JP2015138419A - Efficient learning time allocation device and learning time allocation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a learning time allocation device and a learning time allocation program capable of optimally allocating learning time before an actual test date on the basis of test subjects and a learner's characteristics so that the learner can achieve the best result in the test.SOLUTION: Provided is a learning time allocation device which comprises the steps of: calculating learning time effective for improving academic performance on the basis of information about a learner's learning time and test subjects; creating a learning curve suited for the learner and the test subjects by changing a predetermined standard learning curve; calculating expected scores for each test subject by putting the learning time effective for improving academic performance for each test subject in the learning curve; and calculating an expected highest total score at which the total of expected scores for each test subject is the highest and presenting to the learner a learning time allocation for each test subject for achieving the expected highest total score as the best allocation of learning time.

Description

  The present invention relates to a learning time distribution apparatus and a learning time distribution program for presenting a learning time distribution with the highest learning effect according to the characteristics of a test subject and a learner.

  In conventional learning, each learner predicts based on his / her own intuition about how to allocate the limited amount of study time until the exam to each subject to increase the score most efficiently. The allocation of learning time was decided. As a result, many learners do inefficient learning, such as allocating learning time to subjects that do not efficiently lead to an increase in scores, or not allocating learning time to subjects that can expect an increase in scores, A great deal of time has been wasted.

  Therefore, a technique is known in which the learnable time is calculated based on the content of the learner's behavior per week, and a suitable learning plan is guided to the learner based on the learnable time and the period until the test. (For example, refer to Patent Document 1).

  However, in the technique shown in Patent Document 1, the type of examination is determined in advance, such as a specific civil servant examination or national examination, and the course for the learner is also determined in advance. Could not be applied to various examinations such as entrance examination, qualification examination, national examination or certification examination

JP 2009-47804 A

  The present invention has been made to solve the above problem, and based on the test subject and the learning time of the learner, the expected score or the expected deviation value that can be obtained by the learner in this test is calculated. It is an object of the present invention to provide a learning time distribution device and a learning time distribution program for presenting the learner with an optimal distribution of learning time up to the test date so that the results of the test are the highest.

  In order to achieve the above object, the present invention calculates a storage unit for storing data, an input unit to which data is input, and an optimal allocation result of time spent by the learner on learning of test subjects based on the data The data stored in the storage unit or the data input to the input unit is expected to be spent by the learner to improve the grades of the test subjects by the final examination. Numerical value of the learning time, the current stage deviation value or current stage score that the learner has acquired in the main examination or the mock test in the past, and the difficulty in improving the score or deviation value expected to be obtained for the test subject It is the learning ease, and the control unit calculates a learning curve with the learning time as a variable, and based on the inverse function of the learning curve and the current stage deviation value or the current stage score, the test subject of the learner at the current stage Learning for By calculating a numerical value indicating the degree of progress as a learning achievement and substituting the learning time into a learning curve that combines the learning ease and the learning achievement, the score or deviation value that the learner can acquire for the test subject in this test is obtained. The expected score or expected deviation value is calculated, and the expected score or expected deviation value is stored in the storage unit.

  Further, according to the present invention, in the above-described improved invention, when there are a plurality of test subjects, when the learner learns one test subject, the correlation information affects the learning of the other test subject. Is stored in the storage unit as a learning effect correlation matrix, the learning time allocated to each test subject is set as the learning time per subject for the test subject, and the learning time per subject is set in increments of one or more parameters. The value obtained by subtracting the grade maintenance learning time necessary for the learner to maintain the current score or deviation value for the test subject is calculated as the effective learning time for each test subject. By multiplying the effective learning time for each test subject by the learning effect correlation matrix, the effect learning time for each test subject is calculated, and the effect learning is performed on a learning curve that combines the ease of learning and learning achievement. By substituting time, the expected score or expected deviation value for each test subject is calculated, and the total expected score or expected deviation value for all test subjects is calculated and distributed to the test subjects according to the parameter increments. Calculate the total of expected scores or expected deviation values for all test subjects for every pattern of learning time for each subject, and the expected score with the largest value among the calculated expected scores or expected deviation values Alternatively, the sum of the expected deviation values is stored in the storage unit as the highest expected total score or the highest expected deviation value, and the learning time allocation result for each test subject corresponding to the highest expected total score or the highest expected total deviation value is the optimum learning time. The distribution is stored in the storage unit.

  Using a computer having a storage unit for storing data, an input unit for inputting data, and a control unit for processing the data, a learner calculates an optimal distribution result of time spent for learning the test subjects In the learning time distribution program, the data stored in the storage unit or the data input to the input unit is the learning time that the learner is expected to spend for the improvement of the test subjects by the final examination, and the learner This is the degree of ease of learning that shows the current stage deviation value or current stage score obtained in the main examination or mock test in the past and the difficulty of improving the score or deviation value expected to be obtained for the test subject in numerical values, and is controlled The learning curve with the learning time as a variable, the inverse function of the learning curve, and the current stage deviation value or the current stage score. Scores that learners can acquire for test subjects in this exam by substituting learning time into a learning curve that combines learning progress and learning achievement, and a step that calculates a numerical value indicating the progress of learning as learning achievement Alternatively, the computer causes the computer to execute a step of calculating the deviation value as an expected score or an expected deviation value and a step of storing the expected score or the expected deviation value in the storage unit.

  Further, in the above-described improved invention, the present invention relates to correlation information that affects learning of another test subject when a learner learns one test subject when there are a plurality of test subjects in the control unit. Is stored in the storage unit as a learning effect correlation matrix, and the learning time allocated to each test subject is set as the learning time per subject for the test subject, and the learning time per subject is determined by one or more parameters. Effective learning is performed for each test subject by subtracting the grade change learning time required for the learner to maintain the current score or deviation value for the test subject from the step to change in steps of each subject and the study time for each subject. A step of calculating time, a step of calculating an effect learning time for each test subject by multiplying an effective learning time for each test subject by a learning effect correlation matrix; By substituting the effect learning time into the learning curve that combines ease and learning achievement, the expected score or expected deviation value for each test subject is calculated, and the total expected score or expected deviation value for all test subjects is calculated. A step of calculating, and a step of calculating a total of expected scores or deviation values for all test subjects for every pattern of learning time for each subject allocated to the test subjects according to the parameter increments. Storing the highest expected total score or expected deviation value in the storage unit as the highest expected total score or highest expected deviation value, and the highest expected total score or highest Storing a learning time distribution result for each test subject corresponding to the expected total deviation value in a storage unit as an optimal learning time distribution, and a computer To be executed.

  The learning time distribution apparatus of the present invention can calculate a learning curve corresponding to a test subject, and can also calculate the learning ease of the test subject and the learner's learning achievement level. The learning time distribution device can calculate the expected score or the expected deviation value that can be obtained by the learner in the main examination with high accuracy based on the learning curve, the learning ease, the learning achievement degree, and the effect learning time. . Further, the learning time distribution device calculates the distribution of the learning time for each subject to each test subject from which the learner can obtain the highest expected total score or the highest expected total deviation value, and stores and outputs this distribution. Thereby, the learner can know the allocation of the learning time for each subject to each test subject that can efficiently improve the grade without depending on intuition.

