JP2872024B2 - Problem generator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a computer-based learning system (hereinafter referred to as CAI) in which a learner grasps a setting situation from a question sentence, selects an appropriate formula, and performs a mathematical processing operation. For the type of problem that leads to the solution,
The present invention relates to a problem generator having a function of generating various questions and their answers.

[0002]

[Prior art]

(1) In the CAI system, it is indispensable to present a problem and obtain a response from the learner in order to grasp the state of understanding of the learner. In particular, when trying to realize an individualized learning environment according to the learner's situation, it is necessary to have a function that can generate various problems such as problems created from various viewpoints and problems with many concrete examples Becomes

FIG. 11 is a diagram schematically showing a conventional problem generator 41. As shown in FIG. In this device 41, question data 4 including question sentences, question diagrams, answers, answer sentences, etc.
2 are separately prepared, and output a part of them when receiving a presentation instruction.

(2) As one of the learning modes of CAI, there is one called a drill & practice style.
In this mode, the CAI system performs automatic generation of questions, generation of answers, diagnosis of learner responses, control of jump destinations, and the like.

[0005]

The above prior arts (1) and (2) have the following problems or problems. According to the prior art (1), with regard to a sentence problem involving the operation of a mathematical expression, all items to be presented to the learner such as a problem sentence, an answer sentence, and a response to an incorrect answer are prepared in advance. I have. (See FIG. 11).

Therefore, even in the case of the same situation setting problem, it is necessary to separately prepare answer data for each of the variables to be obtained and specific examples, which is sufficient to enable the generation of the above-mentioned various problems. The creation and management of the data becomes extremely complicated. As a result, in reality, the types of questions that can be provided to learners are limited to a small number.

In the prior art (2), the function for automatically generating a question and an answer, which has been conventionally realized, uses a random number generation function to replace a numerical part in a question sentence or an expression, and the answer is also converted to data. Is calculated in accordance with the expression described as, and the numerical value portion is embedded and output. For this reason, the format of the problem is extremely limited, and the above-described various problem generations cannot be realized.

The present invention has been made in view of such circumstances, and solves the above-mentioned problems of the prior arts (1) and (2) and eliminates redundancy and abstracted data in a CAI system. An object of the present invention is to provide a problem generating apparatus having a function of generating various question sentences and answer sentences using the same.

[0009]

SUMMARY OF THE INVENTION According to the present invention, this object is achieved by: a. Question including variable list, relational expression, status statement template
Memory means for title data; b. Described the specific names and values of variables appearing in the problem
Memory means for problem specific data; c. Problem generation with desired and known variables
Depending on the directive, the above question type data and question specific data
Generate a question sentence that is a variable that determines the variable specified above.
A problem generating device for a learning system, which includes a computer and generates a problem title and an answer using a computer ; d . For the above relational expression, a variable determined by combining the operations of selecting the expression, eliminating redundant variables in the expression, eliminating unknown variables by integrating the expression, is expressed using known variables, and Solution deriving means for obtaining a value; e. Answer an outline of process in the solution deriving means
An answer sentence generating means for generating an answer sentence described in detail level based on the detail level designation data ;

[0010]

The question sentence generating means creates a sentence expressing the situation by applying the name given in the name list of the question specific data to the symbol display part of the situation sentence template of the question type data. The solution deriving means derives the value of the variable to be obtained by selecting an equation, deleting the variable, and the like. An outline of the process of deriving the solution is based on the data specifying the level of detail of the solution.
The answer sentence is generated by the answer sentence generating means as an answer sentence described in detail . As a result, various problem sentences and their answer sentences can be automatically generated and presented to the learner using data that is easy to create and manage.

[0011]

FIG. 1 is a functional block diagram schematically showing a CAI system using a problem generating apparatus according to the present invention, and FIG. 2 is a functional block diagram showing one embodiment of a problem generating apparatus according to the present invention.

In FIG. 1, reference numeral 10 denotes a CAI system, which includes memory means for storing data and operation programs,
It is composed of arithmetic means and input / output interfaces. FIG. 1 is a block diagram showing a function of the arithmetic means operating by software (operation program).

In FIG. 1, reference numeral 11 denotes a learning progress control device which controls generation of teaching materials to be presented to a learner and how to proceed with learning in accordance with an input from the learner, and 12 denotes an explanation generating device which generates learning materials for explaining learning contents. 1 is a problem generation device according to the present invention, 13
Is a presentation device that performs signal processing for visually and audibly presenting the output of the explanation generation device 12 and the problem generation device 1 to the learner, and 14 is a learning device that inputs answers, requests, questions, and the like from the learner. This is a learner input recognition device that recognizes the intention of the learner and outputs the result to the learning progress control device 11.

