JP5437211B2 - E-learning system with problem extraction function considering question frequency and learner's weakness - Google Patents

Description

  In the self-supporting learning system using the Internet environment, the present invention automatically extracts a problem based on a function of analyzing a learner's weaknesses in order to increase learning efficiency and the number of registered weaknesses and the number of registered problems. The present invention relates to a learning system that gives questions to learners.

As IT technology advances, e-learning systems that perform learning on personal computers are becoming common learning methods. The advantage of the e-learning system is that you can study by yourself without the need for a special lecturer. There are already various systems for learning problems as e-learning.
There are many methods that the system issues, in the form of tests, to let users learn. There are some questions that are fixed at each learning, but the e-learning system analyzes the weak points and extracts the problems along the weak points to improve the learning effect. There are many systems that let you answer.
In the case of these technologies, the problem data and the question content data are managed separately, and it is possible to search the registered question content according to the content judged as a weak point. It is a mechanism to extract problems. (Patent Document 1, Patent Document 2)

Japanese Patent Laid-Open No. 2005-066222 JP 2005-164965 A

The above-described technique is characterized by analyzing weak point classification based on a wrong problem and performing learning along the weak point. Considering the wrong problem as a weak point of the learner and supplementing its classification can be considered as an effective means to improve the quality of learning.
However, even if the classification determined to be a weak point is learned intensively and the weak point is overcome, if the classification with a low question frequency in the actual examination is a weak point, the learning effect may be reduced.
In addition, in the case of the above technology, when adding a problem, not only the problem data, but also data for recording which problem is presented in the learning as data that associates the problem with the learning content data must be registered. I must. For this reason, when it is necessary to replace a problem, it is considered that not only the problem data but also the data of the question contents must be paid attention.
The problem to be solved by the present invention is to provide a learning support system that not only considers the weaknesses of learners but also considers the frequency of questions.

As a means for solving the above problems, the system according to the present invention stores problems, correct answers and explanations, a classification of the problems, and a problem date indicating that the problem is a test problem presented in a previous main examination. A recording device having problem data, learning result recording data storing the learning status and achievement rate and achievement level of each classification in the learner, and a system setting file that holds parameters necessary for calculation during learning;
A data processing device that extracts a problem for a learner, calculates a learning result, and stores it in the recording device, and an e-learning system,
(A) The system setting file is:
(A1) An upper limit of increase in achievement rate for one learning that sets an upper limit for increasing the achievement rate in one learning;
(A2) an assumed achievement rate maximum value for setting an assumed maximum value of the achievement rate in iterative learning;
(A3) holding the achievement degree achievement rate corresponding value in which the correspondence between the achievement degree and the range of the achievement percentage is set,
(B) The data processing device includes:
(B1) A question trend analysis processing unit that analyzes a question tendency based on the problem data, extracts all test questions with the question date with reference to the problem data, and extracts the extracted test questions for each classification Each class is a product of the means for calculating the ratio of the number of test questions in each category to the total number of test questions extracted and extracted, and the product of the total number of questions that can be answered in repeated learning and the ratio of the number of test questions in each class A question tendency analysis processing unit having means for calculating a question tendency tendency coefficient of
(B2) An achievement rate increase value calculation unit that calculates an increase rate of achievement rate per question of each classification based on the question tendency analyzed by the question trend analysis processing unit, and the assumption that the system setting file holds The achievement rate increase value calculation unit having means for calculating the achievement rate increase value of each classification, which is a value obtained by dividing the achievement rate maximum value by the question tendency coefficient of each classification;
(B3) A weak point learning problem extraction unit that extracts a question to be given to the learner based on the achievement rate increase value calculated by the achievement rate increase value calculation unit, wherein the achievement level included in the learning result recording data is By referencing, all the classifications of the minimum achievement level are extracted, and the learning 1 held in the system setting file in order from the classification of the lowest achievement level to the lowest achievement rate among the extracted classifications of the minimum achievement level Means for determining the number of problems to be extracted based on a value obtained by dividing the upper limit of the achievement rate increase of the number of times by the increase value of the achievement rate of the classification, and the problem for weak point learning of each classification according to the determined number of problems Means for extracting from the data, the weak point learning problem extraction unit having,
(B4) A weakness learning result calculation unit for scoring the answer to the weakness learning problem by the learner and judging the ability, the correct answer number in the classification, and the achievement rate increase value of the classification held in the system setting file The means for updating the achievement rate of each classification included in the learning result record data by adding the product of, and the achievement degree achievement rate corresponding value held in the updated achievement rate and the system setting file And a means for updating the achievement level of each classification included in the learning result recording data .

