Information processing device and program

Abstract

To confirm whether points are appropriately allocated to questions constituting a test.SOLUTION: An information processor acquires point allocation information about points allocated to each of the questions constituting a test. The information processor also acquires a total score on the test. Then, the information processor calculates the sum of points allocated to each of the questions constituting the test on the basis of the acquired point allocation information. The information processor determines whether the calculated sum of points allocated to each of the questions constituting the test matches the total score of the acquired test and outputs an alert when it is determined that they do not match each other.SELECTED DRAWING: Figure 21

Description

Technical Field

[0001] The present invention relates to a technique for checking test scores.

Background Art

[0002] Conventionally, a system for automatically scoring tests has been known. Patent Document 1 discloses a system for calculating a base score according to the score ratio for each question and the relative difficulty of the question. Patent Document 1 also discloses a system for dynamically changing the score during the examination based on the answer results and answer times of a plurality of learners with respect to the base score.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the invention according to Patent Document 1, it is necessary to set in advance the score ratio and difficulty level for each question, which places a burden on the learning supporter. In addition, the invention according to Patent Document 1 determines the score only based on the relative relationship of each question in one test, and similar questions in other tests are not considered.

[0005] In addition, the learning supporter may set the score by himself / herself referring to the automatically calculated score. In this case, problems such as noticing and correcting a mistake in setting the score during or after the test scoring have occurred.

[0006] The present invention is made, for example, to solve the above problems, and an object thereof is to provide an information processing apparatus for checking whether the scores of the questions constituting a test are appropriate.

Means for Solving the Problems

[0007] In one aspect of the present invention, the information processing device includes: a scoring information acquisition unit that acquires scoring information relating to the scoring of each question constituting the test; a total score acquisition unit that acquires the total score of the test; and an alert output unit that determines, based on the scoring information, whether the sum of the scores of each question constituting the test matches the total score of the test, and outputs an alert if it is determined that they do not match. , a memory unit that stores information about multiple problems, a target problem information acquisition unit that acquires target problem information about target problems that constitute a test and are subject to scoring, a similarity information acquisition unit that acquires similarity information about similar problems similar to each target problem from the memory unit based on the target problem information, a scoring calculation unit that calculates the score for each target problem based on the target problem information, the similarity information for each target problem, and a function used for scoring calculation, and an output unit that outputs the scores, Equipped with ru According to this embodiment, the information processing device can determine whether the point allocation for each question constituting the test is appropriate, and can prevent point allocation adjustments and the resulting problems. Furthermore, according to this embodiment, the information processing device can automatically calculate the score for each target question that makes up the test, based on the target question information, the similarity information of each target question, and the function used for calculating the score. Therefore, the score can be set based on objective evaluation, and the burden on learning support staff can be reduced.

[0009] In one embodiment of the above-described information processing device, the storage unit stores the text of the problem, the attribute information of the problem, the difficulty level of the problem, and the correct answer rate of the problem in association with each other. The target problem information includes the text of the target problem, the attribute information of the target problem, and the weights related to the scoring of the target problem. The similarity information acquisition unit acquires the difficulty level and correct answer rate of similar problems similar to each target problem from the storage unit as similarity information based on the target problem information. The scoring calculation unit calculates the score for each target problem based on the weights related to the scoring of each target problem, the difficulty level and correct answer rate of similar problems similar to each target problem, and the function used for scoring calculation. According to this embodiment, the information processing device can automatically calculate the score for each problem constituting the test based on the weights related to the scoring of each target problem, the difficulty level and correct answer rate of similar problems similar to each target problem, and the function used for scoring calculation. Therefore, it is possible to calculate the score for each target problem according to the weights related to the scoring of each target problem and the difficulty level and correct answer rate of similar problems.

[0010] In one embodiment of the above-described information processing device, an average value calculation unit is provided that calculates the average difficulty level and average correct answer rate of similar problems similar to each target problem based on the similar information, and the scoring calculation unit calculates the scoring for each target problem by substituting the average difficulty level and average correct answer rate corresponding to each target problem into the function and multiplying by a value indicating the weight corresponding to each target problem. According to this embodiment, the information processing device can calculate the scoring for each target problem by substituting the average difficulty level and average correct answer rate of similar problems similar to the target problem into a function used for scoring calculation and multiplying the calculated value by a value indicating the weight for the scoring of each target problem.

[0011] In one embodiment of the above-described information processing device, a parameter information acquisition unit is provided that acquires the parameter values ​​of the function as parameter information, and the scoring calculation unit calculates the scoring for each target problem by further substituting the parameter values ​​into the function. According to this embodiment, the information processing device can calculate the scoring for each type of problem using a function with adjusted parameter values.

[0012] In another aspect of the present invention, computer And a memory unit that stores information about multiple problems A program executed by an information processing device comprising: a scoring information acquisition unit that acquires scoring information regarding the scoring of each question constituting a test; a total score acquisition unit that acquires the total score of the test; and an alert output unit that determines whether the sum of the scores of each question constituting the test matches the total score of the test based on the scoring information, and outputs an alert if it is determined that they do not match. A target question information acquisition unit that constructs the test and acquires target question information for which scoring is to be set; a similar information acquisition unit that acquires similar information for similar questions similar to each target question from the storage unit based on the target question information; a scoring calculation unit that calculates the score for each target question based on the target question information, the similar information for each target question, and a function used for scoring calculation; and an output unit that outputs the scores. The computer is made to function as described above. By installing and running this program on the computer, the information processing device according to the present invention can be configured. [Effects of the Invention]

[0013] According to the information processing device of the present invention, by confirming whether the point distribution of the questions that make up the test is appropriate, it is possible to prevent corrections to the point distribution and the resulting problems. [Brief explanation of the drawing]

[0014] [Figure 1] Shows the configuration of the learning support system to which the server of the present invention is applied. [Figure 2] It is a block diagram showing the hardware configuration of the server. [Figure 3] It is an example of the data configuration of the problem DB. [Figure 4] It is a block diagram showing the functional configuration of the server in the first embodiment. [Figure 5] It is an example of the scoring screen in the first embodiment. [Figure 6] It is an example of the reference screen in the first embodiment. [Figure 7] It is an example of the parameter setting screen in the first embodiment. [Figure 8] It is a diagram for explaining the score before normalization. [Figure 9] It is a diagram for explaining the normalization method in the first embodiment. [Figure 10] It is a diagram showing the score after normalization in the first embodiment. [Figure 11] It is a flowchart of the score calculation process in the first embodiment. [Figure 12] It is a block diagram showing the functional configuration of the server in the second embodiment. [Figure 13] It is an example of the scoring screen in the second embodiment. [Figure 14] It is an example of the reference screen in the second embodiment. [Figure 15] It is an example of the parameter setting screen in the second embodiment. [Figure 16] It is a diagram for explaining the score before normalization considering weights. [Figure 17] It is a diagram for explaining the normalization method in the second embodiment. [Figure 18] It is a diagram showing the score before normalization and the score after normalization in the second embodiment. [Figure 19] It is an example of the confirmation screen in the second embodiment. <opposite0opposite [Figure 20]This is a flowchart of the scoring process in the second embodiment. [Figure 21] This is a flowchart of the verification process in the second embodiment. [Figure 22] This diagram illustrates the identification of similarity problems using text similarity. [Figure 23] This is a block diagram showing the hardware configuration of the server in the third modified example. [Modes for carrying out the invention]

[0015] Embodiments of the present invention will be described below with reference to the drawings. <First Embodiment> [Overall structure] Figure 1 shows the configuration of a learning support system to which the server of the present invention is applied. The learning support system 100 is a system that acquires similar problems to the problems that make up the test and automatically calculates the points allocated to the problems according to the correct answer rate and difficulty level of the similar problems. The learning support system 100 is configured so that a plurality of terminal devices 10 and a server 20 can communicate via a network 5 such as the Internet. In this embodiment, the test and the problems that make up the test are input by a teacher who is a learning supporter, and the test is taken by learners such as students.

