JP2002229431A - Learning system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a learning system in which a learner objectively recognizes a field or a unit in which the learner is weak and can get a high learning effect by learning the field or the unit more often than other fields or units. SOLUTION: The learning system is constituted of a question making means for making a question with at least one attribute, an answering means for inputting a learner's answer to the given question, a correct/incorrect decision means for deciding whether the learner's answer is correct or incorrect by comparing the correct answer prepared beforehand with the learner's answer, a summing up means for summing up the number of correct/incorrect answers decided by the decision means in each of the attribute, a weak point recognition means for recognizing the learner's weak attribute with fewer correct answers or more incorrect answers on the basis of the summing up result, and a question selection means for selecting a question from questions with the learner's weak attribute.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a learning system, and more particularly, to a learning system for answering questions.

[0002]

2. Description of the Related Art In recent years, individual learning systems using computers have become widespread. For example, a math question is displayed on the display, and when the learner's answer is entered, the answer is compared with the correct answer prepared in advance to determine whether the answer is correct, and if correct, the next question is displayed. ,
If you make a mistake, the correct answer will be displayed. If a lot of questions are prepared in each field or unit, the learner can repeat the learning several times even after completing the learning.

[0003]

However, in the conventional learning system, even if there is a field or unit where the learner is strong or weak, the question is given without reflecting the field or unit. You will spend the same amount of time learning repeatedly, and you will not be able to focus on learning to overcome your weak areas and units. Alternatively, the only choice is to repeatedly study the field or unit that the learner chooses. Therefore, it has been difficult to achieve a high learning effect with the conventional learning system.

[0004] Accordingly, an object of the present invention is to provide a learning system that objectively recognizes a field or a unit that a learner is not good at and learns it more than others to obtain a high learning effect. .

[0005]

SUMMARY OF THE INVENTION The object of the present invention is as follows.
An answering means for asking a question having at least one attribute; an answering means for inputting a learner's answer to the question; and comparing a correct answer prepared in advance with the learner's answer to determine whether the learner's answer is correct or incorrect. Correctness determination means for determining
Tallying means for counting the number of correct and incorrect by the determination for each attribute of the problem, weak recognition means for recognizing poor attributes with few correct answers or many wrong answers from the result of the counting,
This is achieved by a learning system including question selection means for selecting a question from the question having the weak attribute. The learning system according to claim 1, wherein the problem is classified into a plurality of hierarchies in a tree shape, and an arbitrary hierarchy can be selected from these hierarchies, whereby a higher learning effect can be obtained.

[0006] The attribute of a problem referred to here is a target, a range, and a feature of the problem. The degree of learning of the problem can be measured and recognized from the viewpoint of these attributes. The objects, ranges, and features are subjects, fields, units, small units, items obtained by further subdividing the small units, and the like, and can be classified into a tree-like hierarchy. Specifically, the subjects are "mathematics", "national language", "society", "English", "music", etc. As a unit, in the case of "mathematics", "planar figures", "positive and negative numbers" , "Equation", etc., "language"
In the case of "grammar", "sentence reading", "composition", etc., in the case of "society", "industrial revolution", "Meiji Restoration", etc., "English"
In the case of `` current completion '', `` past completion '', etc., in the case of `` music '', `` singing '', `` instrument performance '', etc. For example, "Japanese language" such as "Area of figure", "Move figure", "Point and line"
For "grammar", "particle", "verb", etc. For "industrial revolution" of "society", "technological innovation", "capitalism", etc.
For "English" and "English conversation", "Phrasebook", "Greetings"
For "songs" of "music", such as "abbreviated form" and "pronunciation", "minor", "modulation", "two-beat", etc. In the case of "area of figure", "division of decimal point" in case of "area of triangle", "rounding off", "calculation of unit", etc. In the case of "promotion" in the case of "technological innovation" in the "industrial revolution" of "society", "pronunciation", etc. Examples include "r", "l", "i:" (pronunciation symbols), "doll with blue eyes" in the case of "song" of "singing" of "music", "small autumn found", and the like. According to the present invention, weaknesses and strengths can be recognized, and a question having an attribute of weakness that is objectively selected for a learner is given, so that the learner can focus on weaknesses, The learning effect can be greatly improved. When having a plurality of attributes, it may have a plurality of attributes of a certain hierarchy or may have an attribute of a plurality of hierarchies. When recognizing a weak point, it can be recognized from a multifaceted viewpoint, which is more preferable.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings. FIG. 1 shows a basic learning system of the present invention. The present invention provides a question answering means 1 for asking a question having at least one attribute, an answering means 2 for inputting a learner's answer to the question, and a comparison between a correct answer prepared in advance and the learner's answer to determine the learner's answer. Correctness determination means 3 for determining correctness of the answer, and totalization means 4 for totalizing the correctness number for each attribute of the question.
And weak question recognizing means 5 for recognizing weak attributes, and question selecting means 6 for selecting questions from questions having weak attributes.