The above system is characterized in that it has a function that can analyze weak points in consideration of information on the number of registered problems (probing questions by continuing to include only problems).
In the above system, the problem data stored in the recording device further includes a weakness analysis flag for determining whether to give a question as a weakness analysis problem at the start of learning,
(A) The system setting file is:
(A4) holding an initial achievement rate in which the correspondence between the number of the weak point analysis problems at the start of learning and the number of correct answers is set in advance;
(B) The data processing device includes:
(B5) a weak point analysis problem extracting unit that extracts problem data having the weak point analysis flag of 1 to measure a learner's learning ability at the start of learning, and outputs the problem data as a weak point analysis problem;
(B6) A weak point analysis result calculation unit that determines the ability by scoring the result of the learner solving the weak point analysis problem that has been output, and totalizes the number of correct answers for each classification for the weak point analysis problem Means for recording the initial achievement rate of each classification of the learner in the learning result recording data based on the result and the initial achievement ratio held in the system setting file, the initial achievement ratio of each recorded classification, and the system setting file The weak point analysis result calculation unit having a means for recording the initial achievement degree of each classification in the learning result record data based on the achievement degree achievement rate correspondence value held by the computer.

  According to the above invention, at the beginning of a series of learning, a student who considers the frequency of classification questions by analyzing the registered test questions and determining a standard value for evaluation based on the number of questions for each classification It becomes possible to evaluate the problem, and the problem for effective learning can be extracted by using this evaluation and the evaluation standard.

  In addition, the problem is extracted dynamically based on the current evaluation, so new questions can be automatically targeted by adding problem data, and the frequency of questions is updated. Will also be able to do.

It is a system configuration figure showing an embodiment of the present invention. 2 is a flowchart of learning performed by the system shown in FIG. 1. It is a flowchart of the process in the initial setting (23 of FIG. 2) of scoring and a learner's data. It is a flowchart of the process in extracting the problem of weak point learning and making a learner perform learning (2 of FIG. 2). It is a flowchart of the process in a scoring and the update of a learner's data (25 of FIG. 2). 4 is a configuration diagram of problem data stored in a problem data storage area 41. FIG. 4 is a configuration diagram of classification data stored in a classification data storage area 42. FIG. 4 is a configuration diagram of learner data stored in a learner data storage area 43. FIG. 4 is a configuration diagram of learning result data of a learner stored in a learning result data storage area 44. FIG. 4 is a configuration diagram of data stored in a learning history storage area 45. FIG. 4 is a configuration diagram of data stored in an answer data storage area 46. FIG. FIG. 2 is a data configuration diagram of a system setting file 47 that stores data necessary for operating the system of FIG. 1. FIG. 5 is a data structure diagram of a question trend information temporary storage file 48, which is a file that is temporarily saved when a question tendency is determined from registered problem data in learning.

  Hereinafter, an embodiment of a learning system to which the present invention is applied will be described.

FIG. 1 is a system configuration diagram showing an embodiment of the present invention. In the present specification, elements displayed as XX part and XX means indicate that a CPU of a computer reads a necessary program and executes processing.
This system includes a data processing unit 1 that mainly performs processing at the time of learning, an input device 2 that is a means for a learner to input a correct answer to the system, processing results in the data processing unit, and learner input Is output to the outside and a data storage unit 4 for storing the problem data for learning and the learning result of the learner, which are physically connected as shown in the figure.
In this configuration example, everything is physically connected, but when implementing this embodiment, even if a network is used to exchange each function and storage area data, etc. May be used, or each function may be distributed.