[0016] The terminal device 10 is used by learning support staff such as teachers, and is an information processing device such as a tablet or a personal computer terminal. Specifically, the terminal device 10 transmits information about the target questions and parameter information related to the questions that constitute the test and for which scoring is set (hereinafter also referred to as "target questions"), and receives and displays the scoring screen.

[0017] Server 20 is an information processing device that processes, stores, and transmits various types of information, and is, for example, a server device, a personal computer, or a general-purpose tablet PC (personal computer).

[0018] [Server Configuration] Figure 2 is a block diagram showing the hardware configuration of server 20. Server 20 comprises a communication unit 11, a control unit 12, a storage unit 13, a recording medium 14, a display unit 15, and an input unit 16. These components, the problem database (hereinafter referred to as "database" or "DB") 31, and the results DB 32 are interconnected via a bus 19.

[0019] Server 20 can run on a single computer, or it can run in a distributed manner across multiple computers, or it can run in a distributed manner across virtual machines.

[0020] The communication unit 11 is a communication unit for communicating with the terminal device 10 via the network 5. Specifically, the communication unit 11 receives target problem information and parameter information from the terminal device 10, and transmits the scoring screen to the terminal device 10.

[0021] The control unit 12 includes arithmetic processing units such as a CPU (Central Processing Unit), MPU (Micro-Processing Unit), and GPU (Graphics Processing Unit), and performs various information processing and control processing related to the server 20 by reading and executing programs stored in the memory unit 13. The programs can be deployed on a single computer or site, or distributed across multiple sites and executed on multiple computers interconnected by a communication network. Although Figure 2 describes the control unit 12 as a single processor, it may also be a multi-processor system.

[0022] The storage unit 13 includes memory elements such as RAM (Random Access Memory) and ROM (Read Only Memory), and stores programs or data necessary for the control unit 12 to execute processing. The storage unit 13 also temporarily stores data necessary for the control unit 12 to execute arithmetic processing.

[0023] The recording medium 14 is a non-volatile, non-temporary recording medium such as a disk-shaped recording medium or semiconductor memory, and is configured to be detachable from the server 20. The recording medium 14 stores various programs that the control unit 12 executes. When the server 20 performs score calculation processing or verification processing, the programs stored on the recording medium 14 are loaded into the storage unit 13 and executed by the control unit 12.

[0024] The display unit 15 is a liquid crystal display or an organic EL (electroluminescence) display, etc., and displays various information according to the instructions of the control unit 12. The input unit 16 is an input device such as a mouse, keyboard, touch panel, or buttons, and outputs the received operation information to the control unit 12.

[0025] Problem DB31 stores text data (hereinafter simply referred to as "text") of multiple pre-created problems, along with attribute information and difficulty levels for each problem. Attribute information is information that classifies the learning content corresponding to a problem, and specifically includes learning element name, learning element ID, curriculum guideline name, curriculum guideline code, unit, and subject. Figure 3 shows an example of the data structure of Problem DB31. Figure 3(a) is Problem DB31x using a learning element list as attribute information, and Figure 3(b) is Problem DB31y using a curriculum guideline as attribute information.

[0026] As shown in Figure 3(a), the problem DB31x stores information such as learning element name, learning element ID, example problem, problem ID, and difficulty level in an associated manner. Here, a learning element is a subdivision of learning content and textbook explanations into the most appropriate level of detail, taking into account learning units and other learning groups, for purposes such as linking textbooks and teaching materials, understanding learning progress, and creating instructional plans. The learning element name is the name that represents the learning content of the learning element, and the learning element ID is the identification information for each learning element. The learning element list is mainly maintained by the Japan Educational Information Promotion Association, and learning element IDs are pre-associated with learning element names. The example problem is an example of a problem corresponding to the learning content of each learning element. The problem ID is the identification information for each problem, and the difficulty level is the difficulty level for each problem.

[0027] Here, difficulty level is a numerical value indicating the difficulty of a problem. The difficulty level is based on any concept, such as the difficulty of the problem or the level of correct answer rate, and may be set by evaluating each problem using Item Response Theory (IRT), or it may be set artificially in advance. Specifically, in this embodiment, the difficulty level is calculated based on performance information including the correctness of answers to problems by multiple students, as described later. However, as shown in Figure 3(a), it may be stored in the problem DB31x and extracted when needed, or it may not be stored in the problem DB31x and may be calculated each time based on the performance information as needed.

[0028] In this embodiment, one example problem is stored for each learning element, but the present invention is not limited thereto, and multiple example problems may be stored for each learning element.

[0029] As shown in Figure 3(b), the problem DB31y stores information such as the name of the curriculum guideline, the curriculum guideline code, example problems, problem IDs, and difficulty levels, all associated with each other. Here, the curriculum guideline is a set of guidelines used by schools to standardize educational data, which subdivides learning content as a basis for organizing the curriculum. The name of the curriculum guideline is the name that represents the learning content of the curriculum guideline, and the curriculum guideline code is the identification information for each curriculum guideline. The curriculum guideline is mainly developed by the Ministry of Education, Culture, Sports, Science and Technology, and a curriculum guideline code is pre-assigned to each curriculum guideline name. The example problems are examples of problems corresponding to the learning content of each curriculum guideline. The problem ID is the identification information for each problem, and the difficulty level is the difficulty level for each problem.

[0030] In this embodiment, one example problem is stored for each curriculum guideline, but the present invention is not limited thereto, and multiple example problems may be stored for each curriculum guideline.

[0031] As will be described in detail later, in the scoring calculation process of this embodiment, information about similar problems is obtained from problem DB31x based on the learning element name and learning element ID. However, the present invention is not limited to this, and information about similar problems may also be obtained from problem DB31y based on the curriculum guideline name and curriculum guideline code. Thus, the attribute information for obtaining information about similar problems can be arbitrarily set individually or in combination of multiple items from among the learning element name, learning element ID, curriculum guideline name, curriculum guideline code, unit, subject, etc.

[0032] The Grade DB32 stores grade information by associating the problem ID of the problem a student has attempted with the correctness of their answer to that problem, linked together with the student ID that identifies the student. The correct answer rate is the percentage of students who answered a given problem correctly. Specifically, in this embodiment, the correct answer rate is calculated based on the correctness of the answers of multiple students to the problem included in the grade information, and is stored in the Grade DB32 linked to the problem ID. However, the present invention is not limited to this, and the correct answer rate may not be stored in the Grade DB32, but may be calculated each time it is needed based on the grade information. Alternatively, the correct answer rate may be stored in the Problem DB31 linked to the problem ID.

