question start waiting state in which the processing of step S2801 of FIG. 28 and the processing of step S2507 of FIG. 25 are repeated, steps S2802 to S280 are performed.
5, steps S2601 to S2610 of FIG. 26 → S26
A series of processes of 11 → S2806 is executed. These series of processes are performed in steps S308 to S311 of FIG. 3 and steps S401 to S of FIG. 4 in the first embodiment.
This is the same processing as the series of processing consisting of 410, and the question having the minimum question frequency coefficient m (i) is searched and the question is given.

When the question is given in this way, first,
After the answer waiting time TT is cleared in step S2806, a value 1 for validating the operation of the timer interrupt routine is assigned to the timer flag TF in step S2807. Then, by repeating steps S2808 and S2809, it is determined whether there is a response from the learner or whether the answer waiting time TT has reached the time limit time k. The timer flag TF is sequentially counted up by the timer interrupt processing of FIG. 30 described later.

If the learner responds within the time limit time, the determination in step S2809 becomes YES, and step S28 depends on whether the response is YES or NO.
After branching in step 10, step S2901 in FIG. 29 is executed. The subsequent processing of steps S2901 to S2904 is exactly the same as the processing of S502 to S507 of FIG. 5 in the first embodiment.

On the other hand, if the time limit time elapses without the learner responding, the determination in step S2808 is YES.
It becomes S, and a NO answer is forcibly made and step S290
1 is executed. After that, the above-mentioned steps S2901 to S2901
The process of S2904 is executed.

Finally, in step S2907, depending on whether the current mode is the normal mode or the auto mode, the process returns to step S2507 in FIG.
Whether the process returns to step S2802 of FIG. 28 is branched.

Finally, FIG. 30 is an operation flowchart of the timer interrupt process for counting the answer waiting time TT. The process of this operation flowchart is executed at regular time intervals based on an interrupt from a hardware timer (not shown). And step S
Only when the value of the timer flag TF is 1 in 3001, that is, only when the answer waiting state is entered in the auto mode, the answer waiting time TT is incremented by +1 in step S3002 (step S280 in FIG. 28).
7, see step S2508 in FIG. 25).

As described above, the time limit can be set for the answer to the question given. Other Embodiments In the embodiments described above, each parameter for changing the question frequency coefficient m (i) is set to the same initial value for all questions at the start of learning, but depending on the difficulty level of the questions, etc. The initial value may be changed for each problem.

Further, in the exponential function type mathematical formula for determining the question frequency coefficient m (i), various forms of the exponent part can be applied.

[0110]

According to the present invention, the characteristic of an exponential function having a negative slope when determining a question frequency coefficient indicates a parameter indicating the number of times of learning or elapsed time of a question and the degree of correctness of the answer result of the question. By controlling by the parameter and, for example, a question for which the correct answer was given at the time of the previous question becomes difficult to be given after the next question, and a question which has not been given in the past becomes easy to be given, and it is erroneous at the time of the last question. The question that has been answered can be controlled so that it will be more likely to be given after the next question and iterative learning will be performed.
It is possible to set the question according to the learner's memory.

On the other hand, if this characteristic is changed by the changing means, the learner can adjust the questioning tendency. In addition, if the above-mentioned characteristics are controlled by a parameter indicating the ratio of the correct answer to the problem in the past and a parameter indicating the ratio of the incorrect answer to the problem in the past, Depending on how much the answer is correct or incorrect, the question frequency of the question after the next question can be changed.

Further, if the above-mentioned characteristics are controlled also by the response at the time of the answer of the learner, for example, the parameter indicating the touch response of the switch, the problem may also occur depending on whether or not the learner was confident at the time of the answer. It is possible to change the question frequency after the next question. On the other hand, when searching for the minimum value of the question frequency coefficient for a question, if the search is performed for the minimum value of the question frequency coefficient corresponding to only the question corresponding to the designated question level among the plurality of questions, The learner can select a problem according to his or her level.

Furthermore, if the learner does not respond within a predetermined time after the question is given, a predetermined answer can be forcibly given, and a time limit can be set for the answer to the question given. Become.

In addition, if the initial value of the question frequency coefficient is determined in accordance with each of a plurality of questions, it becomes possible to control the question tendency according to the difficulty of the question.

[Brief description of drawings]

FIG. 1 is an external view of a learning device according to the present invention.

FIG. 2 is a functional block diagram showing a functional configuration of the first embodiment.

FIG. 3 is an operation flowchart of the first embodiment (part 1).
Is.

FIG. 4 is an operation flowchart of the first embodiment (part 2).
Is.

FIG. 5 is an operation flowchart of the first embodiment (part 3).
Is.

FIG. 6 is an operation flowchart of the second embodiment (part 1).
Is.

FIG. 7 is an operation flowchart of the second embodiment (part 2).
Is.