The data processor 1 has a weak point analysis problem extraction unit 11 that outputs a test (hereinafter referred to as “weak point analysis problem”) for measuring the learner's learning ability at the start of learning, and the learner has solved. A weak point analysis result calculation unit 12 for scoring the result of the weak point analysis problem and judging the ability; a question tendency analysis processing unit 13 for analyzing the question tendency based on the problem data registered in the system after storing the result of the weak point analysis; An achievement rate increase value calculation unit 14 that calculates an increase rate of achievement rate per question of each classification based on the analyzed question tendency, and weak point learning problem extraction that automatically extracts questions that are given according to the state of the learner And a weak point learning result calculation unit 16 for scoring the result of the weak point learning problem solved by the learner and determining the ability.
The input device 2 includes a mouse, a keyboard, a touch panel, and the like, which are devices suitable for inputting a learner's answer to the system.
The output device 3 includes a display, a projector, and the like, which are devices suitable for displaying the result processed by the data processing unit for the learner.

  The data storage unit 4 records a problem data storage area 41 for recording problems held by the system, a classification data storage area 42 for recording classification data registered in the system, and a learner's name and ID. Learner data storage area 43, learning result data storage area 44 that records the situation of each class of learners, learning history storage area 45 that records the degree of achievement of the learner for each learning of the learner, and learning Answer data storage area 46 for recording the answer results at the time, a system setting file 47 for recording the number of learnings and the evaluation criteria of the learning situation, and a question trend information temporary storage file for temporarily recording the question tendency 48.

  Specific data examples of these data storage units 4 have contents as shown in FIGS. The contents of the data storage unit 4 are expressed as a storage area or a file. However, if the data can be stored as shown in the figure, there is no problem even if other storage methods such as a database are used. There is no.

  As shown in FIG. 6, the problem data storage area 41 in the data recording unit 4 has a problem classification ID 601, problem ID 602, difficulty 603, problem sentence 604, correct answer 605, explanation sentence 606, type 607, and learning performed at the beginning of learning. It is composed of a weak point analysis flag 608 for determining whether or not to give a question as a weak point analysis of a person, and a problem date 609, and is registered so as to be uniquely identified by a classification ID and a problem ID. Among these data, data having a weak point analysis flag 608 of “1” is data set as a weak point analysis test. In order to identify whether or not the problem data is an actual test question, a question date 609 is set. If this item is set, it is an actual test question, and if not, it is not a test question. Identified.

As shown in FIG. 7, the classification data storage area 42 includes a classification ID 701 and a classification name 702 of the classification ID, and the classification ID of the problem data is registered as information that links the classification data storage area 42.
The learner data storage area 43 is an area for holding data of learners who use the system as shown in FIG. 8, and includes learner IDs 801, surnames 802, first names 803, and learning progress 804 of learners. It is uniquely identified by ID801. The learning progress 804 is an item for storing the current number of times of learning, and is an item that is updated every time after completion of learning.

The learning result data storage area 44 is an area for holding the current learning result in each classification for each learner. As shown in FIG. 9, the learner ID 901, the classification ID 902, the achievement rate increase value 903, the achievement rate 904, and the achievement The degree 905 is uniquely identified by the learner ID 901 and the classification ID 902. Among these items, the achievement rate 904 is a value for recording a learner's level of proficiency as an internal value of the system, and is a value that increases as learning progresses. The achievement rate 904 increases by a certain numerical value only in the relationship between the number of questions and the number of correct answers in the initial state confirmation test, and is set to the achievement rate increase value 903 of the corresponding field every time one question is correctly answered in subsequent repeated learning. To rise. The achievement rate increase value 903 represents the value of the achievement rate that increases when one question is correctly answered during weak point learning. The achievement rate increase value 903 is an item to be calculated based on the number of registered exam questions, that is, the test frequency of the tests up to the present, and the increasing value is calculated based on the test frequency. Learning can be performed with the frequency added to the viewpoint. The achievement level 905 is a value calculated using an achievement rate 904 based on a value of the system setting file 45 described later, and is a value indicating the level of the learner.
The learning result data is initially set at the beginning of learning using this system, and the achievement rate 904 and the achievement level 905 are updated each time learning is repeated. The history data of the individual learner is uniquely identified by the learner ID 901 as in the range 906.

  The learning history data storage area 45 is an area for storing a result of learning performed by the learner, and includes a learner ID 1001, a learning count 1002, a classification ID 1003, and an achievement degree 1004 as shown in FIG. This data is added each time the learner learns, and one learning result is uniquely identified by the number of learning times 1002 as in the range 1005. Further, the entire history data of the learner is uniquely identified by the learner ID 1001 as in the range of 1006.