[0033] The storage configurations of each DB described above are merely examples, and other storage configurations may be used as long as the relationships between the data are maintained. Furthermore, each DB can be implemented using a recording medium such as an HDD (Hard Disk Drive) or SSD (Solid State Drive). In this embodiment, the storage unit 13 and the various DBs may be configured as a single storage device, or they may be separate storage devices. Additionally, the various DBs may be external storage devices connected to the server 20, and their configuration can be arbitrarily set.

[0034] Figure 4 is a block diagram showing the functional configuration of server 20. Functionally, server 20 comprises a problem DB 31, a performance DB 32, a target problem information acquisition unit 41, a parameter information acquisition unit 42, a similar information acquisition unit 43, an average value calculation unit 44, a point allocation calculation unit 45, a normalization unit 46, and a screen output unit 47.

[0035] The target problem information acquisition unit 41, parameter information acquisition unit 42, similarity information acquisition unit 43, average value calculation unit 44, point allocation calculation unit 45, normalization unit 46, and screen output unit 47 are realized by the control unit 12 executing a program.

[0036] The target problem information acquisition unit 41 acquires information about the target problem as target problem information. Specifically, the target problem information includes the text of the problems that make up the test (hereinafter also referred to as "problem text") and attribute information of the problem, and is information that is entered by the teacher using the scoring screen described later. In this embodiment, the server 20 is equipped with problem DB 31x shown in Figure 3(a) and problem DB 31y shown in Figure 3(b), and it is assumed that the teacher has entered one or more of the following as attribute information of the problem: learning element name, learning element ID, curriculum guideline name, and curriculum guideline code.

[0037] The parameter information acquisition unit 42 acquires the values ​​of various parameters of the function used for calculating the scoring (hereinafter also referred to as the "scoring calculation function") as parameter information. In this embodiment, the parameter information is information entered by the teacher using the scoring screen described later, but the present invention is not limited thereto and may be set in advance.

[0038] The similarity information acquisition unit 43 acquires information about similar problems that are similar to each target problem from various databases based on the target problem information. The similarity information acquisition unit 43 includes a similar problem text acquisition unit 51, a similar problem difficulty level acquisition unit 52, and a similar problem correct answer rate acquisition unit 53.

[0039] The Similar Problem Text Acquisition Unit 51 acquires corresponding example problems from Problem DB 31x and Problem DB 31y as similar problem texts to the target problem, based on the learning element name, learning element ID, curriculum guideline name, and curriculum guideline code included in the target problem information. For example, if the target problem information includes the learning element name "Integer Division," the Similar Problem Text Acquisition Unit 51 acquires the example problem "10÷4" associated with the learning element name "Integer Division" from Problem DB 31x as similar problem text. In this way, the Similar Problem Text Acquisition Unit 51 acquires example problems associated with each learning element from Problem DB 31x as similar problems to each target problem, based on the learning element name and learning element ID corresponding to each target problem.

[0040] The similar problem difficulty acquisition unit 52 acquires the difficulty level of each similar problem from the problem DB 31x. The similar problem correct answer rate acquisition unit 53 acquires the correct answer rate of each similar problem from the score DB 32 based on the problem ID of the similar problem.

[0041] In this embodiment, the similar information corresponding to each target problem includes the text of the similar problem acquired by the similar problem text acquisition unit 51, the difficulty level of the similar problem acquired by the similar problem difficulty level acquisition unit 52, and the correct answer rate of the similar problem acquired by the similar problem correct answer rate acquisition unit 53. However, the present invention is not limited thereto, and the information contained in the similar information can be arbitrarily set.

[0042] The average value calculation unit 44 calculates the average difficulty level and average correct answer rate of similar problems based on the similar information corresponding to the target problem, when there are multiple similar problems to a given target problem. The average value calculation unit 44 includes a difficulty level average value calculation unit 54 and a correct answer rate average value calculation unit 55.

[0043] The difficulty average calculation unit 54 calculates the average difficulty of similar problems based on the numerical value indicating the difficulty of similar problems included in the similar information corresponding to each target problem. If there is only one similar problem included in the similar information, the difficulty average calculation unit 54 may consider the numerical value indicating the difficulty of that similar problem as the difficulty average. The correct answer rate average calculation unit 55 calculates the correct answer rate average of similar problems based on the correct answer rates of similar problems included in the similar information corresponding to each target problem. If there is only one similar problem included in the similar information, the correct answer rate average calculation unit 55 may consider the correct answer rate of that similar problem as the correct answer rate average.

[0044] The scoring unit 45 calculates the score for each target question based on the target question information and similarity information for each target question. Specifically, the scoring unit 45 uses a scoring function to calculate the score for each target question according to the difficulty level and correct answer rate of similar questions. The scoring function may be predetermined as one, or multiple options may be provided for the teacher to select the appropriate one. In this embodiment, the scoring unit 45 calculates the score for each target question by substituting the average difficulty level and average correct answer rate corresponding to each target question into the scoring function. The scoring unit 45 may also calculate the score for each target question by further substituting the parameter values, which have been previously entered by the teacher based on parameter information, into the same function.

[0045] The normalization unit 46 obtains the total test score (hereinafter also referred to as "total test score") and normalizes it so that the sum of the points allocated to each target question equals the total test score. The total test score may be set in advance, or it may be entered by the teacher using the scoring screen.

[0046] The screen output unit 47 creates a scoring screen that includes the point values ​​for each target question normalized by the normalization unit 46, and transmits screen information related to the scoring screen to the terminal device 10. By displaying the scoring screen based on the screen information received by the terminal device 10, the teacher can confirm the point values ​​for each target question.

[0047] The screen output unit 47 may also create a screen for the teacher to input information about the target problems and parameters, and transmit screen information related to that screen to the terminal device 10. In this embodiment, although the details will be described later, the screen output unit 47 will create a scoring screen, a reference screen, a parameter setting screen, etc., to realize input of target problem information and parameter information, output of normalized scoring for each target problem, and referencing of similar information, and will transmit information related to the various screens to the terminal device 10 in response to login and various requests.

[0048] [Scoring screen] Next, the scoring screen created and output by the screen output unit 47 will be described. The teacher displays the scoring screen by logging into a designated site using the terminal device 10 and inputs the text of the target questions that make up the designated test, as well as the attribute information of those target questions. The teacher also inputs parameter values ​​for the scoring calculation function from the scoring screen as needed. Furthermore, the teacher requests the scoring by pressing a designated button on the scoring screen, and the scoring for each target question is displayed on the screen.

[0049] Figure 5 shows an example of a scoring screen. As shown in Figure 5, the scoring screen has a target question input area 71, an add question button 72, a similar question reference button 73, a parameter setting button 74, a score calculation button 75, and a score output area 76.