[0008] In the normal learning mode of the present learning system, as in the conventional learning system, the questions are uniformly distributed over a predetermined range such as a certain field or unit. By 4, tallying is performed for each attribute. In order to distinguish the present invention by means 1 to 6 from ordinary learning, the learning by means 1 to 6 is hereinafter referred to as a weak mode. By selecting the weakness overcome mode of the present invention, weaknesses can be focused on from the results learned up to that point.

The problem is that there is at least one attribute as described above. Having a plurality of attributes makes it possible to recognize weaknesses or strengths from more viewpoints, so that the learning effect can be further improved.

FIG. 2 is a block diagram schematically showing the configuration of a learning system 10 according to a preferred embodiment of the present invention.

As shown in FIG. 2, a learning system 10 according to the present embodiment has a CP for controlling the operation of the entire apparatus.
U (central processing unit) 21, a memory 22 storing a program to be executed by the CPU 21, a problem database, a correct answer database, and the like, a keyboard 20 for the learner to input an answer, , A pointing device 26 for receiving an instruction from the learner, and a bus 29 for interconnecting these devices.

Further, a microphone 23 for inputting answers by voice, an A / D converter 24 for converting analog data input from the microphone 23 into digital data, and a speaker for performing questions and correct answers by voice. 27 and a D / A converter 28 for converting digital data into analog data.

Since such a hardware configuration is a configuration usually provided in a general personal computer,
The learning system 10 according to the present embodiment can be configured by a general personal computer.

The memory 22 stores programs necessary for the operation of the learning system 10. The programs required for the operation of the learning system 10 include a program (OS) for controlling the operation of the entire apparatus, a learner, A program for comparing the answer input from the user with the correct answer data (correctness judgment program), a program for totaling each attribute of the question (totaling program), and a program for recognizing weak attributes from the totaled results (poor recognition) Program) and a program for selecting the next question according to the question selection method (question selection program).

In addition, there may be a program for converting character data into voice data so that the problem or correct answer is not displayed on the display but is performed by voice from the speaker 27. Instead of using the keyboard 20 or the pointing device 26 to answer the above, there may be a program for converting voice data into character data in order to perform the answer by voice from the microphone 23.
Further, it is preferable that a program that is not essential for learning but is visually and audibly enjoyable is included in order to increase the motivation of the learner.

Further, the memory 22 includes a question database in which questions to be set are stored, a correct database in which correct answers corresponding to the respective questions are stored, and a tabulation table for counting the correct / incorrect number for each attribute. It is.

Each of the question data in the question database may include a message attached to the question, an answer example of the three-choice question, and the like. These may be separated from the question data, and may have a message database or an answer example database in which messages and answer examples corresponding to the respective questions are separately stored. The memory 22 of the present invention refers to the following. That is, a rewritable recording device such as a semiconductor memory, a magnetic disk, an optical disk, etc., which stores work data such as a program, a correct / incorrect tabulation table, etc., and a read-only optical disk, which contains a problem database or a correct answer database, etc. Can be. Also,
Data such as problem data and a correct / incorrect tabulation table may be stored via a communication circuit.