  As shown in FIG. 11, the answer data storage area 46 includes a learner ID 1101, a learning count 1102, a classification ID 1103, a question ID 1104, a question number 1105, a learner's answer 1106 for the question, and a correct / incorrect 1107 for the answer. The answer data is uniquely identified by a learner ID 1101, a learning count 1102, a classification ID 1103, and a problem ID 1104. The answer data in each learning is identified by whether the number of learnings 1102 is the same as in the range 1108, and the question number is identified by the question number 1105. The entire answer of the learner is identified by the learner ID 1101 as in the range 1109.

  As shown in FIG. 12, the system setting file 47 is a file that holds parameters necessary for calculation during learning. The system setting file 47 stores a learning possible number of times 1201 for storing the number of times of repeated learning, and per learning when performing repeated learning. The number of questions for one learning 1202 that stores the number of questions of the learning, the upper limit 1203 for increasing the achievement rate in one learning, and the upper limit 1203 for increasing the achievement rate in one learning, Among them, an assumed achievement rate maximum value 1204 for saving the assumed maximum value of the achievement rate 904, a weak point analysis achievement rate 1205 in which correspondence between the number of questions and the number of correct answers in the initial state confirmation test is saved, and a maximum value of the achievement degree 905 Is stored, and the achievement degree corresponding value 1207 for recording the correspondence between the achievement degree maximum value 1206 and the achievement ratio 904 for determining the achievement degree 905.

As shown in FIG. 13, the question trend information temporary recording file 48 includes a classification ID 1301, a problem number 1302 currently registered, and a problem number ratio calculated based on the number of problems registered in the problem data storage area 41. 1303 and a question tendency coefficient 1304 used in the calculation reflecting the question tendency in the system. The question trend information temporary recording file is generated for temporary storage when the learning result data storage area 44 is initialized after the weak point analysis.
As points to note when developing this system, the number of problems registered with the problem date 609 stored in the problem data storage area 41, that is, the number of data that are test questions, is included in each category. One question must exist. (Because it is a classification where there is no problem, it is contrary to the direction of the system that reinforces the weak point for each classification)
In addition, it is necessary that data having the weak point analysis flag 608 of “1” is not registered more than three questions for each classification. This is because, as shown in the system setting file 47, the number of problems in each classification in the weak point analysis is not assumed to be 3 or more.

According to the learning system of the present embodiment, the following effects can be expected.
(1) It is possible to evaluate weak points in consideration of the frequency of questions by determining the increase value of the achievement rate, which is the standard for judging the state of the learner, based on the number of registered exam questions. It becomes like this.
(2) Since the achievement rate is calculated based on the number of registered problems, weak points can be automatically evaluated in consideration of the frequency of questions even when the number of problems increases or decreases.

  Hereinafter, the flow of processing in the embodiment configured as described above will be described.

FIG. 2 is a flow of learning in the embodiment.
First, the learner data of this system is registered in the learner data storage area 43 of this system. (Step 21)
After registering the learner data, the system uses the weak point analysis problem extraction unit 11 to extract the weak point analysis problem in order to evaluate the learner's ability. The problems extracted here are those registered in the problem data storage area 41 for which the weak point analysis flag 608 is “1”. At this time, problem numbers are assigned in order from the top of the extraction result. (Step 22)
The extracted problem is displayed to the learner by the output device 3. The learner who sees the displayed problem inputs an answer to the system using the input device 2.
When the learner inputs a scoring instruction to the system after inputting the answers to all the questions, the system starts the initial setting of the learner's data by using the scoring of the weak point analysis question and the result. (Step 23)