[0050] The target problem input area 71 is an area where the teacher inputs the text of the target problem, the name of the learning element, the ID of the learning element, etc. Information about the target problem entered in the target problem input area 71 is acquired by the server 20 as target problem information. Specifically, in the target problem input area 71, the teacher inputs the text of the target problem "4200÷6" as the problem text corresponding to number (1), and "Integer division" as the name of the learning element. At this time, the teacher may also input the learning element ID associated with "Integer division". In this embodiment, since the test consists of a total of 5 questions, the teacher adds a row for number (2) to the target problem input area 71 by pressing the add question button 72. Then, in the target problem input area 71, the teacher inputs the text of the target problem, the name of the learning element, and the ID of the learning element corresponding to number (2). After that, the teacher inputs target problem information for 5 target problems by repeatedly pressing the add question button 72 and inputting the target problem text, learning element name, and learning element ID corresponding to the number, as shown in the figure.

[0051] The similar problem reference button 73, when pressed, allows a request for a reference screen to be sent to the server 20. Figure 6 shows an example of a reference screen. As shown in Figure 6, the reference screen displays the number of the problem that makes up the test, the target problem, similar problems to the target problem, the average difficulty level of the similar problems, and the average correct answer rate in association with each other. When the server 20 receives a request for a reference screen from the terminal device 10, it creates the reference screen and sends screen information related to the reference screen to the terminal device 10. By displaying the reference screen based on the screen information received by the terminal device 10, the teacher can easily check similar problems, their average difficulty level, average correct answer rate, etc. for each target problem that they have entered.

[0052] The parameter setting button 74, when pressed, can request the server 20 to access the parameter setting screen. Figure 7 shows an example of the parameter setting screen. As shown in Figure 7, the parameter setting screen includes a graph 60 representing a function for calculating the score, a difficulty range item 61, a score range item 62, a correct answer rate range item 63, an x-axis slope item 64, a y-axis slope item 65, and a total score item 66.

[0053] Graph 60 represents a function for calculating the scoring, and as shown in the figure, it is a three-dimensional graph composed of x, y, and z axes. Specifically, the x axis represents difficulty, the y axis represents the correct answer rate, and the z axis represents the scoring. Server 20 uses the equation of the plane in Graph 60, "z = ax + by + c", to calculate the scoring for each target problem. Difficulty Range Item 61 is an item for setting and inputting a numerical range indicating the difficulty level. Scoring Range Item 62 is an item for setting and inputting a range for the scoring. Correct Answer Rate Range Item 63 is an item for setting and inputting a range for the correct answer rate. The x-axis Slope Item 64 is "a" in the equation of the plane, and is an item for setting and inputting the value of the parameter corresponding to the difficulty level, i.e., the slope of the x axis. The y-axis Slope Item 65 is "b" in the equation of the plane, and is an item for setting and inputting the value of the parameter corresponding to the correct answer rate, i.e., the slope of the y axis. The "Total Score" item 66 is where you set and input the sum of the points allocated to each question, i.e., the total test score. In this embodiment, the total test score is set to "100 points".

[0054] When server 20 receives a request for a parameter setting screen from terminal device 10, it creates the parameter setting screen and sends screen information related to the parameter setting screen to terminal device 10. By displaying the parameter setting screen based on the screen information received by terminal device 10, the teacher can use the parameter setting screen to send various ranges and parameter values ​​that they have arbitrarily set to server 20.

[0055] The teacher uses terminal device 10 to input the target problem information and parameter information, and then presses the score calculation button 75 to send a request for score output to server 20. Upon receiving the score output request, server 20 uses a score calculation function to calculate the score for each target problem. Server 20 then creates a score screen that includes the score for each target problem and sends screen information related to this score screen to terminal device 10. In this way, when the teacher presses the score calculation button 75, the score for each target problem is displayed on the score screen.

[0056] Specifically, Server 20 calculates the points allocated to each target problem using the equation of the plane "z = ax + by + c" in Graph 60. Server 20 substitutes the average difficulty value for "x" and the average correct answer rate for "y" in the equation of the plane. Server 20 also substitutes the values ​​of the corresponding parameters for "a", "b", and "c" in the equation of the plane based on the parameter information. In this way, Server 20 can calculate the points allocated to "z" in the equation of the plane for each target problem. Figure 8 is a diagram illustrating the points allocated before normalization. As shown in Figure 8, the points allocated to each target problem before normalization are calculated using a point allocation function, according to the average difficulty value and average correct answer rate of similar problems.

[0057] Server 20 obtains the total test score and normalizes it so that the sum of the points allocated to each target question equals the total test score. Specifically, Server 20 considers the sum of the points included in the parameter information as the total test score. Figure 9 illustrates the method of normalizing so that the sum of the points before normalization equals 100 points. Server 20 calculates the "normalized score" by substituting the "sum of points before normalization" and the "points before normalization" for each target question into Equation 70 as shown in the figure. For example, in the case of the score allocation shown in Figure 8, the sum of the points before normalization for target questions (1) to (5) is (10 + 10 + 47 + 33 + 53) = "153", and the point before normalization for target question (1) is "10". Substituting these into Equation 70, the normalized score for target question (1) becomes "6", as shown in Figure 9. Server 20 similarly calculates the normalized scores for target questions (2) to (5).

[0058] Here, if the sum of the normalized scores falls short of 100 points as a result of normalization, the server 20 adjusts by adding the points to any of the target questions. Specifically, in the example in Figure 9, the adjustment is made by adding 1 point each to the scores of target questions numbered (3), (4), and (5). Figure 10 shows the normalized scores for target questions numbered (1) to (5). The normalized scores for the target questions are displayed in the score output area 76 on the right side of the screen, as shown in Figure 5.

[0059] Furthermore, when logging into the designated site, the scoring screen will only display the range indicated by the input on the left side of the screen as shown in Figure 5. Pressing the scoring calculation button 75 will display the range indicated by the output on the right side of the screen. Also, the scoring screen shown in Figure 5, the reference screen shown in Figure 6, and the parameter setting screen shown in Figure 7 are examples, and the present invention is not limited thereto; the configuration of various screens can be set arbitrarily.

[0060] [Score calculation process] Next, we will explain the process of obtaining similar problems to the target problem and calculating the points according to the correct answer rate and difficulty level of those similar problems. Figure 11 is a flowchart of the point calculation process performed by Server 20. This process is achieved by Server 20 executing a pre-prepared program.

[0061] The teacher uses terminal device 10 to log in to a designated site to display the scoring screen, where they input target question information and parameter information, and refer to similar information. Once input and referencing are complete, the teacher presses a designated button on the scoring screen to send the target question information and parameter information to server 20 and request the output of the scoring for each target question.

[0062] Server 20 obtains target problem information from terminal device 10 (step S11). Server 20 also obtains parameter information from terminal device 10 (step S12). Then, based on the attribute information of the target problem included in the target problem information, Server 20 obtains the text and difficulty level of similar problems from problem DB 31 (step S13). Server 20 also obtains the correct answer rate of the similar problems from score DB 32 (step S14). Then, Server 20 calculates the average difficulty level and average correct answer rate of the target problem (step S15). Server 20 calculates the score of the target problem by substituting the average difficulty level, average correct answer rate, and the values ​​of various parameters included in the parameter information into a function for calculating the score (step S16).