As shown in FIG. 3, a learning button 30, a weak button 31 and an end button 32 are displayed on the display 25. The learner uses the pointing device 26 (see FIG. 2) to select one of them. Can be pressed.

Next, the operation of the learning system having such a configuration will be described. When the learner presses the learning button displayed on the display 25 and selects the learning mode, a question selected from a plurality of question databases stored in the memory 22 under the control of the CPU 21 is displayed. It is displayed on the display 25.

In the learning mode, the method of selecting a question is a method of selecting an attribute to be set from a preset range without fail, and a question having an attribute selected without being selected is selected. For example, when the range to be learned by the learning system includes a plurality of small units, attributes representing the small units are sequentially selected, and a question having the attribute is set. For example, if you are learning junior high school mathematics, the attributes are “area of figure”, “movement of figure”, “point and line”, “large and small”,
They are "reciprocal", "calculation including four rules", "property of equation", and "linear equation solving method". The selection of the question is performed by the CPU 21 executing the question selection program.

It is desirable that consideration be given so that effective learning can be performed even when the number of questions is limited. For example, when selecting a question with a certain attribute, it is desirable that there are enough questions stored in the question database for each attribute and that even if the learner repeatedly learns, a new question is always given, but if it is not enough, Since it is necessary to select again from the questions that have already been set, it is desirable not to set the questions in the same order as the order set before. If the questions are presented in the same order, the learner may remember the correct answer order, and the learning effect is reduced.

In response to the question, the learner inputs the answer using the keyboard 20 or the like. The learner's answer is compared with the correct answer corresponding to the relevant question stored in the correct answer database of the memory 22, and if they match, the correct answer is determined. This comparison / judgment is performed by the CPU 21 executing the above-described right / wrong judgment program.

The result determined in this way is the bus 29
Is supplied to the display 25 via the
5 shows the result as “correct answer” as shown in FIGS. 4 and 5,
"Incorrect answer" is displayed, and in the case of an incorrect answer, the correct answer is also displayed. When the learner finishes learning the question, the learner presses the forward button 33 displayed below the display 25 to return to the display of FIG.

On the other hand, the memory 22 is provided with a correct / incorrect tabulation table for each attribute. Then, the correct or incorrect result of the answer is supplied via the bus 29, and the points of the correct or incorrect answer are added to all the attributes of the question given, and the total number of correct or incorrect answers for each attribute, the correct answer rate Alternatively, the wrong answer rate is tabulated. The result of the tallying may be displayed on the display 25 in a visually easy-to-understand manner as needed by the learner. This counting is performed by the CPU 21 executing the counting program described above.

In the case of continuing learning in the normal learning mode after completing the above-mentioned problem, when the learning button displayed on the display 25 is pressed by using the pointing device 26, the question selection program is executed, and again, The questions are uniformly selected from the learning target, range, and characteristics. The tallying process is performed inside the learning system, but the result is not reflected and the problem is selected.

On the other hand, if the learner finishes the above-mentioned problem and presses the weakness overcoming button displayed on the display 25 to select the weakness overcoming mode, the learner will be in the weakness overcoming mode, and control by the CPU 21 will also be possible. And the memory 22
The questions selected from the question database stored in the database 25 by reflecting the results already tabulated are displayed on the display 25 via the bus 29.

That is, in the weak mode, the memory 2
2, the number of incorrect answers is compared for each attribute from the correct / incorrect tabulation table for each attribute in which the results of the learning so far are stored, and an attribute having a large number of incorrect answers is selected. Alternatively, an attribute having a high wrong answer rate may be selected. The number of attributes to be selected may be set by the learner, and a plurality may be selected at a time. By doing so, you can learn multiple units or items that you are not good at once. Selection of these weak attributes is performed by the CPU 21 executing a weak recognition program.

Subsequently, a question including the selected attribute is selected from the question database and set as the next question. These questions are selected by the CPU 21 executing a question selection program.