The block diagram of the learning time distribution apparatus which concerns on embodiment of this invention. The figure which shows an example of the learning effect correlation matrix which concerns on this embodiment. The flowchart of the learning curve etc. calculation process which concerns on this embodiment. (A) thru | or (e) are the flowchart which shows the process of the subroutine A which concerns on this embodiment, and the figure which shows the function regarding this test standard deviation. The flowchart which shows the process of the subroutine B which concerns on this embodiment. (A) And (b) is a figure which shows the standard learning curve and learning curve which concern on this embodiment. The flowchart which shows the process of the subroutine C which concerns on this embodiment. (A) And (b) is a figure which shows the flowchart and exponential function which show the process of the subroutine D which concerns on this embodiment. (A) And (b) is a figure which shows the learning curve which concerns on this embodiment. The figure which shows the relationship between the effect learning time which concerns on this embodiment, and an expected score. These are the flowcharts of the optimal learning time distribution calculation process which concerns on this embodiment. (A) thru | or (c) are flowcharts which show the process of the subroutines E thru | or G which concern on this embodiment. The figure which shows the output example of the optimal learning time distribution which concerns on this embodiment. The figure which shows the relationship between the learning time for every subject which concerns on this embodiment, and the highest expected total score. The flowchart of the process which calculates the optimal learning time distribution in the case where this test concerning this embodiment is divided into a plurality of stages.

  Hereinafter, a learning time distribution device according to an embodiment of the present invention will be described with reference to the drawings. FIG. 1 shows a configuration of a device 1 (abbreviated as a device) according to the present embodiment. The device 1 includes a CPU 2 (control unit) that controls the device 1, a storage unit 3 that stores data, an input unit 4 such as a mouse or a keyboard to which data is input, a display unit 5 that displays data, and a storage. A storage medium connection unit 7 that reads / writes data from / to the medium 6 and a wireless transmission / reception unit 11 that transmits / receives data to / from the external terminal 9 or the printer 10 via the wireless 8 are connected to the bus 12. Yes.

  Here, the data includes the test subject 13, the current stage deviation value 14, the final test deviation value 15, the subject-specific learning time 16, the grade maintenance learning time 17, the effective learning time 18, the effect learning time 19, the trial average score 20, Main test average score 21, main test standard deviation 22, current stage score 23, learning achievement level 24, learning ease 25, expected score 26, parameter 27, highest expected total score 28, next best expected total score 29, optimum learning time The distribution 30, the next best learning time distribution 31, and the learning effect correlation matrix 32. The test subjects 13 to the learning effect correlation matrix 32 are stored in the storage unit 3. In the present embodiment, an example in which the device 1 calculates the highest expected total score 28 in this test will be described. However, the device 1 calculates the highest expected total deviation value 28 instead of the highest expected total score 28. It is also possible. The apparatus 1 sets the current stage score 23, the expected score 26, the highest expected total score 28, and the next best expected total score 29 to the current stage deviation value 14, the expected deviation value 26, the highest expected total deviation value 28, and the next By replacing with the good expected total deviation value 29, the highest expected total deviation value 28 can be calculated.

  The test subject 13 is a type of test subject such as English, mathematics, or physics. The current stage deviation value 14 is a deviation value for the test subject 13 acquired by the learner in the past main examination or trial. The main test deviation value 15 is a deviation value that serves as an index of the difficulty level of the test subject 13 necessary for passing the main test such as an entrance examination, a national examination, and a qualification examination. The learning time 16 for each subject is the time spent for studying each test subject 13 among the learning time that the learner can spend for studying the exam from the present time to the final exam. The grade maintenance learning time 17 is a study time required to maintain the grade of the test subject 13 at the present time, although the grade of the test subject 13 cannot be improved. The effective learning time 18 is a study time for improving the grade of the test subject 13 and is a value obtained by subtracting the grade maintenance learning time 17 from the subject-specific study time 16. The effect learning time 19 is an effective learning time 18, which is a study time in which the effective learning time 18 of another test subject 13 that affects the improvement of the grade of the predetermined test subject 13 is taken into consideration.

  The trial average 20 is the average score of all examinees in the trial taken by the learner. The average test score 21 is the average score of all examinees in this test. This test standard deviation 22 is the standard deviation of the scores of all examinees in this test. The current stage score 23 is a score for the test subject 13 acquired by the learner in the past main examination questions, or a score estimated that the learner can acquire in the main examination at the present stage from the result of the trial. The learning achievement level 24 is an index indicating how much learning the learner has progressed with respect to the final examination at the present time. The learning ease 25 is an index indicating the degree to which the learning achievement level 24 is improved when the learner learns for a certain period of time. The smaller the numerical value of the learning ease 25, the more the learner needs to spend a longer time for learning in order to improve the grade. The expected score 26 is a score that is estimated to be acquired in the main test when the learner spends the learning time 19 by the main test.

  The parameter 27 is a predetermined percentage. Here, the parameter 27 is, for example, 10%, 5%, and 2.5%. The parameter 27 is used when the effect learning time 19 is allocated to the test subject 13. For example, when the parameter 27 is 10%, the apparatus 1 calculates the optimal learning time distribution with an accuracy of 10% for the test subject 13 by changing the learning time 16 for each subject in increments of 10%. The highest expected total score 28 is the highest total score among the total points of the expected score 26 estimated that the learner can acquire for each test subject 13 in the final examination. The next best expected total score 29 is the highest total score next to the highest expected total score 28 among the total points of the expected score 26 estimated that the learner can obtain in the main examination.

  The optimal learning time allocation 30 is an allocation of the learning time 16 for each subject to each test subject 13 calculated when the apparatus 1 estimates that the highest expected total score 28 can be obtained. The optimum learning time distribution 30 is output as a numerical value, a graphic, a graph, or a sentence. Here, the output is, for example, being displayed on the display unit 5, being written in the storage medium 6, being transmitted to the external terminal 9, or being printed out by the printer 10. The next best learning time distribution 31 is a distribution of the learning time 16 per subject calculated for each test subject 13 when the apparatus 1 estimates that the next best expected total score 29 can be obtained.

  Next, the learning effect correlation matrix 32 will be described with reference to FIG. When there are a plurality of test subjects 13, the learning of a specific test subject 13 by a learner may have a learning effect on the improvement of the results of other test subjects 13. In FIG. 2, there are test subjects A to D as the test subject 13. For example, in the case of university entrance examinations, “Mathematics for Center Examination” and “Mathematics for Secondary Examination” are different subjects, but their learning contents overlap. Therefore, the learning effect correlation matrix 32 shows the relationship between the effective learning times 18 that affect each other in advance for the test subjects A to D by numerical values. For example, when the learner learns the test subject B for one hour, the learner determines that the value of the element at the intersection of the test subject A and the test subject B in the learning effect correlation matrix 32 is 0.3. This means that the subject A has been learned for 18 minutes from 1 hour × 0.3 = 18 minutes. When there are a plurality of test subjects 13, the apparatus 1 can reflect the relationship between the test subjects 13 and the effective learning time 18 in accordance with the substance by using the learning effect correlation matrix 32.