[0063] Server 20 determines whether or not it has calculated the points for all target questions (step S17). If it determines that it has not calculated the points for all target questions (step S17; No), Server 20 returns to the process in step S13 and calculates the points for the target questions that have not yet been calculated. On the other hand, if it determines that it has calculated the points for all target questions (step S17; Yes), Server 20 normalizes the points so that the sum of the points equals the total test score. Then, Server 20 creates a scoring screen including the normalized points and sends screen information related to the scoring screen to Terminal Device 10 (step S18). This completes the scoring calculation process. Terminal Device 10 displays the scoring screen based on the screen information received from Server 20.

[0064] The learning support system 100 of this embodiment can automatically calculate and output the points allocated to each question based on the correct answer rate and difficulty level by acquiring similar questions that make up the test. Therefore, the points allocated to each question can be set based on objective evaluations such as the correct answer rate and difficulty level, rather than the subjective evaluation of the learning support worker. Furthermore, by setting the points allocated to the test to reflect the correct answer rate and difficulty level of similar questions, the discrepancy between the learner's academic ability and their score can be reduced.

[0065] Furthermore, assigning points to each question is a burdensome task for learning support staff. However, since the learning support system 100 automatically calculates and outputs the points for each question, learning support staff can easily check and set appropriate point assignments for questions they have arbitrarily selected, taking into account the difficulty level and correct answer rate. Therefore, the burden on learning support staff can be reduced.

[0066] Furthermore, since teachers can arbitrarily set the parameters of the function used to calculate the point distribution, they can easily make adjustments such as giving higher points to "difficult questions with a low correct answer rate" that many students cannot answer correctly, or giving higher points to "basic questions with a low correct answer rate" that they want students to understand.

[0067] <Second Embodiment> In the first embodiment, the server 20 calculates the points allocated to each question in the test according to the correct answer rate and difficulty level. However, the present invention is not limited to this, and the server may also calculate the points by considering the weight of each question's points (hereinafter simply referred to as "weight") set by the learning supporter. Furthermore, if the teacher sets the points allocated to each question based on the points allocated to each question calculated according to the correct answer rate, difficulty level, and weight, the server 20 may also verify whether the teacher's points are appropriate.

[0068] The learning support system 100x in the second embodiment is a system that acquires similar problems to each problem that makes up the test, and automatically calculates the score for each problem according to the correct answer rate and difficulty level of the similar problems and the weight of the points set for each problem. The learning support system 100x is also a system that verifies whether the points set for each problem are appropriate. Note that the overall configuration and the server hardware configuration are the same as in the first embodiment, so for convenience, the explanation will be omitted.

[0069] First, the functional configuration of the server in the second embodiment will be described. Figure 12 is a block diagram showing the functional configuration of server 20x. Functionally, server 20x includes a problem DB 31, a score DB 32, a target problem information acquisition unit 41x, a parameter information acquisition unit 42x, a similar information acquisition unit 43, an average value calculation unit 44, a point allocation calculation unit 45x, a normalization unit 46x, a screen output unit 47x, and a total score confirmation unit 48.

[0070] For convenience, the same functions as in the first embodiment will not be explained. In addition, the target problem information acquisition unit 41x, parameter information acquisition unit 42x, similar information acquisition unit 43, average value calculation unit 44, point allocation calculation unit 45x, normalization unit 46x, screen output unit 47x, and total score confirmation unit 48 are realized by the control unit 12 executing a program.

[0071] The target problem information acquisition unit 41x acquires target problem information. Specifically, the target problem information includes the problem text of the problems that make up the test, the weight of the points allocated to the problem, attribute information of the problem, and information such as whether or not it is normalized, and is information that is entered by the teacher using the scoring screen described later. The weight is a numerical value set by the teacher or the like, and if the teacher wants to allocate more points to a particular problem in the test, the numerical value indicating the weight of that problem is set high. On the other hand, if the teacher wants to allocate less points to a particular problem in the test, the numerical value indicating the weight of that problem is set low. In this embodiment, the server 20x is equipped with the problem DB 31x shown in Figure 3(a) and the problem DB 31y shown in Figure 3(b), and it is assumed that the teacher has entered one or more of the following as attribute information of the problem: learning element name, learning element ID, curriculum guideline name, and curriculum guideline code.

[0072] The parameter information acquisition unit 42x acquires the values ​​of various parameters of the scoring calculation function, which will be described later, as parameter information. Specifically, the parameter information is information entered by the teacher using the scoring screen, which will be described later.

[0073] The similar information acquisition unit 43 and the average value calculation unit 44 are the same as in the first embodiment, so for convenience, their explanation will be omitted.

[0074] The scoring unit 45x calculates the score for each target question based on the target question information and the similarity information of each target question. Specifically, the scoring unit 45x uses a scoring function to calculate the score for each target question from the difficulty level, the correct answer rate, and the weight. In this embodiment, the scoring unit 45x calculates the score for each target question by substituting the average difficulty level, the average correct answer rate, and the weight into a function. Alternatively, the scoring unit 45x may calculate the score for each target question by further substituting the parameter values, which have been previously entered by the teacher, into the same function based on the parameter information.

[0075] The normalization unit 46x, if the target question information includes a normalization request, retrieves the total test score and normalizes it so that the sum of the points allocated to each target question equals the total test score. The total test score is entered by the teacher using the scoring screen or parameter setting screen.

[0076] The screen output unit 47x creates a scoring screen that includes the points allocated to each target question calculated by the scoring calculation unit 45x, and transmits image information related to the scoring screen to the terminal device 10. However, if the target question information includes a normalization request, the screen output unit 47x creates a scoring screen that includes the points allocated to each target question normalized by the normalization unit 46x, and transmits image information related to the scoring screen to the terminal device 10. By displaying the scoring screen based on the screen information received by the terminal device 10, the teacher can confirm the points allocated to each target question.

[0077] The screen output unit 47x may also create screens for teachers to input information about target problems and parameters, and screens to confirm whether the points set for each problem are appropriate, and transmit screen information related to these various screens to the terminal device 10. In this embodiment, although the details will be described later, the screen output unit 47x will create a scoring screen, a reference screen, a parameter setting screen, a confirmation screen, etc., to enable input of target problem information and parameter information, output of points for each target problem, referencing similar information, and confirmation of whether the points are appropriate, and will transmit information related to these various screens to the terminal device 10 in response to login and various requests.

[0078] The total score verification unit 48 obtains the total test score and verifies whether the point allocation for each target question is appropriate. The total score verification unit 48 includes a total test score acquisition unit 56, a point allocation information acquisition unit 57, and an alert output unit 58.

[0079] The test total score acquisition unit 56 acquires the total test score from the terminal device 10. The total test score is entered by the teacher using the scoring screen, parameter setting screen, confirmation screen, etc. The scoring information acquisition unit 57 acquires the scoring information of each question set by the teacher. Specifically, the scoring information is information entered by the teacher using the confirmation screen described later. The alert output unit 58 determines, based on the scoring information, whether the sum of the points for each question constituting the test matches the total test score, and outputs an alert to the terminal device 10 if they do not match. The alert output may include, for example, sending a warning message or sound to the terminal device 10 indicating that the sum of the points does not match the total test score.