FIG. 8 is a flowchart of the operation of the second embodiment (No. 3).
Is.

FIG. 9 is an operation flowchart of the second embodiment (part 4).
Is.

FIG. 10 is an operation flowchart (No. 1) of the third embodiment.

FIG. 11 is an operation flowchart (2) of the third embodiment.

FIG. 12 is an operation flowchart (Part 3) of the third embodiment.

FIG. 13 is an operation flowchart (part 1) of the fourth embodiment.

FIG. 14 is an operation flowchart (part 2) of the fourth embodiment.

FIG. 15 is an operation flowchart (part 3) of the fourth embodiment.

FIG. 16 is an operation flowchart (No. 1) of the fifth embodiment.

FIG. 17 is an operation flowchart (No. 2) of the fifth embodiment.

FIG. 18 is an operation flowchart (No. 3) of the fifth embodiment.

FIG. 19 is an operational flowchart (Part 4) of the fifth embodiment.

FIG. 20 is an operational flowchart (No. 5) of the fifth embodiment.

FIG. 21 is an operation flowchart (6) of the fifth embodiment.

FIG. 22 is an operation flowchart (No. 1) of the sixth embodiment.

FIG. 23 is an operation flowchart (No. 2) of the sixth embodiment.

FIG. 24 is an operational flowchart (No. 3) of the sixth embodiment.

FIG. 25 is an operational flowchart (No. 1) of the seventh embodiment.

FIG. 26 is an operational flowchart (No. 2) of the seventh embodiment.

FIG. 27 is an operational flowchart (No. 3) of the seventh embodiment.

FIG. 28 is an operation flowchart (No. 4) of the seventh embodiment.

FIG. 29 is an operational flowchart (No. 5) of the seventh embodiment.

FIG. 30 is an operational flowchart (6) of the seventh embodiment.

[Explanation of symbols]

 101 display unit 102 operation unit 103 speaker 104 operation key 105 NO button 106 YES button 107 questioning button 108 questioning method changeover switch 201 minimum questioning frequency coefficient searching unit 202 questioning question determining unit 203 questioning unit 204 answering unit 205 answer result judging unit 206 questioning Frequency coefficient changing section

Claims (8)

[Claims] 1. A minimum question frequency coefficient retrieving means for retrieving a minimum value of a question frequency coefficient representing a question frequency of each of a plurality of questions, and a minimum question frequency coefficient retrieved by the minimum question frequency coefficient retrieving means. A question question determining means for determining a question question by means of a question, a question questioning means for questioning the question question decided by the question question determining means, and an answering means for causing a learner to answer the question given by the question questioning means, An answer result judging means for judging the answer result of the learner by the answering means, a parameter indicating the number of times of learning of the question, and a parameter indicating a correctness degree of the answer result of the question judged by the answer result judging means. And a question frequency coefficient changing means for changing the question frequency coefficient of each question according to the characteristics of the exponential function having a negative slope and being included in the input, and a learning device. 2. A minimum question frequency coefficient retrieving means for retrieving a minimum value of a question frequency coefficient representing a question frequency of each of a plurality of problems, and a minimum question frequency coefficient retrieved by the minimum question frequency coefficient retrieving means. A question question determining means for determining a question question by means of a question, a question questioning means for questioning the question question decided by the question question determining means, and an answering means for causing a learner to answer the question given by the question questioning means, An answer result determining means for determining the answer result of the learner by the answering means, a parameter indicating the elapsed time of the question, and a parameter indicating the degree of correctness of the answer result of the question determined by the answer result determining means. And a question frequency coefficient changing means for changing the question frequency coefficient of the problem in accordance with the characteristic of an exponential function having a negative slope. 3. The changing means for changing the question frequency coefficient further comprises a changing means for changing a characteristic of the negative exponential function when the question frequency coefficient of the question is changed. The learning device according to any one of claims. 4. The question frequency coefficient changing means further inputs a parameter indicating a ratio of an answer to the question being a correct answer in the past and a parameter indicating a ratio of an answer to the question being an incorrect answer in the past. 4. The learning device according to claim 1, wherein the question frequency coefficient of the question is changed according to the characteristic of the exponential function having a negative slope. 5. The question frequency coefficient changing means further includes, as an input, a parameter indicating a response at the time of the learner's answer by the answering means, and the question frequency coefficient of the question according to the characteristic of the exponential function having a negative slope. The learning device according to claim 1, wherein the learning device is changed. 6. The minimum question frequency coefficient searching means searches only the minimum value of the question frequency coefficient corresponding to only questions corresponding to a specified question level among the plurality of questions. The learning device according to any one of claims 1 to 5. 7. The answering means forcibly makes a predetermined answer when the learner does not respond within a predetermined time after the questioning means has given the question. The learning device according to any one of 1 to 6. 8. The learning device according to claim 1, wherein an initial value of the question frequency coefficient is determined according to each of the plurality of questions.