The details of the scoring and learner data initial setting process are as shown in FIG.
In FIG. 3, from 301 to 305, the result of weak point analysis is calculated.
First, the system setting file 47 is read. (Step 301)
Next, the weak point analysis result calculation unit 12 performs correct / incorrect determination on the learner's data input by the input device 2 based on the answer 605 in the question data storage area. The determination is made in the order of the problem numbers given in step 22. The result of correct / incorrect determination is recorded in the answer data storage area 46. At the time of recording, “0” learning is performed so that the item of learning count 1102 can be determined to be a record of weak point analysis. (Step 302)
Thereafter, based on the classification data stored in the classification data storage area 42, the number of questions and the number of correct answers in each classification in the weak point analysis test are totalized. (Step 303)
After the number of correct answers for each category is completed, the initial achievement rate for each category is calculated based on the number of correct answers. The initial achievement rate is set based on the value 1205 stored in the system setting file 47. For example, the initial achievement rate when the number of correct answers in the classification of three questions in the weak point analysis is 2 is “25” set to 1205-n−2. Perform this procedure for all classifications to determine the initial achievement rate. (Step 304)

Next, the achievement level is calculated from the previously determined achievement rate. The achievement level is calculated using the achievement level corresponding value 1207 stored in the system setting file 47. In the setting, when the achievement rate is less than the value of 1207-a that is the first value of the achievement rate achievement rate corresponding value 1207, the achievement rate is 0, and thereafter, the value is equal to or greater than the value of 1207-a and less than the value of 1207-b. If there is a degree of achievement 1, a value of 1207-b or more and less than 1207-c, the degree of achievement 2 is judged. If the last degree of achievement is a value of 1207-n or more, the final achievement degree is obtained. It becomes 5. In the case of this example, when the achievement rate of a certain classification is 20, it is 15 or more and less than 30, so the achievement level is judged as 1. (Step 305)
From 306 to 309, the achievement rate increase value is calculated in consideration of the question tendency and stored in the learning result recording area.

First, the question tendency analysis unit 13 analyzes the question tendency.
The problem data in which the problem date 609 exists is extracted from the problem data registered in the problem data storage area. As described above, since all questions having the problem date 609 of the problem data are registered as previous test questions, only the test questions are extracted here. In this case, it is also possible to allow the learner to input a date and target only questions that have been given as a test during a predetermined period. (Step 306)
The extracted problems are tabulated for each classification ID 601 and the number of test questions for each classification is calculated. In this way, the number of test questions for each category registered in the system can be tabulated. (Step 307)

The necessary information is stored in the question trend information temporary storage file 48 using the number of questions thus tabulated. At the time of saving, the classification ID 1301 and the number of problems 1302 are saved as they are in step 307. The question number ratio 1303 stores “number of questions 1302 for each class” ÷ “total number of test questions extracted in step 306”.
The question tendency coefficient 1304 is obtained by first obtaining “the number of possible learnings 1201” × “the number of questions 1202 once learned” + “twice the number of questions presented in the weakness analysis” in the system setting file read in step 301. The total number of questions that can be answered in a series of studies. The reason why the number of questions presented in the weak point analysis is doubled is that the increase in the achievement rate by the weak point analysis in this system is a fixed value, which is larger than in normal learning. Then, the product of the total number of questions that can be answered and the question number ratio 1303 is calculated to obtain a question tendency coefficient 1304. The meaning of the question tendency coefficient is an approximate number that indicates how much the classification problem can be solved in a series of learning. (Step 308)
Then, the achievement rate increase value calculation unit 14 is used to calculate the achievement rate increase value of each classification based on the data stored in the question trend information temporary storage file 48. This calculation is obtained by “maximum assumed achievement rate 1204” ÷ “question tendency coefficient 1304”. (Step 309)

Finally, all the values obtained in steps 301 to 309 are stored in the learning result storage area 44, and the initial setting of the scoring and learning result recording file is completed.
After scoring the weak point analysis and initial setting of the learner's data, a problem for repeated learning called weak point learning is extracted and the learner is made to learn. (Step 24)