[0080] In the above configuration, the problem DB31 and score DB32 of server 20x are examples of the storage unit of the present invention. Furthermore, in the above configuration, the target problem information acquisition unit 41x, parameter information acquisition unit 42x, similar information acquisition unit 43, average value calculation unit 44, score calculation unit 45x, screen output unit 47x, test total score acquisition unit 56, score information acquisition unit 57, and alert output unit 58 of server 20 are examples of the target problem information acquisition unit, parameter information acquisition unit, similar information acquisition unit, average value calculation unit, score calculation unit, output unit, total score acquisition unit, score information acquisition unit, and alert output unit of the present invention, respectively.

[0081] [Scoring screen] Next, the scoring screen created and output by the screen output unit 47x will be described. The teacher displays the scoring screen by logging into a designated site using the terminal device 10, and inputs the text of the target questions that make up the designated test, the attribute information of the target questions, the weight of the points allocated to the target questions, and whether or not normalization is performed. The teacher also inputs the parameter values ​​of the function for calculating the points from the parameter setting screen as needed. Furthermore, by pressing a designated button on the scoring screen, the teacher can request that the points allocated to each entered target question before normalization or after normalization be displayed on the screen.

[0082] Figure 13 shows an example of a scoring screen. As shown in Figure 13, the scoring screen has a target question input area 71x, an add question button 72, a similar question reference button 73, a parameter setting button 74, a score calculation button 75 and a score output area 76, a normalization selection button 77 and a total test score input area 78.

[0083] The normalization selection button 77 is a button that allows the teacher to select whether or not to normalize the points allocated to each target question, calculated by pressing the point allocation calculation button 75, with the total test score. If normalization is desired, the teacher selects the radio button for normalization selection button 77a shown in Figure 13 and enters the total test score in the total test score input area 78. On the other hand, if normalization is not desired, the teacher selects the radio button for normalization selection button 77b. If the radio button for normalization selection button 77a is selected, the target question information will include the normalization request and the total test score. Note that the total test score may be the value entered in the sum of points item 66 on the parameter setting screen described later.

[0084] The target problem input area 71x is an area where the teacher inputs the text, weight, learning element name, learning element ID, etc., of the target problem. The information about the target problem entered in the target problem input area 71x is acquired by the server 20x as target problem information. Specifically, in the target problem input area 71x, the teacher inputs the text of the target problem "4200÷6" as the problem text corresponding to number (1), "integer division" as the learning element name, and "0.8" as the weight. At this time, the teacher may also input the learning element ID associated with "integer division". In this embodiment, since the test consists of a total of 5 questions, the teacher adds a row for number (2) to the target problem input area 71x by pressing the add question button 72. Then, in the target problem input area 71x, the teacher inputs the problem text, weight, learning element name, and learning element ID corresponding to number (2). Subsequently, the teacher repeatedly presses the "Add Problem" button 72 and inputs the problem text, weight, learning element name, and learning element ID corresponding to the number, thereby inputting the problem information for the five target problems as shown in the diagram.

[0085] The similar problem reference button 73, when pressed, allows a request for a reference screen to be sent to the server 20x. Figure 14 shows an example of a reference screen. As shown in Figure 14, the reference screen displays the number of the problem that makes up the test, the target problem, the weight of the points allocated to that target problem, similar problems to that target problem, the average difficulty level of the similar problems, and the average correct answer rate in association with each other. When the server 20x receives a request for a reference screen from the terminal device 10, it creates the reference screen and sends screen information related to the reference screen to the terminal device 10. By displaying the reference screen based on the screen information received by the terminal device 10, the teacher can easily check similar problems, their average difficulty level, average correct answer rate, etc. for each target problem that they have entered.

[0086] The parameter setting button 74, when pressed, can request the parameter setting screen from the server 20x. Figure 15 shows an example of the parameter setting screen. As shown in Figure 15, the parameter setting screen includes a graph 60x representing a function for calculating the score, a difficulty range item 61, a score range item 62, a correct answer rate range item 63, an x-axis slope item 64, a y-axis slope item 65, and a total score item 66.

[0087] Graph 60x represents the function for calculating the scoring, and as shown in the figure, it is a three-dimensional graph composed of x, y, and z axes. Specifically, the x axis represents difficulty, the y axis represents the correct answer rate, and the z axis represents the scoring. Server 20x calculates the scoring for each target problem by multiplying the value obtained from the plane equation "z=ax+by+c" in graph 60x by the weight "w" for each target problem. Difficulty Range Item 61 is an item for setting and inputting the range of numerical values ​​indicating difficulty. Scoring Range Item 62 is an item for setting and inputting the range of scoring. Correct Answer Rate Range Item 63 is an item for setting and inputting the range of correct answer rates. x-axis Slope Item 64 is "a" in the plane equation, and is an item for setting and inputting the value of the parameter corresponding to difficulty, i.e., the slope of the x axis. y-axis Slope Item 65 is "b" in the plane equation, and is an item for setting and inputting the value of the parameter corresponding to the correct answer rate, i.e., the slope of the y axis. The "Total Score" item 66 is where you set and input the sum of the points allocated to each question, i.e., the total score of the test. In this embodiment, the total test score is set to "100 points".

[0088] When server 20x receives a request for a parameter setting screen from terminal device 10, it creates the parameter setting screen and sends screen information related to the parameter setting screen to terminal device 10. By displaying the parameter setting screen based on the screen information received by terminal device 10, the teacher can use the parameter setting screen to send various ranges and parameter values ​​that they have arbitrarily set to server 20x.

[0089] The teacher uses terminal device 10 to input the target problem information and parameter information, and then presses the score calculation button 75 to send a request for score output to server 20x. Upon receiving the score output request, server 20x uses a score calculation function to calculate the score for each target problem. Server 20x then creates a score screen that includes the score for each target problem and sends screen information related to this score screen to terminal device 10. In this way, when the teacher presses the score calculation button 75, the score for each target problem is displayed on the score screen.

[0090] Specifically, Server 20x calculates the points allocated to each target problem by multiplying the value obtained from the plane equation "z = ax + by + c" in Graph 60x by a weight "w" for each target problem. Server 20x substitutes the average difficulty value for "x" and the average correct answer rate for "y" in the plane equation. Server 20 calculates the "z" of the plane equation corresponding to each target problem by substituting the corresponding parameter values ​​for "a", "b", and "c" in the plane equation based on the parameter information. Server 20x then calculates the points allocated to each target problem by multiplying the calculated value "z" by a value representing the weight "w" for the points allocated to each target problem. Figure 16(a) is a diagram illustrating the points allocated before weighting. As shown in Figure 16(a), the points allocated to each target problem before weighting are calculated by a function for calculating points, according to the average difficulty value and average correct answer rate of similar problems. Figure 16(b) is a diagram illustrating the points allocated before normalization. As shown in Figure 16(b), the score for each target problem before normalization is calculated by multiplying the score before weighting by the weight "w" for each target problem.