  Next, a learning curve calculation process for calculating the learning achievement degree 24, the learning ease 25, and the learning curve 34 will be described as a stage before the apparatus 1 calculates the optimal learning time distribution 30. Here, the learning curve 34 is a curve used for calculating the expected score 26 using the learning time 16 for each subject as a variable. FIG. 3 shows a flow of a learning curve calculation process in the apparatus 1. The CPU 2 inputs the current stage deviation value 14 via the input unit 4, the storage medium connection unit 7, or the wireless transmission / reception unit 11, or refers to the current stage deviation value 14 stored in advance in the storage unit 3. Thus, the current stage deviation value 14 is acquired (S101). The CPU 2 receives the test subject 13 through the input unit 4, the storage medium connection unit 7, or the wireless transmission / reception unit 11, or refers to the test subject 13 stored in advance in the storage unit 3, A type such as physics is acquired as the test subject 13 (S102).

  The CPU 2 determines whether or not there are a plurality of test subjects 13 (S103). If there are a plurality of test subjects 13, the CPU 2 executes a loop process from step S104 to S111 for each test subject 13 (S112). . On the other hand, when a plurality of test subjects 13 do not exist, the CPU 2 executes the processes of steps S104 to S111 only once and ends the learning curve calculation process. In step S104, the CPU 2 determines whether or not the main test standard deviation 22 for the test subject 13 can be acquired (S104). The CPU 2 receives the main test standard deviation 22 for the test subject 13 via the input unit 4, the storage medium connection unit 7, or the wireless transmission / reception unit 11, or the main test standard deviation 22 is stored in the storage unit 3 in advance. If this is the case (Yes in S104), the test standard deviation 22 is acquired and step S106 is executed. On the other hand, the CPU 2 calculates the main test standard deviation 22 when the main test standard deviation 22 cannot be acquired via the input unit 4, the storage medium connection unit 7, the wireless transmission / reception unit 11, and the storage unit 3 (No in S104). (S105). Here, the process of step S105 is referred to as a subroutine A.

  The subroutine A will be described with reference to FIG. 4A to 4E show functions related to the processing of the subroutine A and the test standard deviation 22. 4A shows the flow of the subroutine A, FIG. 4B shows the transition of the main test standard deviation 22 for each test subject 13 in the main test conducted a plurality of times, and FIG. FIG. 4D shows the relationship between the test standard deviation 22 and the distribution of the test test deviation value 15, and FIG. 4D shows the relationship between the test standard deviation 22 and the test average point 21 when the test subject 13 is the number IIIC. FIG. 4 (e) shows the relationship between the main standard deviation 22 and the main test average point 21 when the test subject 13 is earth science. With reference to FIG. 4A, the CPU 2 calculates the main test standard deviation 22 based on the linear function shown in FIG. 4D or FIG. 4E suitable for each test subject 13.

  4B to 4E show the main test standard deviation 22 and the test subject 13 in advance so that the CPU 2 can calculate the main test standard deviation 22 by estimation when the CPU 2 cannot acquire the main test standard deviation 22. The example which investigated the correlation with this test deviation value 15 and this test average point 21 is shown. As a result, as shown in FIG. 4B, it has been found that there is a correlation between the test standard deviation 22 and the test subject 13. On the other hand, as shown in FIG. 4C, no correlation was found between the distribution of the test standard deviation 22 and the test test deviation value 15. Further, as shown in FIGS. 4D and 4E, it is found that there is a correlation that can be modeled by a linear function between the test standard deviation 22 and the test average point 21 for individual test subjects 13. did. The CPU 2 can calculate the main test standard deviation 22 by using this linear function. This linear function can also be approximated by log.

  Referring to FIG. 3, when CPU 2 calculates main test standard deviation 22 in step S105, CPU 2 obtains the current stage obtained when the learner solved the past questions of the main test or took the main test in the past. It is confirmed whether or not the score 23 exists in the storage unit 3 or the storage medium 6 (S106). The current stage score 23 in step S106 is a score obtained when the learner has solved the past questions of the main examination or has taken the main examination in the past, and is not a score obtained by the learner in the trial. The CPU 2 executes the process of step S108 when the storage unit 3 or the storage medium 6 has a past question of the main test or a current stage score 23 about the past main test (Yes in S106). On the other hand, when the current stage score 23 for the past question of the main test or the past main test does not exist in the storage unit 3 or the storage medium 6 (No in S106), the CPU 2 obtains the current stage score 23 for the test subject 13. Calculate (S107). Here, the processing in step S107 is referred to as subroutine B. The subroutine B will be described with reference to FIG. FIG. 5 shows the processing of subroutine B. The CPU 2 calculates the current stage score 23 based on the main test average point 21, the current stage deviation value 14, the main test deviation value 15, and the main test standard deviation 22.

  Referring to FIG. 3, the CPU 2 calculates a learning curve 34 based on the main test average point 21 and the standard learning curve 33 (S108). Referring to FIG. 6, FIG. 6 (a) shows a standard learning curve 33, and FIG. 6 (b) shows a learning curve 34.

With reference to FIGS. 6A and 6B, the CPU 2 calculates a learning curve 34 by modifying the standard learning curve 33. Referring to FIG. 6A, the horizontal direction is the x-axis, the vertical direction is the G-axis, and the standard learning curve 33 is a curve defined by Equation 1, for example. The x-axis is an index indicating the length of time, and the G-axis is an index indicating the score rate or the height of the deviation value expected to be acquired by the learner such as the expected score 26. Whether the G axis is used as a score or a deviation value is determined depending on whether the index of the desired result is the expected score 26 or the expected deviation value 26. Here, the G axis is used as the score.

  In Equation 1, Gs indicates an example of a function of the standard learning curve 33 with the G axis as a score, and x is the effect learning time 19. The effect learning time 19 is a numerical value indicating the length of time that the learner learns to improve the grade of the test subject 13 until the final examination, and is a variable here. The standard learning curve 33 is a curve that becomes a model of the learning curve 34 to be described later. When the G-axis is used as the point rate, an arbitrary value passing through the coordinates (0, 0.5) and the coordinates (1, 0.6) is used. It is a curve. When the G axis is set as the deviation value, the curve is an arbitrary curve passing through the coordinates (0, 50) and the coordinates (1, 60). The value of the G axis at x = 0 represents an average score when the G axis is a score rate, and represents an average deviation value when the G axis is a deviation value. In FIG. 6A, the G-axis is a score rate, and the average score is represented by a value from 0 to 1. Therefore, the standard learning curve 33 indicates that the average score of the test is 0.5. Since the main test average point 21 in the main test is often not 0.5, the CPU 2 deforms the standard learning curve 33 so that the average point in the standard learning curve 33 matches the main test average point 21. When the G axis is used as the deviation value, the deviation value corresponding to the average point is 50, so there is no need to deform the curve. That is, the standard learning curve 33 and the learning curve 34 match.

FIG. 6B shows a learning curve 34 when the test average score 21 is 0.3. The learning curve 34 is defined by Equation 2, for example. Here, G1 represents a function of the learning curve 34. In the case where the G-axis is used as the point ratio, a is the actual test average point 21. In the case where the G axis is the deviation value, a is 0.5.

