JPH0627868A - Cai system - Google Patents

Abstract

PURPOSE:To provide the suitable CAI system which changes teaching materials to be selected at the time of learning between a learner who has some knowledge about the teaching materials and a complete beginner and selects problems in consideration of the intention of the learners. CONSTITUTION:A teaching material selection part 15 selects the teaching materials according to learning information on a learner model 12 and rules in a teaching material selection rule base 13. A teaching material execution part 16 selects the teaching materials in a teaching material file 11 and outputs and displays them, and then accepts the input of learner's answers. A diagnostic part 17 diagnoses the learner's answers and updates the learner model 12. A self-declaration part 18 inputs learner's knowledge and the intention of learning and performs inference by referring to a model generation rule base 14 to update the learner model 12 with the obtained learning information. The learner can efficiently learn necessary knowledge by self-delcaring the learner's previous knowledge or the correction by the self-declaration.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a CAI system, and more particularly to a CAI system for selecting teaching materials based on a learner model.

[0002]

2. Description of the Related Art Conventional teaching materials for a CAI system are selected from learning result data of a learner corresponding to a learner model and learning of a learner as in Japanese Patent Application No. 1-17998 and "Human Interface". The state data and the input are used to refer to the rule data in the teaching material selection rule file to select the teaching material to be provided next.

[0003]

In the above-mentioned prior art, since the learner model is determined by solving the problem (execution result of the teaching material), regardless of the learner's ability before the start of learning by the system, I had to start with empty grades and learning. For this reason, there is a problem that a learner who has knowledge needs to solve a familiar task and wastes time. During learning, the next task was decided based on the contents of the learner model of the system, regardless of the learner's intention. Therefore, it is not possible for the learner to determine the next appropriate task for the learner, such as a task for specialized learning or a task for intensive learning.
There is a problem. In view of the above circumstances, the purpose of the present invention is to
Change the learning material selected at the time of learning between a learner who has some knowledge of the teaching material and a completely beginner, and
CAI suitable for selecting a task considering the learner's intention
To provide a system.

[0004]

[Means for Solving the Problems] A teaching material file that stores teaching materials related to learning items, a learner model that stores learning information related to a learner's understanding of learning items, and for selecting teaching materials based on the learner model A storage device having a teaching material selection rule base that stores the rules, and a teaching material selection unit that selects a teaching material to be provided by referring to the learning material selection rule base corresponding to a learning state from a learner model,
The selected teaching material is output to the output device, and the teaching material execution unit that inputs the answer corresponding to the output from the input device, and the processing device that has the diagnosis unit that diagnoses the input answer and updates the learner model In the CAI system, the learner model stores learning information about a self-report of a learner, and a model creation rule base that stores rules for inferring learning information stored in the learner model from self-report, and learning. A self-declaration unit is provided which inputs learning information regarding a self-report of a learner from an input device and outputs learning information to a learner model by referring to a model creation rule base in response to the input.

[0005]

[Function] At the start of learning, the learner's preliminary knowledge or information about correction by self-report can be input to the self-report section, the learner model can be rewritten based on the information, and the teaching material to be given at the start of learning can be changed. Furthermore, during learning, information about the degree of understanding of the knowledge and learning items or the learner's intention is input from the self-reporting section, the learner model is rewritten based on this input, and the learning material considering the learner's intention is selected. .