The problem to be selected may be a problem that has never been learned, but it is important to re-learn a wrong problem in the weakness overcome mode, so that the learner can re-learn when desired. It is preferable to provide a means for providing a re-learning selection button on the display 25 so that a wrong question recorded in the memory 22 can be repeated. In that case, do not try to answer the questions in the same order as the previous questions. If the questions are presented in the same order, the learner may remember the correct answer order, and the learning effect is reduced.

Under the control of the CPU 21, the following problem is displayed on the display 25.

In response, the learner inputs an answer using the keyboard 20. The learner's answer is compared with the correct answer corresponding to this question stored in the correct answer database of the memory 22, and if they match, the correct answer is determined, and if they do not match, it is determined as the wrong answer. This comparison / judgment is performed by the CPU 21 executing the above-described right / wrong judgment program.

The result determined in this manner is transmitted to the bus 29
Is supplied to the display 25 via the
5 shows the result as “correct answer” as shown in FIGS.
"Incorrect answer" is displayed, and in the case of an incorrect answer, the correct answer is also displayed. When the learner has completed the study on this question, a forward button is displayed below the display 25. When the learner presses this button, the learner returns to the display of FIG.

Next, a description will be given of a case where a hierarchy is provided for attributes. The attribute in question may have several hierarchies.
The attributes referred to here are the target, range, and characteristics of the problem.
The degree of learning of the problem can be measured and recognized from the viewpoint of these attributes. The objects, ranges, and features are subjects, fields, units, small units, items obtained by further subdividing the small units, and the like, and the problems can be classified into a tree-like hierarchy. For example, the first hierarchy can be a large unit and the second hierarchy can be a small unit. In Table 1,
Here is an example when the subject to be learned is junior high school mathematics. At this time, the attributes of the first hierarchy are “plane figures”, which are large units,
In "positive / negative numbers" and "equations", the attributes of the second hierarchy are their small units. In the case of "planar figures", "area of figures", "movement of figures", "points and lines" Yes, "positive / negative" means "large and small", "reciprocal", and "calculations involving four rules". "Equation" means "property of equations" and "how to solve linear equations". .

An item obtained by further subdividing a small unit may be provided as a third hierarchy. For example, a small unit "area of figure"
In the case of, "area of a triangle", "area of a square", and "area of a circle".

If the subject to be learned in the learning system is all subjects of junior high school students, a further hierarchy may be provided above the hierarchy composed of large units. For example, the subject is the first hierarchy, and its attributes are “mathematics”, "Science", "Society", "English",
“National language” and “music”, in which case the second hierarchy may be a large unit and the third hierarchy may be a small unit.

[Table 1]

As described above, if the problem has attributes of several layers, weaknesses and specialties in each layer can be recognized. In other words, it is possible to objectively recognize a weak large unit, a weak small unit, or a weak subject from the learning result and focus on learning.

Further, it is preferable that the learner set the hierarchy at which he / she is weak. If the hierarchy is designated using the keyboard 20 or the pointing device 26, the weak attribute can be selected from the attribute of the designated hierarchy. For example, if the hierarchy 2 is designated, the small unit which is weak is recognized and the question is given from the range. Therefore, it is possible to freely set a range in which learning is to be focused.

Next, a description will be given of an embodiment of the present invention in which the correct answer is composed of a plurality of elements and the corresponding answer is composed of a plurality of elements, which is applied to learning of English. The English learning system here means that the learner converts the Japanese sentence displayed on the display into an English sentence, and when this is read out by the learner, it is compared with a plurality of correct answer examples prepared in advance and one of them It is a system that determines whether the pronunciation is correct or not, and displays “correct” or “wrong” on a display. But with traditional learning systems,
Even if it was determined to be an error, it was not known whether the reason was that an appropriate word was not used or that the pronunciation of the learner was incorrect. In the English learning system of the present invention, even if the whole English sentence is incorrect, it is determined whether or not the sound is correct for each word, so that the wrong word is known, and if the problem has the word of the correct example as an attribute, You can recognize weak words.

In the English learning system according to the present invention, when a learner expresses an English sentence composed of a plurality of words by voice, it is converted into character data by using voice recognition means, and the character data is converted to correct example data and word data. Comparison and judgment are made for each.