  The coordinate indicating the average point in the standard learning curve 33 is P1, and the coordinate indicating that the test average point 21 is 0.3 is P2. The learning curve 34 is a curve obtained by modifying the standard learning curve 33 so that the G coordinate becomes the final test average point 21 at x = 0.

Referring to FIG. 3, CPU 2 calculates learning achievement level 24 from the inverse function of learning curve 34 (step S109). Here, the process of step S109 is referred to as a subroutine C. The subroutine C will be described with reference to FIG. FIG. 7 shows the processing of subroutine C. For example, the CPU 2 calculates the learning achievement degree 24 based on the inverse function defined by Equation 3. Here, t is the learning achievement level 24.

  Referring to FIG. 3, in step S109, when CPU 2 calculates learning achievement level 24, CPU 2 calculates learning ease 25 using main test deviation value 15 (S110). Here, the process of step S110 is referred to as a subroutine D. The subroutine D will be described with reference to FIG. FIG. 8A shows a flow of the subroutine D, and FIG. 8B shows an exponential function 35 with a learning ease of 25. FIG. The learner has different learning times for improving the grade depending on each test subject 13. The difficulty of learning is indicated by a numerical value indicating the difficulty in improving the score. Here, the learning ease 25 means that the smaller the value, the more the learner needs to spend learning time. As illustrated in FIG. 8A, the CPU 2 calculates the learning ease 25 based on the exponential function 35 suitable for each test subject 13.

  With reference to FIG. 8B, the CPU 2 calculates an exponential function 35 suitable for each according to the type of the test subject 13. The CPU 2 calculates the learning ease 25 from the calculated exponential function 35 and the final test deviation value 15. Here, the exponential function 35 for calculating the learning ease 25 is expressed by Equation 4, for example.

  Here, b is the learning ease 25, and h is the test deviation value 15. k 1 is a learning ease coefficient for the test subject 13. When the G-axis of the learning curve 34 is the score, k2 is 50 to 60 in the main test in which the value of the main test deviation 15 is 50 and the value of the main average of the test 21 is 50. It is the effect learning time 19 required to increase the R is the effect learning time 19 required to raise the score from 50 to 60 in the main test where the main test deviation value 15 is 60 and the main test average score 21 is 50. This is a value obtained by dividing the effect learning time 19 required to raise the score from 50 to 60 in the main test in which the value of deviation 15 is 50 and the value of test average point 21 is 50. When the G axis of the learning curve 34 is a deviation value, k2 is the effect learning time 19 required to increase the learner's deviation value from 50 to 60, and R is 1. The CPU 2 can calculate the learning ease 25 for each test subject 13 by using the exponential function 35.

Referring to FIG. 3, CPU 2 combines learning achievement level 24 and learning ease 25 with learning curve 34 (S111). With reference to FIGS. 9A and 9B, the process of step S111 will be described. FIG. 9A shows a modified learning curve 36 in which the learning achievement level 24 is combined with the learning curve 34, and FIG. 9B shows the learning curve 34 with the learning achievement degree 24 and the learning ease 25 on the learning curve 34. A combined characteristic learning curve 37 is shown. In FIG. 9A, the CPU 2 translates the learning curve 34 from the coordinate P2 to the coordinate P3 in the negative x-axis direction by t, where the calculated learning achievement degree 24 is t. A curve obtained by combining this learning curve 34 with the learning achievement level 24 is defined as a deformation learning curve 36, and the deformation learning curve 36 is defined by, for example, Expression 5. Here, G2 represents the function of the deformation learning curve 36, and t is the learning achievement level 24.

Referring to FIG. 9B, the CPU 2 calculates, based on the learning ease 25, how many hours each unit time in the deformation learning curve 36 corresponds to when displayed in time (60 minutes). Thereby, the CPU 2 can calculate the specific learning curve 37. Here, the specific learning curve 37 is defined by, for example, Equation 6, G indicates a function of the specific learning curve 37, and b indicates the learning ease 25.

  The characteristic learning curve 37 is used when the CPU 2 calculates the expected score 26. Referring to FIG. 3, after executing the loop process from step S103 to step S112 for all test subjects 13 (S112), the CPU 2 ends the learning curve calculation process.

  FIG. 10 shows the relationship between the effect learning time 19 and the expected score 26. A to D in FIG. 10 indicate that they correspond to the processes of subroutines A to D in FIG. 3, respectively. The CPU 2 calculates a main test standard deviation 22 based on the test subject 13 and the main test average point 21. The CPU 2 does not calculate the main test standard deviation 22 when the main test standard deviation 22 is stored in the storage unit 3. The CPU 2 calculates a learning ease 25 based on the main test subject 13 and the main test deviation value 15. The CPU 2 calculates the current stage score 23 based on the current stage deviation value 14, the main test standard deviation 22, and the main test deviation value 15. The CPU 2 does not calculate the current stage score 23 when the current stage score 23 is stored in the storage unit 3.

  The CPU 2 calculates a learning achievement degree 24 based on the current stage score 23. Further, the CPU 2 calculates a specific learning curve 37 based on the learning achievement degree 24, the learning ease 25, and the effect learning time 19. The CPU 2 calculates the expected score 26 based on the specific learning curve 37. As described above, the apparatus 1 can calculate the expected score 26 based on the test subject 13, the final test average point 21, the final test deviation value 15, and the effect learning time 19.

  Next, the process in which the apparatus 1 calculates the optimal learning time distribution 30 will be described. FIG. 11 shows a flow of optimal learning time distribution calculation processing in the apparatus 1. E to G in FIG. 11 indicate that the process of step S201 is subroutine E, the processes of steps S203, S207, and S211 are subroutine F, and the processes of steps S204, S208, and S212 are subroutines. Indicates G. The CPU 2 calculates the sum of the learning time 16 for each subject based on information input via the input unit 4, the storage medium connection unit 7, or the wireless transmission / reception unit 11, or information stored in advance in the storage unit 3. (S201). With reference to FIG. 12, the processing of subroutine E will be described. 12A shows the flow of processing of subroutine E, FIG. 12B shows the flow of processing of subroutine F, and FIG. 12C shows the flow of processing of subroutine G.

  With reference to Fig.12 (a), CPU2 calculates the remaining days from this time to this test based on the calendar information which the memory | storage part 3 memorize | stores (S301). It is assumed that the date on which this test is performed is input in advance via the input unit 4 or the wireless transmission / reception unit 11 or stored in the storage unit 3 in advance. In addition, it is assumed that the learner inputs information related to the learning time to the apparatus 1 in advance via the input unit 4 and the wireless transmission / reception unit 11. Here, for example, the learning time information can be assigned 3 hours per day as learning time from Monday to Friday, and 6 hours per day as learning time on Saturdays, Sundays, holidays and holidays. It is information that can be done. The CPU 2 calculates the sum of the learning time 16 for each subject based on the information on the day of the week, holiday information, the number of remaining days until the main examination, and the learning time held in the calendar information (S302). The sum is stored in the storage unit 3.