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a functional configuration diagram of an embodiment of the present invention. First, the hardware configuration of this embodiment will be described. The processing device 21 executes a program that supports learning by a computer. The storage device 22 stores the contents of the teaching material, learning state data, and the like. The input device 23 is a keyboard, a mouse or the like for inputting the learning result of the learner. Display device 2
Reference numeral 4 is a display device. The processing device 21 operates according to a built-in program, receives an input from the input device 23, performs processing for learning using the information stored in the storage device 22, and outputs the result to the display device 24. Each component will be described. In the processing device 21, the learning material selection 15, the learning material execution 16, the diagnosis 17, and the self-report 18 are executed. The storage device 22 stores a teaching material file 11, a learner model 12, a learning material selection rule base 13, and a model creation rule base 14. The teaching material file 11 contains a plurality of teaching materials used for learning. The learner model 12 contains learning state data of the learner. The learning material selection rule base 13 contains rule data for selecting a learning material to be provided according to the learning state of the learner. The model creation rule base 14 contains rule data for rewriting the learner model by input from the learner. The learning material selection unit 15 searches the learning state data of the learner model 12 and the rule data of the selection rule base 13 to select a learning material number, and generates guidance suitable for the learning material. The teaching material executing unit 16 extracts the teaching material from the teaching material file 11 based on the selection by the teaching material selecting unit 15 and provides it to the learner.
The diagnosis unit 17 diagnoses the learner's answer input in the teaching material execution unit 16 and updates the data of the learner model 12 with the learning state data obtained as a result. The self-report section 18 inputs the learner's preliminary knowledge and learning intentions, infers using the model creation rule base 14, and updates the learner model 12 with the obtained data. An example of the teaching material file 11 used in the present embodiment is shown in FIG. 2, an example of the learner model 12 is shown in FIG. 3, an example of the teaching material selection rule base 13 is shown in FIG. 4, and an example of the model creation rule base 14 is shown in FIG. Shown respectively. In this embodiment, it is assumed that C language is learned. The teaching material file 11 is composed of a table 21 in which a set of a teaching material number 21a and a practice question 21b related to C language is stored. The learner model 12 includes a learning item 31a, a table 31 storing a score 31b indicating the degree of comprehension for each learning problem, and a learner's desire 32 for the item to be learned. At the start of learning, the score 31b contains zero. The learner's request 32 is empty of data. The learning material selection rule base 13 is a learning material number 41a.
And a table 4 storing learning items 41b related to each question
1 and a group of rules 42 and below for selecting a teaching material. For example, in the rule group 43, when the “data type” of the related learning item 41b is lower than the score of 0.5 or the “function” is lower than the score of 0.5, the teaching material number (question) 1 is set. select. Modeling rule base 1
4 is a learning item 5 related to the preliminary knowledge 51a (if section)
1b and a table 51 storing a score 51c (then part) indicating the degree of comprehension.

Next, the processing executed in the processing device 21 will be described with reference to FIG. Details will be described later for each step. First, a learning material is selected based on the learner model 12 and the learning material selection rule base 13. Guidance is generated for the selected teaching material and output to the display device 24 (step 15). Next, either inputting the input device 23 to continue learning, interrupt learning, or terminate learning is input (step 61). Step 1 if you want to continue learning
Perform steps 6 and 17. First, the teaching material selected in step 15 is searched from the teaching material file 11 and displayed on the display device 24. The learner inputs the answer using the input device 23 (step 16). The inputted answer is diagnosed, and the score stored in the learner model 12 is updated with the data obtained as a result (step 17). When learning is interrupted in step 61, step 18 is executed. First, the learner model 12 is output to the display device 24. Then, the learner's self-report is accepted from the input device 23. The score stored in the learner model 12 is updated with the resulting data (step 18).

The process of each step of FIG. 6 will be described in detail with reference to FIGS. 7 to 10. First, the processing of the teaching material selection 15 is shown in FIG.
Will be explained. First, the number of the teaching material is selected based on the learning state data of the learner model 12 and the rule of the teaching material selection rule base 13 (step 71). For example, it is assumed that the selected learning material number is Q3 of the table 41. From the tables 31 and 41, the learning state is searched for the learning item related to the selected teaching material (step 72). For example, the learning item related to the selected teaching material is "data type",
They are "functions" and "pointers", and their learning states are points 1.0, 0.4, and 0.8. It is divided into two groups according to the level of the score (step 73). Items with high scores are group A, and items with low scores are group B.
Here, the group A has a high degree of comprehension and the group B has a low degree of comprehension. The next step is determined by comparing the learning item desired by the learner with the learning item of the selected teaching material (step 74). The guidance shown in step 75 is output to the display device 24 if the learner's desired learning item is not among the learning items of the selected teaching material, step 76 if it is in the A group, and step 77 if it is in the B group. . The processing of the teaching material execution 16 will be described with reference to FIG. The learning material number and the learning material file 11 selected by the learning material selection unit 15
Search for teaching materials from (step 81). The teaching material is output to the display device 21 (step 82). Learner input device 2
The answer to the teaching material is input to 3 (step 83). Next, the processing of the diagnosis 17 will be described with reference to FIG. The result of the learner's execution (answer) is diagnosed and the degree of understanding of the learner is calculated (step 91). The learning state data of the learner model 12 is updated (step 92). Next, the processing of the self-report 18 will be described with reference to FIG. First, the current learner model 12 is output to the display device 24 (step 101).
It is confirmed whether or not there is an input of prior knowledge or correction by self-report (step 103), and if there is input of prior knowledge, data is analyzed using fuzzy inference to construct a learner model (step 102). Details of step 102 will be described later. If the changed portion is input, the correction content is input from the input device 24 (step 104). Then, the understanding level (table 31) of the learner model 12 is updated based on the input (step 105). For example, when the learner recognizes that the degree of understanding is higher than the score 0.4 of the displayed learning item “function” when the table 31 is displayed on the display device 24, the score is 0.7. To fix. Then, the understanding level (table 31) of the learner model 12 is updated to a score of 0.7. If the changed portion is not input, the process proceeds to step 106. Next, check if there is any input for the learning item you want to study (step 106), and if not, self-report 1
8 is completed. If the desired item is input, the desired item is input from the input device 23 (step 107). The table 32 of the learner model 12 is updated based on the input (step 108).