At this time, if the question has a word constituting an example of a correct answer as an attribute, the correct / incorrect number is counted in the true / false tabulation table for each attribute of the word according to the comparison of each word. Can be done.

Furthermore, if the problem has phonetic symbols of the words constituting the correct answer example as attributes, the number of correct and incorrect words for each attribute of the phonetic symbols is counted according to the comparison of each word. Can be recognized. It is preferable that the words constituting the correct answer example have characteristics of the word as attributes in addition to the phonetic symbols of the word, so that a weaker person can be recognized from a different viewpoint. Word features include, for example, "verb"
"Plural", "past".

The outline of the learning system of the present invention is basically the same as the configuration of FIG. 2, except that a memory recognition program (speech recognition program) and a character generated by the speech recognition are further stored in the memory 22. A program (correctness determination program) for comparing data and correct answer data for each word and determining correctness is required. In addition to this, it is more preferable that the pronunciation of the learner himself can be listened to again, but a program (recording program) for recording the data supplied from the microphone 23 is necessary. Further, the memory 22 includes a pronunciation database corresponding to each correct answer example in addition to a question database and a correct / incorrect tabulation table.

Although the above has been a description of learning English, it goes without saying that the present invention can be applied not only to English but also to other foreign languages.

[0044]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described in detail.

Embodiment 1 A junior high school mathematics learning system will be described as an example. The problem is to have a plurality of junior high school students' small units of mathematics as attributes. The attributes in this example are "area of figure", "movement of figure", "point and line",
These are "the magnitude of the number", "the reciprocal", "the calculation including the four rules", "the property of the equation", and "the method of solving the linear equation". Table 2, Table 3
Has a, b, c, d, e,
Expressed by symbols f, g and h. In Table 2, the attributes having a problem are indicated with an asterisk. For example, question 1 is a, c, b
Attribute. Table 3 shows a result of correct / incorrect judgment of the answer to the question and a correct / incorrect tabulation table for each attribute. Since the question 1 was a wrong answer, the judgment was "x" and the corresponding a,
Point 1 was added to the attributes of c and b.

According to Table 3, at the end of Question 10,
Attributes that are not good at 2 or more wrong answers are a, c, d, f
It can be seen that the weakest attribute is c.
At this point, the weakness overcome mode is executed. For example, if the number of attributes is set to 4 as the number of attributes to overcome the weaknesses, the questions having the above four attributes are respectively selected, and the learning of those attributes can be performed. In other words, the small unit "area of figure", "points and lines", "large and small numbers", "calculations including four rules"
Questions will be asked respectively. In addition, the attribute can be selected not based on the number of wrong answers but on the wrong answer rate.

[Table 2]

[Table 3]

Embodiment 2 An example of a junior high school mathematics learning system will be described. The major unit of mathematics of junior high school students was defined as the attribute of hierarchy 1 and the small unit was defined as the attribute of hierarchy 2. In this example, the attributes of the first layer are "planar figures", "positive and negative numbers", and "equations", and the attributes of the second layer are "area of figures", "movement of figures", "points and points". Lines, magnitudes of numbers, reciprocals, calculations involving the four rules, properties of equations, and how to solve linear equations. In Tables 4 and 5, the attributes of the first hierarchy are a, b, c, respectively, and the attributes of the second hierarchy are a1, a2, a3, b1, b2, b3, respectively.
It was represented by the symbols c1 and c2. The problem has multiple attributes with different hierarchies. In Table 4, the attributes of the problem are indicated by *. For example, question 1 has attributes a and a2. Table 5 is a table showing the results of correct / incorrect judgments on the answers to the questions and a correct / incorrect tabulation table for each attribute. For example, since question 1 was an incorrect answer, the judgment was "x" and the corresponding a, a
Point 1 was added to the attribute of 2.