  Referring to FIG. 11, the CPU 2 calculates the optimal learning time distribution 30 in which the total score of each test subject 13 in the main test becomes the highest expected total score 28 by executing the processing of steps S202 to S213. The CPU 2 refers to the storage unit 3 and acquires the parameter 27 (S202). The parameter 27 acquired here is 10%. CPU 2 does not change the sum of the learning time 16 per subject calculated in the processing of subroutine E, and changes the distribution of the learning time 16 per subject 16 for each test subject 13 in increments of 10%. The process of assigning each learning time 16 is repeated (S202 to S205). The CPU 2 calculates an effective learning time 19 after calculating an effective learning time 18 for each subject learning time 16 assigned to each test subject 13 (S203).

  With reference to FIG. 12B, the processing of the subroutine F will be described. The CPU 2 repeats the process of calculating the effective learning time 18 for each test subject 13 (S401 to S403). The CPU 2 calculates the effective learning time 18 for the test subject 13 by subtracting the grade maintenance learning time 17 from the learning time 16 for each subject assigned to the test subject 13 (S402). When the CPU 2 calculates the effective learning time 18 for all the test subjects 13 (S403), the CPU 2 calculates the effect learning time 19 for each test subject 13 by multiplying the effective learning time 18 by the learning effect correlation matrix 32 (S404). ). Referring to FIG. 11, CPU 2 calculates optimal learning time distribution 30 and next-best learning time distribution 31 based on effect learning time 19 and records them in storage unit 3 (S204).

  With reference to FIG. 12C, the processing of the subroutine G will be described. The CPU 2 calculates the expected score 26 of the test subject 13 by substituting the effect learning time 19 of the test subject 13 into the specific learning curve 37, and calculates the total of the expected scores 26 for all the test subjects 13 (S501). ). The CPU 2 compares the total of the expected scores 26 with the highest expected total score 28 and the next best expected total score 29 stored in the storage unit 3 (S502). When the total of the newly calculated expected score 26 is lower than the highest expected total score 28 and the next best expected total score 29 (No in S502), the CPU 2 ends the processing of the subroutine G.

  On the other hand, when the total of the newly calculated expected score 26 is higher than the highest expected total score 28 and the next best expected total score 29 (Yes in S502), the CPU 2 sets the newly calculated expected score 26 to the highest The expected total score 28 or the next best expected total score 29 is recorded in the storage unit 3. Thus, the storage unit 3 can always store the highest score value as the highest expected total score 28 out of the total expected score 26 and store the second highest value as the next best expected total score 29. it can. Further, the CPU 2 records the distribution of the learning time 16 for each subject assigned to each test subject 13 when the highest expected total score 28 is reached as the optimal learning time distribution 30 in the storage unit 3 (S503). Similarly to the optimal learning time distribution 30, the CPU 2 stores the distribution of the learning time 16 for each subject assigned to each test subject 13 when the next best expected total score 29 is reached as the next best learning time distribution 31. Record in section 3 (S503).

  Referring to FIG. 11, when CPU 2 repeats the processes of steps S202 to S205, it executes the processes of steps S206 to S209. The CPU 2 refers to the storage unit 3 and acquires the parameter 26 (S206). The parameter 26 acquired here is 5%. The CPU 2 increases or decreases the learning time 16 per subject allocated to each test subject 13 by 5% or 5% with respect to the optimal learning time allocation 30 and the next best learning time allocation 31 stored in the storage unit 3. Each time it is changed, the processes of the subroutines F and G are executed for the learning time 16 for each subject allocated to each test subject 13 in each case (S207, S208). As a result, the CPU 2 can obtain the highest expected total score 28 and the next best expected total score 29 more than the optimum learning time distribution 30 and the next best learning time distribution 31 calculated in step S204 with fewer calculation steps. 30 and suboptimal learning time allocation 31 can be calculated with an accuracy of 5%.

  When the CPU 2 repeats the processes of steps S205 to S209, it executes the processes of steps S210 to S213. The CPU 2 refers to the storage unit 3 and acquires the parameter 26 (S210). The parameter 26 acquired here is 2.5%. The CPU 2 increases the learning time 16 for each subject allocated to each test subject 13 by 2.5% with respect to the optimal learning time allocation 30 and the next best learning time allocation 31 stored in the storage unit 3 or 2.5. Every time the change is made by decreasing the percentage, the processes of subroutines F and G are executed for the learning time 16 for each subject allocated to each test subject 13 in each case (S211 and S212). Thereby, the CPU 2 can obtain the highest expected total score 28 and the next best expected total score 29 more than the optimum learning time distribution 30 and the next best learning time distribution 31 calculated in step S208 with few calculation steps. 30 and suboptimal learning time allocation 31 can be calculated with an accuracy of 2.5%.

  The CPU 2 outputs the optimum learning time distribution 30 at the time when step S213 is completed as a calculation result together with figures and sentences (S214), and ends the optimum learning time distribution calculation process. FIG. 13 shows an output example of the optimal learning time distribution 30. Here, the optimal learning time distribution 30 is output as a pie chart, and the total of the learning time 16 for each subject up to the test day is 268 hours, and the expected score 26 is output as a score after studying. Furthermore, a score that has been weighted and added so that the score after studying has the same score ratio as in the main exam is output as a score after the graded score. As described above, the apparatus 1 effectively improves the performance without depending on the intuition of the learner by executing the learning curve calculation process shown in FIG. 3 and the optimum learning time distribution calculation process shown in FIG. It is possible to present to the learner the distribution of learning time for each test subject 13 that can be made. Thereby, the learner can know the optimal distribution of the learning time to each test subject 13 based on his / her own current performance and the learning time specific to himself / herself until the final examination.

  FIG. 14 shows the relationship between the subject learning time 16 and the highest expected total score 28. E and F in FIG. 14 indicate that they correspond to the processes of subroutines E and F in FIG. 11, respectively. The CPU 2 calculates the sum of the learning time 16 for each subject and the number of remaining days until the main test based on information related to the learning time of the learner, the date on which the main test is performed, and calendar information. The CPU 2 changes the subject learning time 16 for the predetermined test subject 13 based on the value of the parameter 27. Here, the value of the parameter 27 is, for example, 10%, 5%, and 2.5%. The CPU 2 calculates an effective learning time 18 and an effect learning time 19 for the predetermined test subject 13. The CPU 2 calculates the expected score 26 of the test subject 13 by substituting the effect learning time 19 for the predetermined test subject 13 into the specific learning curve 37. The CPU 2 calculates the highest expected total score 28 by summing up the expected scores 26 for all the test subjects 13.

  Next, the distribution of the learning time 16 per subject when the main test is divided into a plurality of stages as in the primary test and the secondary test will be described. For example, when the main test is divided into a primary test and a secondary test, the allocation of the learning time 16 per subject during the period from the current stage to the primary test is divided according to the subject during the period from the primary test to the secondary test. The distribution of the learning time 16 is affected. Therefore, the apparatus 1 performs the learning curve calculation processing shown in FIG. 3 and the optimum shown in FIG. 11 for all the test subjects 13 in the primary test and the secondary test and the period from the current stage to the secondary test. It is desirable to calculate the optimal learning time distribution 30 by executing the learning time distribution calculation process.