The process of self-report 18 and step 102 will be described in detail with reference to FIG. First, a screen for inputting the learner's preliminary knowledge is output (step 111). The learner inputs his / her knowledge on the screen (step 112). The degree of prior knowledge is calculated from the learner's input (step 11
3). Obtain a learner model using fuzzy reasoning. First, the fitness of the if part of each rule is calculated (step 11
4) The comprehension level of the learning item in the then part of each rule is calculated (step 115). Next, the maximum value of the degree of understanding of each learning item is obtained (step 116). Learner model 12
Of the data is changed (step 117).

An example of execution of this embodiment will be described with reference to the screen example of FIG. First, the learning material is selected based on the learner model 12 and the learning material selection rule base 13 (step 7).
1). Now, since the scores of all the items of the learner model 12 are zero, the question 1 is selected from the rule data 43 of the teaching material selection rule base 13. Guidance is generated with reference to the learner model 12 for the selected teaching material (step 72).
~ 77). Now, the guidance (141) is generated and output. From the input device 23, input is made as to whether to continue the answer, interrupt the answer, or end the learning (step 61). Now, ok (142) is input as the answer is continued. The content of the selected teaching material Q1 is searched from the teaching material file 11 (step 81), and the content (143) is output to the display device 24 (step 82). The learner inputs the answer from the input device 23 (step 83). When the creation of the answer is completed, the diagnosis unit 17 diagnoses the execution result (step 91), and the learner model 12 is updated with the data obtained from the diagnosis result (step 92). Next, returning to step 71, the teaching material is selected based on the new learner model 12. Now, question 3 is selected and guidance (144) is output. In step 61, if the answer to the teaching material selected by the system is interrupted, n
When o (145) is input, the self-report section 18 is activated. First, the learner model 12 is output (step 1
01). The learner refers to the learner model 12 that is output and makes a self-report from the input device 23 (step 1).
03-108). Input whether to self-report the question (146). Assuming that the correction is to be performed now, the correction (147) is selected and input. Steps 104 and 10
In step 5, the contents of the learner model 12 are input as shown in FIG. 4 (147), and in steps 107 and 108, “array” (148) is input in the desired learning item. Next, returning to step 71 once again, the teaching material is selected based on the new learner model 12. Now, with reference to the table 41 by the rule data 42 of the teaching material selection rule base 13, the teaching material regarding “array” is selected. In this embodiment, question 2 is selected in consideration of the learner's intention.

Next, an example will be shown in which the preliminary knowledge is input in step 102. FIG. 12 is a screen for inputting preliminary knowledge, which is output in step 111, and FIG. 13 shows an example of comprehension level calculation of learning items using fuzzy inference. The rule 134 of FIG. 13 is a rule stored in the model creation rule base 14. The calculated value 135 is a calculation result using fuzzy inference. First, a screen for inputting the learner's preliminary knowledge (Fig. 12)
Is output (step 111). The learner moves the bar 123 to an appropriate position with the mouse according to the number of steps and the period of the language etc. which have been used (step 112). For example, 121 is FORTRAN slightly less than 3Kstep or 6
It shows that it has been developed for less than a month.
It indicates that it has been developed over Kstep or over a year. It is assumed that the learner inputs his / her knowledge as shown in FIG. Next, the degree of understanding of the learning items is calculated using fuzzy reasoning. First, the degree of the preliminary knowledge 131 is converted into a numerical value (step 113), and from this, the degree of rule conformance 13
2 is calculated (step 114). The degree of the two pieces of background knowledge 131 of rule 1 is calculated by the calculated value 135 of the background knowledge 131.
As shown in, the rule conformance is 1 and 0.7 and 0.4.
32 takes 0.4 and becomes 0.4. In rule 2, the degrees of the two pieces of preliminary knowledge 131 are both 1.0, and the rule matching degree 132 is 1.0. Next, the comprehension level 133 of each learning item and the rule conformance level 132 for each rule determine the calculated value 135 of the understanding level 133 for each learning item (step 115). Here, the comprehension level of Rule 1 regarding the “data type” is calculated as 0.7 × 0.4 = 0.28. The comprehension of rule 2 is calculated as 0.2 × 1.0 = 0.2. Further, the comprehension level of Rule 1 regarding the “function” is calculated to be 0.32, and the comprehension level of Rule 2 is calculated to be 0.4.
Maximum value of the calculated value thus obtained for each learning item 1
When 36 is larger than the understanding level stored in the current learner model, the value is the understanding level of each learning item (step 116). For example, in the “data type”, the max of 0.28 and 0.20 is 0.28. Normal intellectual CAI
The system does not consider the learner's wishes in the selection of teaching materials.
In the present embodiment, as described above, it is possible to select the learning material most suitable for the learner from the learning materials related to the items desired by the learner. Further, the learner model can be changed by reflecting the learner's preliminary knowledge. In addition, in the usual intelligent CAI system, even a person who has some knowledge of the contents of the teaching material started learning from Q1. In this embodiment, by providing the self-reporting section, the data at the start of learning is changed so that the learner who has the knowledge as shown in FIG. be able to.