[0091] If the target problem information includes a normalization request, server 20x retrieves the total test score and normalizes it so that the sum of the points allocated to each target problem equals the total test score. Specifically, the total test score may be the score entered in the total test score input area 78 on the scoring screen, or the sum of the points included in the parameter information may be considered as the total test score. Figure 17 is a diagram illustrating how to normalize so that the sum of the points allocated before normalization equals 100 points. Server 20x calculates the "normalized score" by substituting the "sum of points allocated before normalization" and the "points allocated before normalization" for each target problem into equation 70 as shown in the figure. For example, in the case of the scoring shown in Figure 16(b), the sum of the scores for target questions (1) to (5) before normalization is (8 + 8 + 33 + 61 + 84) = "194", and the score for target question (1) before normalization is "8". Substituting these into equation 70, the normalized score for target question (1) becomes "4" as shown in the figure. Server 20x similarly calculates the normalized scores for target questions (2) to (5).

[0092] Here, if the sum of the normalized scores falls short of 100 points as a result of normalization, server 20x adjusts by adding it to the score of any target question. Specifically, in the example in Figure 17, the adjustment is made by adding 1 point to the score of target question (5). Figure 18 shows the score of target questions (1) to (5) before and after normalization. If the target question information does not include a normalization request, the score of the target question before normalization is displayed in the score output area 76 on the right side of the score screen shown in Figure 13. On the other hand, if the target question information includes a normalization request, the score of the target question after normalization is displayed in the score output area 76 on the right side of the score screen shown in Figure 13.

[0093] Furthermore, when logging into the designated site, the scoring screen will only display the range indicated by the input on the left side of the screen as shown in Figure 13. Pressing the scoring calculation button 75 will display the range indicated by the output on the right side of the screen. Also, the scoring screen shown in Figure 13, the reference screen shown in Figure 14, and the parameter setting screen shown in Figure 15 are examples, and the present invention is not limited thereto; the configuration of various screens can be set arbitrarily.

[0094] [Confirmation screen] Next, the confirmation screen created and output by the screen output unit 47x will be described. The teacher displays the confirmation screen by logging into a designated site using the terminal device 10 and performs the test editing work. Figure 19 is an example of a confirmation screen. As shown in Figure 19, the confirmation screen has a test layout display area 81, a total test score input area 82, a point allocation input area 83, and an OK button 84. The test layout display area 81 is an area for displaying and editing the test layout, as shown in Figure 19. The total test score input area 82 is an area where the teacher inputs the total test score. The point allocation input area 83 is an area for inputting the point allocation for each question that makes up the test. The numbers displayed in the test layout display area 81 correspond to the question numbers in the point allocation input area 83.

[0095] The teacher uses terminal device 10 to input the scores for each question that makes up the test into the score input area 83 on the confirmation screen, referring to the unnormalized scores displayed on the score screen. The teacher also inputs the total score for any test into the total test score input area 82. After the teacher has finished inputting into the score input area 83 and the total test score input area 82, they press the OK button 84 to send the score information, including information about the scores for each question, and the total test score to the server 20x. Based on the acquired score information and total test score, the server 20x determines whether the sum of the scores for each question that makes up the test matches the total test score, and outputs an alarm to terminal device 10 if they do not match.

[0096] [Score calculation process] Next, we will explain the process of calculating and outputting the points allocated to the test questions based on their correct answer rate, difficulty level, and weight. Figure 20 is a flowchart of the point allocation process performed by server 20x. This process is achieved by server 20x executing a pre-prepared program.

[0097] The teacher uses terminal device 10 to log in to a designated site to display the scoring screen, where they input target question information and parameter information, and refer to similar information. Once input and referencing are complete, the teacher presses a designated button on the scoring screen to send the target question information and parameter information to server 20x and request the output of the scoring for each target question.

[0098] Server 20x obtains target problem information from terminal device 10 (step S21). Server 20x also obtains parameter information from terminal device 10 (step S22). Then, based on the attribute information of the target problem included in the target problem information, Server 20x obtains the text and difficulty level of similar problems from problem DB 31 (step 23). Server 20x also obtains the correct answer rate of the similar problems from score DB 32 (step S24). Then, Server 20x calculates the average difficulty level and average correct answer rate of the target problem (step S25). Server 20x calculates the score of the target problem by multiplying the value obtained by substituting the average difficulty level and average correct answer rate, along with the values ​​of various parameters included in the parameter information, into a function for calculating the score by a value indicating the weight of the score for the target problem (step S26).

[0099] Server 20x determines whether it has calculated the points for all target questions (step S27). If it determines that it has not calculated the points for all target questions (step S27; No), Server 20x returns to the process in step S23 and calculates the points for the target questions that have not yet been calculated. On the other hand, if it determines that it has calculated the points for all target questions (step S27; Yes), Server 20 determines whether the target question information includes a normalization request (step S28). If a normalization request is included (step S28; Yes), Server 20x normalizes the points so that the sum of the points equals the total test score. Then, Server 20x creates a scoring screen including the normalized points and sends the screen information related to the scoring screen to the terminal device 10 (step S30).

[0100] On the other hand, if the target problem information does not include a normalization request (step S28; No), the server 20x creates a scoring screen that includes the scoring before normalization and sends the screen information related to that scoring screen to the terminal device 10 (step S29). This completes the scoring calculation process. The terminal device 10 displays the scoring screen based on the screen information received from the server 20x.

[0101] [Confirmation process] Next, we will explain the verification process to confirm whether the scoring of the test set by the teacher or other person is appropriate. Figure 21 is a flowchart of the verification process performed by server 20x. This process is achieved by server 20x executing a pre-prepared program.

[0102] The teacher uses terminal device 10 to log in to a designated site to display a confirmation screen and enters the total test score and the point distribution for each question that makes up the test. Once the input is complete, the teacher presses a designated button on the confirmation screen to send the point distribution information regarding the total test score and the point distribution for each question to server 20x and requests confirmation of the point distribution.

[0103] Server 20x obtains scoring information from terminal device 10 (step S41). Server 20x also obtains the total test score from terminal device 10 (step S42). Then, based on the obtained scoring information and total test score, server 20x determines whether the sum of the points allocated to each question that makes up the test matches the total test score (step S43). If it determines that they do not match (step S43; No), server 20x outputs an alarm to terminal device 10. This completes the verification process.

[0104] In this embodiment, the verification process obtains the total test score from the scoring information, but the present invention is not limited to this. As long as the scoring calculation process and the verification process are linked, the sum of the scores included in the parameter information may be obtained as the total test score. Furthermore, the total test score may be arbitrarily set in advance.

[0105] The learning support system 100x of this embodiment can calculate the points allocated to each question, taking into account the difficulty level and correct answer rate of the questions, as well as weights arbitrarily set by the teacher or others. Furthermore, if the teacher wishes to normalize the points allocated, the system can output the normalized points, which can then be applied directly to the test. On the other hand, if the teacher does not wish to normalize the points allocated, the system can output the points before normalization, which the teacher can use as a reference to set the test scores.