The problem extraction in step 24 is performed in the weak point learning problem extraction unit 15 along the flow of FIG.
First, learner data in the system setting file 47 and the learning result data storage area 44 are read as data necessary for problem extraction. (Step 401)
Then, the lowest achievement level among the achievement levels for each classification registered in the read learning result data storage area 44 is checked, and the number of the lowest achievement levels is stored. (Step 402)
Calculate how many problems you plan to extract in each category of the lowest achievement you have examined. The calculation is obtained by rounding up the decimals of “the number of problems 1202 per learning in the system setting file 47” ÷ “the number of classifications of the minimum achievement level”. However, be sure to have at least one question. For example, when the number of questions 1202 per learning is 20 and the number of classifications of the minimum achievement level is 4, it is assumed that 5 questions are extracted for each classification. (Step 403)
Then, the classification with the lowest achievement rate among the minimum achievement levels is examined, and the number of problems is determined from this classification. (Step 404)
When confirming the number of problems for each category examined, the number of decimals of “achievement rate increase upper limit value 1203 for one learning in system setting file 47” ÷ “achievement rate increase value 903 of learning result data storage area 44” is calculated. A value obtained from the rounded-up result is calculated, and the smaller of this value and step 403 is extracted as the number of problems to be extracted. For example, when the achievement rate increase upper limit 1203 is 20 and the achievement rate increase value 903 is 12, it is limited not to extract two or more questions, and five questions are scheduled to be extracted in step 403. Even in this case, only two questions will be extracted for this classification. The total number of extracted questions is adjusted so as not to exceed the number of questions 1202 per learning.
By doing this, many categories with large achievement rate increase values are extracted, and a large number of achievement rates are prevented from increasing at once. (Step 405)

When the number of problems in the first classification is determined, the total number of problems for which extraction has been determined is examined. If this is not the number of questions 1202 per learning, the classification of the next lowest achievement rate is examined in the learning result data storage area. If the classification is the same as the previous achievement level, the procedure from step 404 is performed. (Steps 406 → 407 → 408 → 404) In the procedure from step 404, the number of problems to be extracted is the same degree of achievement, so the number of problems is determined as the same value as before. If the degree of achievement is different, the procedure from step 402 is performed based on the next degree of achievement. (Step 406 → 407 → 409 → 402) In this case, since the degree of achievement is different from the previous one, the number of scheduled extraction problems is recalculated. In this case, “the number of problems per learning of the system setting file 47” The number of problems is determined by “difference between 1202 and the number of problems that have been extracted” ÷ “the number of classifications of the minimum achievement”
If the total number of questions for which extraction has been determined exceeds the number of questions 1202 for one learning, or if the total number of problems for which extraction has been determined in the above-described repetition exceeds the number of problems for which extraction has been determined, the process goes through the repetition and proceeds to step 410.
After the repetition, the problem data is extracted for each category from the problem data storage area according to the determined number of extracted problems for each category, and is used as a weak point learning problem. (Step 410)
After extracting the problem, the extracted problem is displayed on the output device 3 as in the weak point analysis, and the learner's answer is input by the input device 2.
Thereafter, the data of the learner of the weak point learning is updated. (Step 25)

The scoring and updating of the learner's data are performed in the procedure of FIG.
First, the system setting file 47 is read, and then learner data is read from the learner data storage area 43 and the learning result data storage area 44. (Step 501)
Thereafter, the correctness / incorrectness of the input answer is determined based on the answer 605 in the question data storage area 41. After the correct / incorrect determination is completed, the result is stored in the answer data storage area 46. The learning count at the time of storage is a value obtained by adding 1 to the value of the learning progress 804 in the learner data storage area 43. Then, the stored scoring results are totaled for each category, and the number of correct answers for each category is examined. (Step 502, Step 503)

  Finally, the learning result storage area 44 is updated by calculating the achievement rate / achievement degree of each classification after learning based on the number of correct answers of each classification. The achievement rate after completion of the learning is obtained by “the achievement rate so far” + “the number of correct answers in the current learning” × “achievement rate increase value 903” for each classification. The achievement level is determined according to the correspondence of the achievement level achievement value corresponding value 1207 of the system setting file 47 based on the achievement rate calculated previously. Further, the learning progress 804 in the learner data storage area 43 is incremented by 1, and the learning is performed once. Further, the current learning result is stored in the learning history storage area 45 by using the achievement level of the current learning result.

  After updating the scoring and learner data, it is checked whether the learning progress 804 added earlier exceeds the learning possible number of times 1201 in the system setting file. If not, the weak point learning is performed again. A series of learning is finished.

  It can be used not only for a stand-alone system but also for a system that distributes processing via a network. It can be used in a system that implements many processes on the server side provided on the web and provides services to users accessing the Internet.