  However, when the apparatus 1 executes the learning ease calculation process and the optimum learning time distribution calculation process for the primary test and the secondary test at a time, the calculation amount becomes enormous and the load on the apparatus 1 becomes extremely large. Therefore, the apparatus 1 executes a learning ease calculation process and an optimal learning time distribution calculation process for each stage, such as a period from the current stage to the primary test, and a period from the primary test to the secondary test. Thereby, the apparatus 1 can calculate the optimal learning time distribution 30 with high accuracy quickly while avoiding a high load.

  FIG. 15 shows a process for calculating the optimum learning time allocation 30 when the test is divided into a plurality of stages. In FIG. 15, the main test is divided into three stages, a primary test, a secondary test, and a tertiary test, and the test dates of the primary test, the secondary test, and the tertiary test are respectively the primary test date, The secondary test date and the third test date. The first learning period 38 indicates the period from the current time to the primary test date, the second learning period 39 indicates the period from the primary test date to the secondary test date, and the third learning period. 40 indicates a period from the secondary test date to the tertiary test date.

  The CPU 2 executes processes from steps S601 to S603 as initial steps, and subsequently executes processes from steps S604 to S606 as second steps. The CPU 2 calculates the distribution of the learning time 16 for each subject in the first learning period 38 in which the expected score 26 of the primary test is maximized (S601). Here, the distribution of the learning time 16 for each subject in the first learning period 38 is calculated by the CPU 2 executing a learning curve calculation process and an optimal learning time distribution calculation process. Next, the CPU 2 maintains the distribution of the learning time 16 for each subject in the first learning period 38 thus calculated, and the learning time 16 for each subject in the second learning period 39 in which the expected score 26 of the secondary examination is maximized. (S602). Furthermore, the CPU 2 maintains the distribution of the learning time 16 for each subject in the calculated first learning period 38 and second learning period 39, and the third learning period 40 in which the expected score 26 of the tertiary examination is maximized. The distribution of the learning time 16 for each subject is calculated (S603).

  After executing the initial step processing, the CPU 2 subsequently executes the second step processing. Thereby, the CPU 2 can calculate the distribution of the learning time 16 for each subject with higher accuracy without applying a high load to the apparatus 1. The CPU 2 maintains the distribution of the learning time 16 for each subject in the calculated second learning period 39 and the third learning period 30, and the subjects in the first learning period 38 in which the expected score 26 of the primary examination is maximized. The distribution of each learning time 16 is newly calculated (S604). Next, the CPU 2 maintains the distribution of the learning time 16 for each subject in the calculated first learning period 38 and third learning period 40, and the second learning period in which the expected score 26 of the secondary examination is maximized. The distribution of the learning time 16 per subject in 39 is newly calculated (S605).

  Further, the CPU 2 maintains the distribution of the learning time 16 for each subject in the newly calculated first learning period 38 and second learning period 39, while the third learning in which the expected score 26 of the tertiary examination is maximized. The distribution of the learning time 16 for each subject in the period 40 is newly calculated (S606). The CPU 2 outputs the distribution of the learning time 16 for each subject in the first learning period 38, the second learning period 39, and the third learning period 40 calculated in the process of the second step as the optimum learning time distribution 30. The process ends. Note that the device 1 may be set to repeat the process of the second step in order to increase the accuracy of the optimal learning time distribution 30. As a result, the device 1 can distribute the optimal learning time even when the main examination is divided into multiple stages, such as a university entrance examination and a national examination, in addition to a single examination such as a junior high school entrance examination and a high school entrance examination. 30 can be calculated.

  Further, the apparatus 1 may be configured to repeatedly execute steps S201 to S214. As a result, the test date of the test is not fixed to one, and even for the test that can be taken multiple times, by repeating steps S201 to S214 for all the test dates, The optimum learning time distribution 30 can be calculated similarly. As a result, the apparatus 1 can be applied to a wide range of tests such as a qualification test that is performed several times a year and an achievement test that is planned to be introduced in a university entrance examination in the future.

  The apparatus 1 according to the present invention is not limited to the configuration of the above embodiment, and various modifications are possible. The current stage score 23, the expected score 26, the highest expected total score 28, and the next best expected total score 29 are The current stage deviation value 14, the expected deviation value 26, the highest expected total deviation value 28, and the second best expected total deviation value 29 may be used. Further, the apparatus 1 may be configured to transmit / receive data to / from the external terminal 9 or the printer 10 via the wireless transmission / reception unit 11, instead of transmitting / receiving data to / from each other through a wired connection. The apparatus 1 is not configured to store the highest expected total score 28, the next best expected total score 29, and the optimum learning time distribution 30 and the next best learning time distribution 31, but the highest expected total score 28 and the optimum learning time. The configuration may be such that only the distribution 30 is stored.

  Further, the apparatus 1 may have a configuration in which the subject learning time 16 or the effective learning time 18 is substituted for the effect learning time 19 in the specific learning curve 37 in FIG. In addition, in FIG. 11, the apparatus 1 may not be configured to execute the processes of steps S202 to S213 but may be configured to execute only the processes of steps S202 to S205 or the processes of steps S202 to S209. Further, the CPU 2 uses the data input via the input unit 4, the storage medium connection unit 7, or the wireless transmission / reception unit 11 instead of the data stored in the storage unit 3, so that the optimal learning time distribution 30 is obtained. It may be configured to calculate Furthermore, the apparatus 1 may be configured to be used for the purpose of determining the possibility of passing in the main test with high accuracy based on the expected deviation value 26 of the learner, for example.

1 Learning time distribution device 2 CPU (control unit)
3 Storage unit 4 Input unit 13 Exam subject 14 Current stage deviation value 16 Course learning time 17 Grade maintenance learning time 18 Effective learning time 19 Effective learning time 23 Current stage score 24 Learning achievement 25 Learning ease 26 Expected score (expected deviation) value)
27 Parameter 28 Maximum expected total score (maximum expected total deviation)
30 Optimal learning time allocation 32 Learning effect correlation matrix 34 Learning curve 37 Specific learning curve (learning curve combining learning ease and learning achievement)

To achieve the above object, the present invention provides a storage unit for storing data, an input unit to which data is input, and learning of a plurality of test subjects performed in a main test taken by a learner based on the data. And a learning time distribution device comprising a control unit that calculates an optimal distribution result of the time spent on the learning unit, the data stored in the storage unit or the data input to the input unit is determined by the learner before the final examination. The amount of study time expected to be spent for improving grades, the current stage deviation value or score obtained by the learner in the final test or mock test, and the test subjects required to pass the final test a present study deviation value, the control unit on the basis of the test subjects and the test deviation, to calculate a learning Simplicity is an indication of the length of the learning time required to improve the performance of the test subjects, It can be acquired in accordance with the learning time and learning time Calculating a learning curve representing the relationship between the expected score or deviation, and the inverse function of the learning curve, based on the current stage deviation or stage scores, progress of the learning to the test subjects learners in stage By calculating a numerical value indicating the degree of learning as a learning achievement and substituting the learning time into a learning curve that combines the learning ease and the learning achievement, a score or deviation value that the learner can acquire for the test subject in this exam is expected. Or it calculates as an expected deviation value, and memorize | stores an expected score or an expected deviation value in a memory | storage part.