The problem generator 1 includes a learning progress controller 11
When a problem generation command that specifies the problem type required for generating a problem, the specific situation, and the variable to be sought is received,
A question sentence and a question diagram are generated and output to the presentation device 13. After that, a solution is derived and an answer sentence is generated, and the answer is output to the learning progress control device 11 for comparison with the learner's answer, and the answer sentence is presented when receiving the answer presentation command from the learning progress control device 11. Output to the device 13.

The problem generator 1 is configured as shown in FIG. In FIG. 2, reference numeral 4 denotes a memory means of question generation data comprising question type data 2 and question specific data 3, and reference numeral 5 denotes a question extraction data 4 extracted from the question generation data 4 based on designation of a question type and a specific situation. This is a question sentence generating means for generating a question sentence using data as a variable for obtaining a designated variable. 6 is a problem diagram generating means for generating a problem diagram in which symbols appearing in the problem sentence are written at appropriate positions in a diagram expressing the situation, and 7 is for selecting and manipulating relational expressions in the problem type data.
By doing so, a solution deriving means for expressing a variable to be obtained by a known variable, and an answer sentence generating means 8 for generating an outline of the process of the solution deriving means 7 as an answer sentence.

Here, the problem type is obtained by classifying specific individual problems based on an abstracted situation and a relational expression used to derive a solution. For example, in the unit of “universal gravitation” in high school physics, there is a problem type that combines universal gravitation and constant velocity circular motion. (This question type is used as an example in the following description). This problem type is concerned with a situation in which two objects are present and one is making a circular motion at constant velocity around the other with the gravitational force as a centripetal force.

In each problem, the name of the object and the numerical value representing the situation, variables to be obtained, combinations of formulas to be used, and the like are variously different. When defining a problem type, prepare the fact that it is a common situation at the above-mentioned abstract level, and prepare basic relational expressions (four in this case, see FIG. 3B described later). It is assumed that a solution can be derived by selecting an appropriate equation from and repeating the mathematical operation.

FIG. 3 is a diagram showing an example of the contents of the question type data 2. In FIG. 3A, reference numeral 20 denotes a situation sentence template used when generating a question sentence, and FIG. Show 2
Reference numeral 1 denotes a relational expression necessary and sufficient for the question type, and reference numeral 22 shown in FIG. 3C denotes a variable list describing variables appearing in the question sentence and the relational expression.

FIG. 4 is a diagram showing an example of the contents of the question specific data 3. In FIG. 4A, reference numeral 23 denotes a list of names, and FIG. This is a list of variable values that give specific values and units or comments.

The problem sentence generating means 5 applies a name given in the name list 23 to the symbol display section ([A] or [B]) of the situation sentence template 20 to generate a sentence expressing the situation. Next, the variables given in the variable value list 24 other than those specified as the variables to be obtained are treated as known variables, and sentences expressing specific setting conditions and what to be obtained are generated.

FIG. 5 is a diagram showing an output example of the question sentence generating means 5, and reference numerals 25 to 28 denote the question type data of FIG.
Is a question sentence generated as a variable for calculating Ra, Rh, g, and v from the question specific data. FIG. 2
Has a problem diagram generating means 6, which can be configured in the same manner as the question sentence generating means 5, and the operation thereof is also the same.

FIG. 6 shows components 71 to 7 of the solution deriving means 7.
FIG. 5 is a diagram for explaining an operation 5 and an operation; In the drawing, reference numeral 15 denotes solution derivation process data, which is composed of input data to the means 7 and output data of the process. 71 is a formula simplifying means for eliminating redundant variables in the formula, 72 is a first formula selecting means for selecting a formula including the variables to be obtained and having the least number of unknown variables, 73
Selects a formula that does not include the variable to be determined and that includes an unknown variable included in the formula after eliminating redundant variables from the output of the first formula selection means 72 by using the formula simplification means 71. An expression selecting unit 74 is an expression shaping unit that transforms the expression into the form of “specified variable =...”, And an expression integrating unit 75 integrates two expressions and eliminates unknown variables.

FIG. 7 is a diagram showing the flow of processing in the solution deriving means 7. Hereinafter, the operation of the solution deriving means 7 when dealing with the problem 26 in which Rh (the height of the satellite orbit) in FIG. 5 is determined will be described with reference to FIGS.