DESCRIPTION OF SYMBOLS 1 Data processing part 2 Input device 3 Output device 4 Data storage part 11 Weak point analysis problem extraction part 12 Weak point analysis result calculation part 13 Question tendency analysis processing part 14 Achievement rate increase value calculation part 15 Weak point learning problem extraction part 16 Weak point learning result calculation Part

Claims (2)

Problem data that stores the problem , correct answer and explanation , classification of the problem, and problem date indicating that the problem is a test problem that has been given in the previous main examination, and the learning status and achievement rate of each classification in the learner And a learning result recording data that stores the achievement level, and a system setting file that holds parameters necessary for calculation during learning ,
Extract the problems for the learner, an e-learning system with a data processing unit to be stored in the recording apparatus calculates the learning result,
(A) The system setting file is:
(A1) An upper limit of increase in achievement rate for one learning that sets an upper limit for increasing the achievement rate in one learning;
(A2) an assumed achievement rate maximum value for setting an assumed maximum value of the achievement rate in iterative learning;
(A3) holding the achievement degree achievement rate corresponding value in which the correspondence between the achievement degree and the range of the achievement percentage is set,
(B) The data processing device includes:
(B1) A question trend analysis processing unit that analyzes a question tendency based on the problem data, extracts all test questions with the question date with reference to the problem data, and extracts the extracted test questions for each classification Each class is a product of the means for calculating the ratio of the number of test questions in each category to the total number of test questions extracted and extracted, and the product of the total number of questions that can be answered in repeated learning and the ratio of the number of test questions in each class A question tendency analysis processing unit having means for calculating a question tendency tendency coefficient of
(B2) An achievement rate increase value calculation unit that calculates an increase rate of achievement rate per question of each classification based on the question tendency analyzed by the question trend analysis processing unit, and the assumption that the system setting file holds The achievement rate increase value calculation unit having means for calculating the achievement rate increase value of each classification, which is a value obtained by dividing the achievement rate maximum value by the question tendency coefficient of each classification;
(B3) A weak point learning problem extraction unit that extracts a question to be given to the learner based on the achievement rate increase value calculated by the achievement rate increase value calculation unit, wherein the achievement level included in the learning result recording data is By referencing, all the classifications of the minimum achievement level are extracted, and the learning 1 held in the system setting file in order from the classification of the lowest achievement level to the lowest achievement rate among the extracted classifications of the minimum achievement level Means for determining the number of problems to be extracted based on a value obtained by dividing the upper limit of the achievement rate increase of the number of times by the increase value of the achievement rate of the classification, and the problem for weak point learning of each classification according to the determined number of problems Means for extracting from the data, the weak point learning problem extraction unit having,
(B4) A weakness learning result calculation unit for scoring the answer to the weakness learning problem by the learner and judging the ability, the correct answer number in the classification, and the achievement rate increase value of the classification held in the system setting file The means for updating the achievement rate of each classification included in the learning result record data by adding the product of, and the achievement degree achievement rate corresponding value held in the updated achievement rate and the system setting file An e-learning system comprising: a weak point learning result calculation unit including: means for updating the achievement level of each classification included in the learning result recording data . An e-learning system according to claim 1,
The problem data stored in the recording device further includes a weak point analysis flag for determining whether to give a question as a weak point analysis problem at the start of learning,
(A) The system setting file is:
(A4) holding an initial achievement rate in which the correspondence between the number of the weak point analysis problems at the start of learning and the number of correct answers is set in advance;
(B) The data processing device includes:
(B5) a weak point analysis problem extracting unit that extracts problem data having the weak point analysis flag of 1 to measure a learner's learning ability at the start of learning, and outputs the problem data as a weak point analysis problem;
(B6) A weak point analysis result calculation unit that determines the ability by scoring the result of the learner solving the weak point analysis problem that has been output, and totalizes the number of correct answers for each classification for the weak point analysis problem Means for recording the initial achievement rate of each classification of the learner in the learning result recording data based on the result and the initial achievement ratio held in the system setting file, the initial achievement ratio of each recorded classification, and the system setting file The weak point analysis result calculation unit comprising: means for recording the initial achievement degree of each classification in the learning result record data based on the achievement degree achievement rate correspondence value held by -learning system.

Images