Further, the present invention provides the improved invention described above, the control unit, when the learner has learned one of the test subjects, affecting correlation information in learning other test subjects, as learning a correlation matrix The learning time is stored in the storage unit, and the learning time allocated to each test subject is taken as the learning time per subject for the test subject, and the learning time per subject is changed in increments of one or more predetermined parameters. The effective learning time for each test subject is calculated by subtracting the result maintenance learning time necessary for the learner to maintain the current score or deviation value for the test subject from the time. by multiplying the time to learning the correlation matrix, and calculates the effect learning time for each test subject, based on the learning Simplicity and academic achievement corrected learning curve, to substitute the effect learning time to learning curve By calculating the expected score or expected deviation value for each test subject, calculating the total expected score or expected deviation value for all test subjects, and learning for each subject allocated to the exam subject according to the parameter increments Calculate the total of expected scores or expected deviation values for all test subjects for all patterns of time, and among the calculated expected scores or expected deviation values, the highest expected score or expected deviation value Is stored in the storage unit as the highest expected total score or the highest expected deviation value, and the learning result allocation result for each test subject corresponding to the highest expected total score or the highest expected total deviation value is stored as the optimal learning time distribution. Remember me.

Using a computer comprising a storage unit for storing data, an input unit for inputting data, and a control unit for processing data, learning of a plurality of test subjects carried out in the main examination taken by the learner In the learning time allocation program that calculates the optimal allocation result of the time spent, the data stored in the storage unit or the data input to the input unit should be spent by the learner to improve the grade of the test subject by the final examination. Expected learning time, current stage deviation value or current stage score obtained by the learner in the main examination or mock examination in the past, and main examination deviation value of the test subjects necessary to pass this examination , Based on the test subject and the main test deviation value , the control unit calculates a learning ease that is an index of the length of the learning time necessary for improving the grade of the test subject, and the learning time and the learning time. Acquire according to A step of calculating a learning curve representing the relationship between the quilt expected score or deviation, and the inverse function of the learning curve, based on the current stage deviation or stage scores, to the test subjects learners in stage The step of calculating the numerical value indicating the degree of learning as a learning achievement, and the learning time is substituted into the learning curve that combines the learning ease and the learning achievement, thereby obtaining a score that the learner can acquire for the test subject in this test or a step of calculating a deviation value as the expected score or expected deviation, to execute the steps of storing the expected score or expected deviation in the storage unit.

Further, the present invention provides the improved invention described above, the control unit, when the learner has learned one of the test subjects, affecting correlation information in learning other test subjects, as learning a correlation matrix A step of storing in the storage unit, a learning time allocated to each test subject as a learning time per subject for the test subject, and a step of changing the learning time per subject in increments of one or more predetermined parameters; , A step in which the learner calculates the effective learning time for each test subject by subtracting the grade maintenance learning time necessary for the learner to maintain the current score or deviation value for the test subject from the study time for each subject, the effective learning time of each test subject by multiplying the learning effect correlation matrix, a step of calculating the effect learning time for each test subject, based on the learning Simplicity and academic achievement The learning curve is corrected Te, by substituting the effective learning time to learning curve, expected for each test subjects scored or by calculating the expected deviation, to calculate the sum of expected scores or expected deviation for all of the test subjects Calculating the total expected scores or expected deviation values for all test subjects for each pattern of the learning time for each subject allocated to the test subjects according to the step of the parameter, and the calculated expectation Storing the highest expected total score or expected deviation value in the storage unit as the highest expected total score or highest expected deviation value among the total of the scores or expected deviation values, and the highest expected total score or highest expected total a step of storing in the storage unit allocation result of the study period for each test subject corresponding to the deviation value as an optimum learning time allocation, is allowed to run.

The trial average 20 is the average score of all examinees in the trial taken by the learner. The average test score 21 is the average score of all examinees in this test. This test standard deviation 22 is the standard deviation of the scores of all examinees in this test. The current stage score 23 is a score for the test subject 13 acquired by the learner in the past main examination questions, or a score estimated that the learner can acquire in the main examination at the present stage from the result of the trial. The learning achievement level 24 is an index indicating how much learning the learner has progressed with respect to the final examination at the present time. The learning ease 25 is an index indicating the degree to which the learning achievement level 24 is improved when the learner learns for a certain period of time. The smaller the numerical value of the learning ease 25, the more the learner needs to spend a longer time for learning in order to improve the grade. A learning ease value of 1 means that learning the test subject 13 for 1 hour increases the expected score 26 from 50 to 60, and the learning ease value 25 is doubled. Then, the learning time 16 per subject necessary for raising the expected score 26 from 50 to 60 is halved. The expected score 26 is a score that is estimated to be acquired in the main test when the learner spends the learning time 19 by the main test.

Referring to FIG. 9B, the CPU 2 calculates, based on the learning ease 25, how many hours each unit time in the deformation learning curve 36 corresponds to when displayed in time (60 minutes). Thereby, the CPU 2 can calculate the specific learning curve 37. Here, the specific learning curve 37 is a curve obtained by multiplying the deformation learning curve 36 by the learning ease 25, and is defined by , for example, Equation 6, G indicates a function of the specific learning curve 37, and b indicates easy learning. Degree 25 is shown.

To achieve the above object, the present invention provides a storage unit for storing data, an input unit to which data is input, and learning of a plurality of test subjects performed in a main test taken by a learner based on the data. And a learning time distribution device comprising a control unit that calculates an optimal distribution result of the time spent on the learning unit, the data stored in the storage unit or the data input to the input unit is determined by the learner before the final examination. The amount of study time expected to be spent for improving grades, the current stage deviation value or score obtained by the learner in the final test or mock test, and the test subjects required to pass the final test This test average value is the average of the test deviation value and the average score of all examinees in this test , and the control unit is required to improve the results of this test subject based on the test subject and this test deviation value. Is an indicator of the length of the learning time Calculating a learning easiness, the relationship between the learning time (x) and the expected score or deviation to be earned in accordance with the learning time (Gs)
Based on the learning curve's inverse function and the current stage deviation value or current stage score, the learning level is calculated as the learning progress of the learner for the test subject at the current stage. By substituting learning time into a learning curve that combines learning ease , learning achievement , and the average score of this exam, the score or deviation value that the learner can acquire for the exam subject in this exam is the expected score or expected deviation value. And the expected score or the expected deviation value is stored in the storage unit.