In step a of FIG. 7, redundant variables included in the relational expression are eliminated by using the expression simplifying means 71. In the expressions of the first and third lines of the relational expression 21 in FIG.
However, in the expression on the second line, m is treated as a redundant variable.
The processing result becomes a in FIG. 8 and is stored as a part of the solution derivation process data 15. It should be noted that the relational expressions are registered as question-type data in a form including redundant variables in order to present to the learner a process in which a solution is derived from the basic expressions.

At step b in FIG. 7, unknown variables including the variable Rh to be obtained from the equation shown in FIG. An expression having a small number of items (excluding variables appearing in the list 24) is selected as the expression F0. In this case, the equation T = 2 * π * (Ra + Rh) / v in the fourth row of FIG. 8A is selected.

In step c of FIG. 7, an unknown variable included in the equation F0 is selected. In this case, it is T. FIG.
In step d, an expression that includes the unknown variable T and does not include the variable Rh to be determined is selected from the expressions shown in FIG. 8A, and this expression is defined as F1 (step e). Absent in this case satisfies the condition expression F1 is returned to step b of Figure 7 along the branch of N _o step e.

In step b of FIG. 7, the number of unknown variables including the variable Rh to be calculated is targeted for three expressions other than the expression on the fourth line of FIG. Is newly selected as the expression F0. In this case, an expression is selected in the first row of FIG.

In step c of FIG. 7, the unknown variable included in the equation F0 is selected as the unknown variable α. here,
The unknown variables are Gr and Ma, but the order in which they appear in the variable list in FIG. 3 is, therefore, Ma is selected as the unknown variable α.

In step d of FIG. 7, the unknown variable α
(= Ma) and not including the variable Rh to be obtained are shown in FIG.
From the formulas shown in a. In this case, the expressions in the second and third lines in FIG. 8A satisfy the condition, but the expression in the second line in which the number of unknown variables in the expression is small is selected as the expression F1 (step e), and Proceed along the branch.

In step e of FIG. 7, the formula F1 is converted into Ma = g * (Ra ＾) using the formula shaping means 74 shown in FIG.
2) Format to the expression / Gr.

In step f of FIG. 7, the above-mentioned shaped equation F1 and the equations in the first and third rows of FIG. 8a are integrated by using the equation integrating means 75 shown in FIG. M
a) is erased, and Gr present as a redundant variable in the equation is eliminated using the equation simplification means 71. At this stage, the four equations in FIG. 8A are stored as a part of the solution derivation process data 15 in the form of FIG. 8B.

In step g of FIG. 7, equation F0 (FIG. 8)
Of examined for the presence of an unknown variable that is included in the first row) in the b, proceeds along the branch of the N _o because it does not exist. If an unknown variable exists, the processing of steps d to g in FIG. 7 is performed.

In step h of FIG. 7, the formula F0 is shaped into the form of “variable to be obtained =...” Using the formula shaping means 74, and the formula shown in FIG. Stored as a part.

In step i of FIG. 7, the known variable R
In a, g, v, and π, a variable value list 24 and a variable list 22
Since the value is given in, the process proceeds along the branch of Yes.

In step i of FIG. 7, calculation is performed by substituting a value into the equation shown in FIG. 8c, and unit conversion is performed as necessary. The result is stored as a part of the solution derivation process data 15 in the form shown in FIG. The answer is 5.5 * 1
0 ＾ 5 m is output to the learning progress control device 11 for comparison with the answer from the learner.

The answer sentence generation means 8 generates an answer sentence from the solution derivation process data based on the answer detail level designation data. Here, the answer detail level designation data indicates any of simple, standard, and detail, and FIG. 9 shows an output example when “standard” is designated. If "succinct" is designated, only the formula and answer used in the solution derivation process are described. When "details" is specified, details of the process of transforming an expression or the process of eliminating unknown variables that are not presented in the case of "standard" are described.

FIG. 10 is a diagram schematically showing another embodiment of the present invention. The configuration of the problem generating device 31 is the same as that of the problem generating device 1 of FIG. 2 described above, except that a detailed explanation command and an explanatory sentence are added to the input and output of the answer sentence generating means 9, respectively.

Here, when the learner requests a detailed explanation on the solution deriving process, the learning progress control unit 11 sends the corresponding location in the solution deriving process data and the understanding state of the learner. The detail level of the explanation is output to the answer sentence generating means 9 with reference to the data and the answer status. Upon receiving this instruction, the answer sentence generating means 9 generates and outputs a description of the mathematical expression processing process at an appropriate level of detail for the part requested by the learner.

The algorithm for deriving a solution shown in FIG. 7 is an example, and it goes without saying that an optimal algorithm for deriving a solution can be prepared according to the target area of the problem and the degree of difficulty. In the above description, any means that does not refer to this concrete realization method, such as the equation simplification means 71 and the equation shaping means 74, can be realized by known techniques.