[0012]

According to the present invention, it is possible to efficiently learn the knowledge necessary for the learner by self-reporting the learner's preliminary knowledge or the self-reported modification. In particular, learners who have knowledge can effectively use the time required for learning,
For example, it is possible to select a task suitable for the learner, such as a task for specialized learning or a task for intensive learning.

[Brief description of drawings]

FIG. 1 is a functional structure diagram of an embodiment of a CAI system according to the present invention.

FIG. 2 is a diagram showing a teaching material file used in this embodiment.

FIG. 3 is a diagram showing a learner model used in this embodiment.

FIG. 4 is a diagram showing a teaching material selection rule base used in this embodiment.

FIG. 5 is a diagram showing a model creation rule base used in this embodiment.

FIG. 6 is a flowchart illustrating the operation of the embodiment.

FIG. 7 is a flowchart illustrating an operation of selecting a teaching material.

FIG. 8 is a flowchart illustrating an operation of executing a teaching material.

FIG. 9 is a flowchart illustrating a diagnosis operation.

FIG. 10 is a flowchart illustrating the self-reporting operation.

FIG. 11 is a flowchart illustrating an input operation of preliminary knowledge.

FIG. 12 is an example of an output screen used to input preliminary knowledge.

FIG. 13 is a table showing a calculation example of learning information.

FIG. 14 is an example of an output screen according to the embodiment.

[Explanation of symbols]

 11 Teaching Material File 12 Learner Model 13 Teaching Material Selection Rule Base 14 Model Creation Rule Base 15 Teaching Material Selection Section 16 Teaching Material Execution Section 17 Diagnostic Section 18 Self-Declaration Section 21 Processing Device 22 Storage Device 23 Input Device 24 Display Device

Claims (7)

[Claims] 1. A storage device for respectively storing a learning material concerning a learning item, learning information concerning a degree of understanding of a learner, and a rule for selecting a learning material, and a teaching material provided by referring to the rule corresponding to the learning information. In the CAI system including a processing device that selects, selects the answer input to the selected teaching material, and updates the learning information,
Accepting the learner's self-declaration and referring to the model-creation rule base corresponding to the input of the self-declaration and the model-creation rule base for storing rules for inferring learning information from the result of the self-declaration, and A CAI system characterized by having a self-reporting section for updating learning information. 2. The CAI system according to claim 1, wherein the self-declaration of the learner is correction based on his / her prior knowledge or self-declaration. 3. The self-assessment unit according to claim 1, including means for accepting an input from an input device, means for converting an input result according to a rule, and means for changing learning information based on the converted information. CAI system characterized by that. 4. The self-assessment section according to claim 1 or claim 3, outputs a screen for inputting the learner's preliminary knowledge,
The CAI is characterized in that the degree of prior knowledge is obtained from the learner's input, the degree of conformity of rules and the degree of understanding of each learning item are calculated, the maximum value of the degree of understanding of each learning item is obtained, and learning information is changed. system. 5. The CAI system according to claim 1, claim 3, or claim 4, wherein the processing result of the self-report section is displayed on a display device. 6. The CAI system according to claim 1, wherein the learning information includes learning state data for each learning item and desired learning item data. 7. The learning item according to claim 1, using the learning state data of the learning item corresponding to the selected teaching material, divides the learning item into a plurality of grooves based on a predetermined degree of comprehension, and outputs a learning message for each groove. CAI characterized by
system.