Using a computer comprising a storage unit for storing data, an input unit for inputting data, and a control unit for processing data, learning of a plurality of test subjects carried out in the main examination taken by the learner In the learning time allocation program that calculates the optimal allocation result of the time spent, the data stored in the storage unit or the data input to the input unit should be spent by the learner to improve the grade of the test subject by the final examination. Expected learning time, current stage deviation value or current stage score obtained by the learner in the main examination or mock test in the past, main examination deviation value of test subjects necessary to pass this examination, and main examination This is the average score of the exams, which is the average of the scores of all candidates in the test, and the control unit determines the length of the learning time necessary for improving the grades of the test subjects based on the test subjects and the test deviation values. Learning capacity as an index A step of calculating a degree, the relationship between the expected score or deviation to be earned in accordance with the learning time (x) and the learning time (Gs)
The learning achievement level is a numerical value indicating the progress of learning for the test subject at the current stage based on the step of calculating the learning curve expressed as , the inverse function of the learning curve, and the current stage deviation value or current stage score. By substituting the learning time into the learning curve that combines the step of calculating as a learning curve, the degree of learning, the achievement of learning , and the average score of the test, we expect the score or deviation that the learner can acquire for the test subject in the test A step of calculating as a score or an expected deviation value and a step of storing the expected score or the expected deviation value in a storage unit are executed.

With reference to FIGS. 6A and 6B, the CPU 2 calculates a learning curve 34 by modifying the standard learning curve 33. Referring to FIG. 6A, the horizontal direction is the x-axis, the vertical direction is the G-axis, and the standard learning curve 33 is a curve defined by Equation 1. The x-axis is an index indicating the length of time, and the G-axis is an index indicating the score rate or the height of the deviation value expected to be acquired by the learner such as the expected score 26. Whether the G axis is used as a score or a deviation value is determined depending on whether the index of the desired result is the expected score 26 or the expected deviation value 26. Here, the G axis is used as the score.

In Formula 1, Gs indicates the function number of the standard learning curve 33, scoring rate G axis, an example of c is 1.5, x is an effective learning time 19. The effect learning time 19 is a numerical value indicating the length of time that the learner learns to improve the grade of the test subject 13 until the final examination, and is a variable here. Standard learning curve 33 is a curve the model learning curve 34 which will be described later, you express by (Equation 1). The value of the G axis at x = 0 represents an average score when the G axis is a score rate, and represents an average deviation value when the G axis is a deviation value. In FIG. 6A, the G-axis is a score rate, and the average score is represented by a value from 0 to 1. Therefore, the standard learning curve 33 indicates that the average score of the test is 0.5. Since the main test average point 21 in the main test is often not 0.5, the CPU 2 deforms the standard learning curve 33 so that the average point in the standard learning curve 33 matches the main test average point 21. When the G axis is used as the deviation value, the deviation value corresponding to the average point is 50, so there is no need to deform the curve. That is, the standard learning curve 33 and the learning curve 34 match.

Claims (4)

A learning time distribution apparatus comprising: a storage unit for storing data; an input unit for inputting data; and a control unit for calculating an optimum distribution result of time spent by the learner for learning the test subjects based on the data In
The data stored in the storage unit or the data input to the input unit includes the learning time that the learner is expected to spend to improve the grade of the test subject before the final examination, A learning ease indicating numerically the current stage deviation value or current stage score obtained in the test or mock test, and the difficulty of improving the score or deviation value expected to be obtained for the test subject,
The controller is
Calculating a learning curve with the learning time as a variable;
Based on the inverse function of the learning curve and the current stage deviation value or the current stage score, a numerical value indicating the degree of learning progress of the learner for the test subject at the current stage is calculated as a learning achievement degree,
By assigning the learning time to the learning curve that combines the ease of learning and the achievement of learning, the score or deviation value that the learner can acquire for the test subject in this test is calculated as an expected score or an expected deviation value. And
The learning time distribution apparatus, wherein the expected score or the expected deviation value is stored in the storage unit. The controller is
If there are multiple test subjects,
When a learner learns one of the test subjects, the correlation information affecting the learning of the other test subjects is stored in the storage unit as a learning effect correlation matrix,
The learning time allocated to each of the test subjects is set as the learning time per subject for the test subject, and the learning time per subject is changed in increments of one or more predetermined parameters.
A value obtained by subtracting the grade maintenance learning time necessary for the learner to maintain the current score or deviation value for the test subject from the learning time for each subject is calculated as the effective learning time for each test subject,
By multiplying the effective learning time for each test subject by the learning effect correlation matrix, the effect learning time for each test subject is calculated,
By substituting the effect learning time into the learning curve that combines the learning ease and the learning achievement degree, the expected score or the expected deviation value for each test subject is calculated, and Calculate the total of the expected score or the expected deviation value,
Calculate the total of the expected score or the expected deviation value for all the test subjects for every pattern of the learning time for each subject allocated to the test subjects according to the increment of the parameter,
Of the calculated expected score or the total of the expected deviation values, the largest expected score or the total of the expected deviation values is stored in the storage unit as the highest expected total score or the highest expected deviation value,
2. The learning according to claim 1, wherein the storage unit stores the learning time distribution result for each test subject corresponding to the highest expected total score or the highest expected total deviation value as an optimal learning time distribution. Time distribution device. Using a computer having a storage unit for storing data, an input unit for inputting data, and a control unit for processing the data, a learner calculates an optimal distribution result of time spent for learning the test subjects In the learning time allocation program,
The data stored in the storage unit or the data input to the input unit includes the learning time that the learner is expected to spend to improve the grade of the test subject before the final examination, A learning ease indicating numerically the current stage deviation value or current stage score obtained in the test or mock test, and the difficulty of improving the score or deviation value expected to be obtained for the test subject,
In the control unit,
Calculating a learning curve with the learning time as a variable;
Based on the inverse function of the learning curve and the current stage deviation value or the current stage score, a step of calculating a numerical value indicating the progress of learning of the learner for the test subject at the current stage as a learning achievement level;
By assigning the learning time to the learning curve that combines the ease of learning and the achievement of learning, the score or deviation value that the learner can acquire for the test subject in this test is calculated as an expected score or an expected deviation value. Step to
A learning time distribution program causing a computer to execute the step of storing the expected score or the expected deviation value in the storage unit. In the control unit,
If there are multiple test subjects,
When the learner learns one of the test subjects, the correlation information affecting the learning of the other test subjects is stored in the storage unit as a learning effect correlation matrix;
The learning time allocated to each test subject is set as a learning time per subject for the test subject, and the learning time per subject is changed in increments of one or more predetermined parameters.
A step of calculating a value obtained by subtracting the grade maintenance learning time necessary for the learner to maintain the current score or deviation value for the test subject from the learning time for each subject as the effective learning time for each test subject. When,
Calculating the effect learning time for each test subject by multiplying the effective learning time for each test subject by the learning effect correlation matrix;
By substituting the effect learning time into the learning curve that combines the learning ease and the learning attainment, the expected score or the expected deviation value for each test subject is calculated, and all of the test subjects are calculated. Calculating the total of the expected score or the expected deviation value;
Calculating the total of the expected score or the expected deviation value for all the test subjects for every pattern of the learning time per subject allocated to the test subjects according to the increment of the parameter;
Storing the expected score or the sum of the expected deviation values having the largest value among the calculated expected scores or the expected deviation values in the storage unit as the highest expected total score or the highest expected deviation value; ,
Causing the computer to execute the step of storing the learning time distribution result for each test subject corresponding to the highest expected total score or the highest expected total deviation value in the storage unit as an optimal learning time distribution. The learning time distribution program according to claim 3.