In the creation of question data, once the question type data is created, the subsequent data creation work and data management are much more difficult than in the conventional method of preparing all question sentences, question diagrams, and answer sentences. Will be easier. By just preparing the question specific data, the question sentences and the answer sentences for the number of variables in the given variable set are created.

Each time the question data creator prepares the data, the question sentence generation means 5, the problem diagram generation means 6, the solution derivation means 7,
The output is input to a question generation simulator in which the answer sentence generation means 8 is incorporated, and each output is checked. At this time, for a problem for which a correct answer cannot be derived due to the specification of the solution deriving means 7, data indicating "solution derivation impossible" is added to the problem specific data. In the CAI system 10, by preventing the problem generator 1 from generating a problem to which “solution unresolvable” data has been added, a problem that the CAI system cannot solve may be presented to the learner and cause confusion. Can be eliminated.

[0042]

According to the first aspect of the present invention, the problem generating data is handled by dividing the problem generating data into abstracted problem data excluding redundancy and problem specific data describing specific names and values. Therefore, there is an effect that the management of data becomes easy, and the work efficiency when performing correction, additional reuse, and the like is improved.

As described above, by using the problem generating data, it is possible to easily generate a problem in which each variable in a given variable set is a "determined variable" and a solution thereto. In addition, answer sentences are set according to the level of understanding of the learner
Described in detail based on the specified "Detailed answer specification data"
Is done. As a result, it is possible to present to the learner various problems such as problems based on various viewpoints and problems of many specific examples in the CAI and their answers,
This has the effect of facilitating continuous learning and appropriate understanding of the learner's understanding.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of a CAI system.

FIG. 2 is a diagram showing an outline of a problem generating apparatus.

FIG. 3 is a diagram showing an example of question type data.

FIG. 4 is a diagram showing an example of problem specific data.

FIG. 5 is a diagram showing an output example of a question sentence generation unit.

FIG. 6 is a diagram illustrating components and operations of a solution deriving unit.

FIG. 7 is a view for explaining the flow of processing of a solution deriving unit;

FIG. 8 is a diagram showing an example of solution derivation process data;

FIG. 9 is a diagram illustrating an output example of an answer sentence generating unit;

FIG. 10 is a diagram showing an outline of another embodiment of the problem generating apparatus.

FIG. 11 is a diagram showing an outline of a conventional problem generating apparatus.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 problem generating device 2 problem type data 3 problem specific data 4 data memory means 5 problem sentence generating means 7 solution deriving means 8 answer sentence generating means 71 formula simplifying means 72 first formula selecting means 73 second formula selection Means 74 shaping means 75 integrating means

Claims (4)

(57) [Claims] 1. A method comprising: a. Variable list, relational expression, status statement template
Memory means for question type data including rates; b. Described the specific names and values of variables appearing in the problem
Memory means for problem specific data; c. Problem generation with desired and known variables
Depending on the directive, the above question type data and question specific data
Generate a question sentence that is a variable that determines the variable specified above.
A problem generating device for a learning system that uses a computer to generate a sentence problem and an answer sentence ; a problem generating device comprising: d . For the above relational expression, a variable determined by combining the operations of selecting the expression, eliminating redundant variables in the expression, eliminating unknown variables by integrating the expression, is expressed using known variables, and Solution deriving means for obtaining a value; e. Answer an outline of process in the solution deriving means
Answer sentence generating means for generating an answer sentence described in detail level based on the detail level designation data . 2. The problem generating apparatus according to claim 1, wherein the solution deriving means includes at least the following means: d1 . Expression simplification means for eliminating redundant variables included in the expression; d2 . First expression selecting means for selecting an expression including a variable to be obtained and having the smallest number of unknown variables from a set of relational expressions; d3 . Second expression selecting means for selecting an expression which does not include the variable to be determined and which includes an unknown variable included in the expression obtained by eliminating redundant variables from the expression selected by the first expression selecting means; d4 . Means for integrating two equations to eliminate unknown variables commonly included in both equations; d5 . An expression shaping unit that transforms a given expression into an expression that represents a specified variable in the expression. 3. The problem generating apparatus according to claim 1, wherein the answer sentence generating means responds to a request from the learner, data indicating the learner's understanding status, or a detailed explanation command based on the answer status. A problem generating apparatus for extracting and outputting a part of the process of (1). 4. The problem generating apparatus according to claim 2, wherein the answer sentence generating means responds to a request from the learner, data indicating the learner's understanding status, and a detailed explanation command based on the answer status. A problem generating apparatus for extracting and outputting a part of the process of (1).