According to Table 4, at the end of Question 16,
The worst large unit to choose by the number of wrong answers is a, and the number of wrong answers is 2
The above-mentioned weak units are a1, b2, and c1. At this point, the weakness overcome mode is executed, the hierarchy 1 is selected from the two hierarchies, the attribute 1 is set as the number of attributes to overcome the weakness, and the number of questions is set to 4, so that 4 questions having the attribute a are presented. . In other words, from the range of the most difficult unit “plane figure”,
Four questions are asked continuously.

[Table 4]

[Table 5]

Embodiment 3 An English learning system will be described as an example. Two levels are provided,
The words that make up the correct answer example are attributes of layer 1 and phonetic symbols are attributes of layer 2. Table 6 shows the results of comparing and judging the results read out from the questions 1 and 2, the attributes of each question, and the correct / incorrect number (error number in this example). For example, the correct answer "I see four people
le. ] Are the attributes “I”, “see”, “four”,
"People" and an attribute ("I") representing a phonetic symbol constituting the pronunciation of each word as an attribute of the second hierarchy
The attributes that represent phonetic symbols that make up the pronunciation of "ai" and "s
The attributes that represent the phonetic symbols that make up the pronunciation of "ee" are "s",
"I:", etc.).

As a result of the student reading the English sentence, "se
Since "e" was incorrect, an erroneous number is added to the attribute "see" of the word, and an erroneous number is added to the attributes "s" and "i:" of the phonetic symbol of the erroneous word "see". The correct / incorrect tabulation summarizes these results each time a question is asked, so that the learner can recognize weak words and poor phonetic symbols and learn problems with those attributes. The effect of learning increases.

[Table 6]

FIG. 6 is a schematic diagram showing a result of reading out an English sentence by a learner on a display. Display 2
In FIG. 5, question 1 was displayed without underlining only "see" to indicate that, because "see" was incorrect. In addition, the display 25 displays a self-pronunciation listening button 34 for listening to the voice read by the user and a pronunciation model listening button 35 for listening to the pronunciation model of the correct example. By pressing it, I was able to confirm the difference in pronunciation. In the case of Problem 2, "lemon" was similarly incorrect, so that only "lemon" is displayed without underlining.

[0052]

As described above, according to the present invention, since the disadvantages can be objectively recognized and the learning can be focused on the disadvantages, the learning effect can be greatly improved. Since a hierarchy can be set and a hierarchy can be selected, a detailed analysis of weak points can be performed, and the learning effect is further improved.

[Brief description of the drawings]

FIG. 1 is a basic illustration of the present invention.

FIG. 2 is a block diagram schematically showing a configuration of a learning system 10 according to a preferred embodiment of the present invention.

FIG. 3 is a schematic diagram illustrating an example of an operation button displayed on a display 25;

FIG. 4 is a schematic diagram showing an example of a case where a judgment displayed on a display 25 is a correct answer.

FIG. 5 is a schematic diagram showing an example of a case where a judgment displayed on a display 25 is an incorrect answer;

FIG. 6 is a schematic diagram illustrating an example of a determination result displayed on a display 25;

[Explanation of symbols]

 DESCRIPTION OF REFERENCE NUMERALS 1 answering means 2 answering means 3 correct / incorrect judgment means 4 totaling means 5 weak recognizing means 6 question selecting means 10 learning system 21 CPU 22 memory 23 microphone 24 A / D converter 25 display 26 pointing device 27 speaker 28 D / A converter 29 Bus 30 Learning button 31 Overcome weak button 32 End button 33 Forward button 34 Self-pronunciation listening button 35 Pronunciation model listening button

Claims (2)

[Claims] An answering means for asking a question having at least one attribute; an answering means for inputting a learner's answer to the question; and comparing the correct answer prepared in advance with the learner's answer, Correctness judgment means for judging correctness of the answer of the person, tallying means for summing up the number of correctness by the judgment for each attribute of the problem, and, based on the result of the tallying, the attribute having a small number of correct answers or a weak attribute having many wrong answers A learning system comprising: a weakness recognition means for recognizing; and a question selection means for selecting a question from the question having the weak attribute. 2. Classifying the problem into a tree shape into a plurality of hierarchies,
The learning system according to claim 1, wherein an arbitrary layer is selected from these layers.