[0106] Furthermore, the learning support system 100x of this embodiment can pre-set the total test score when editing the test layout, etc., and check whether the sum of the points allocated to each question that makes up the test matches the total test score. If they do not match, it can output an alert to draw the attention of the learning supporter. This prevents problems such as noticing and correcting errors in the point allocation settings during or after grading the test. In this way, by preventing the need for recalculation and reanalysis due to point allocation corrections, the learning supporter can quickly perform aggregation and analysis after grading and check the learner's results.

[0107] <Variation> (First variation) In the first and second embodiments described above, similarity information is obtained from the problem DB31 based on the attribute information of the target problem. However, the present invention is not limited thereto, and similarity information may also be obtained from the problem DB31 based on the text similarity of the target problem. Specifically, text similarity is used to identify similar problems to the target problem from among the problems stored in the problem DB31, and information regarding those similar problems is obtained as similarity information.

[0108] Text similarity is a measure of similarity in the feature space, where closer problems have a higher similarity score, approaching 1, and farther problems have a lower similarity score, approaching 0. Figure 22 illustrates the identification of similar problems using text similarity. Server 20, for example, treats equation 85, a predetermined target problem shown in Figure 22(a), as text character by character, such as "y" and "=" as shown in Figure 22(b), and extracts the features of the target problem numerically by using machine learning methods capable of processing time-series information as text, such as LSTM (Long Short Term Memory). Specifically, Server 20 takes "y=4x+6" as input to the LSTM and trains it to output the same text "y=4x+6". After training is complete, Server 20 extracts and utilizes the features of the intermediate layer of the LSTM. Similarly, Server 20 numerically extracts the features of all problems stored in the problem DB 31. The server 20 then calculates the difference between the features of the target problem and the features of each problem stored in the problem database, and identifies problems whose differences are within a threshold as similar problems to the target problem.

[0109] Furthermore, the acquisition of similarity information using text similarity can be applied not only to problems involving mathematical formulas, but also to any problem, such as reading comprehension questions in Japanese language. Also, this modified example is not limited to the first embodiment, but can also be applied to the server 20x of the second embodiment.

[0110] (Second variation) In the first and second embodiments described above, the scoring is calculated based on the difficulty level and correct answer rate of similar problems. However, the present invention is not limited thereto, and the scoring may be calculated based on the difficulty level or correct answer rate of the target problem.

[0111] Furthermore, in the first and second embodiments described above, the scoring is calculated using the average difficulty level and average correct answer rate of similar problems. However, the present invention is not limited thereto, and if the difficulty level and correct answer rate of the target problem are known, the scoring may be calculated using those difficulty levels and correct answer rates.

[0112] Furthermore, while the first and second embodiments described above use the equation of a plane as the function for calculating the scoring, the present invention is not limited thereto, and any other equation, such as the equation of a curved surface, can be applied. In addition, multiple functions that can be used as the function for calculating the scoring may be registered in the server 20 or 20x, and the teacher may select a function from among them to use.

[0113] (Third variation) In the first embodiment described above, the learning supporter uses a terminal device 10, but the present invention is not limited thereto, and the learning supporter may use a teacher terminal 90 that has the functions of a server 20. The teacher terminal 90, like the server 20, is, for example, a personal computer or a general-purpose tablet PC (personal computer).

[0114] Figure 23 shows an example configuration of the learning support system 200 in this case. The teacher terminal 90 is connected to the problem DB 91 and the grade DB 92. The teacher terminal 90 can perform the scoring calculation process that was performed by the server 20, obtain similar information on the target problems, and calculate the points for each target problem according to the difficulty level and correct answer rate. In this case, the teacher terminal 90 is an example of the information processing device of the present invention.

[0115] Furthermore, in the second embodiment described above, the learning supporter may also use a teacher terminal 90 that has the functionality of server 20x. In this case, the teacher terminal 90 can perform the scoring calculation process that was performed by server 20x to obtain similarity information of the target questions and calculate the points for each target question according to the difficulty level, correct answer rate, and weight. In addition, the teacher terminal 90 can perform the verification process that was performed by server 20x to confirm whether the point distribution of the questions that make up the test is appropriate. [Explanation of Symbols]

[0116] 5 Network 10 Terminal devices 20, 20x servers 31, 31x, 31y problem DB 32 Grades DB 41, 41x Target Problem Information Acquisition Unit 42 Parameter Information Acquisition Unit 43 Similar information acquisition unit 44. Average Value Calculation Unit 45, 45x Score calculation unit 46, 46x normalization section 47, 47x screen output section 48. Total Score Confirmation Section 51 Similar Problem Text Acquisition Section 52 Similar problem difficulty level acquisition part 53. Similar Problem Correct Answer Rate Acquisition Section 54. Calculation section for the average difficulty level 55. Calculation unit for the average correct answer rate 56 Test Total Score Acquisition Section 57. Scoring Information Acquisition Department 58 Alert Output Section 100, 100x, 200 Learning Support System

Claims

1. A scoring information acquisition unit that acquires scoring information regarding the point distribution of each question that makes up the test, A total score acquisition unit that acquires the total score of the aforementioned test, Based on the aforementioned scoring information, an alert output unit determines whether the sum of the points allocated to each question constituting the test matches the total score of the test, and outputs an alert if it determines that they do not match. A memory unit that stores information about multiple problems, The aforementioned test comprises a target question information acquisition unit that acquires target question information regarding target questions for which scoring is to be set, Based on the aforementioned target problem information, a similar information acquisition unit acquires similar information about similar problems similar to each target problem from the storage unit, A scoring calculation unit calculates the points allocated to each target problem based on the aforementioned target problem information, similarity information for each target problem, and a function used for calculating the points allocated to each target problem. An output unit that outputs the aforementioned scoring, An information processing device equipped with the following features.

2. The memory unit stores the text of the problem, the attribute information of the problem, the difficulty level of the problem, and the correct answer rate of the problem in association with each other. The aforementioned target problem information includes the text of the target problem, attribute information of the target problem, and weights related to the scoring of the target problem, The similar information acquisition unit acquires the difficulty level and correct answer rate of similar problems that are similar to each target problem from the storage unit based on the target problem information, as similar information. The information processing apparatus according to claim 1, wherein the scoring calculation unit calculates the scoring for each target problem based on the weights related to the scoring of each target problem, the difficulty level and correct answer rate of similar problems similar to each target problem, and the function.

3. Based on the aforementioned similar information, the system includes an average value calculation unit that calculates the average difficulty level and average correct answer rate of similar problems to each target problem. The information processing apparatus according to claim 2, wherein the scoring unit calculates the points allocated to each target question by substituting the average value of the difficulty level and the average value of the correct answer rate corresponding to each target question into the function, and then multiplying by a value indicating the weight corresponding to each target question.

4. The system includes a parameter information acquisition unit that acquires the values ​​of the parameters of the aforementioned function as parameter information, The information processing apparatus according to claim 3, wherein the scoring calculation unit further substitutes the values ​​of the parameters into the function to calculate the scoring for each target problem.

5. A program executed by an information processing device comprising a computer and a memory unit that stores information about multiple problems, A scoring information acquisition unit that obtains scoring information regarding the point distribution of each question that makes up the test. A total score acquisition unit that acquires the total score of the aforementioned test. An alert output unit determines, based on the aforementioned scoring information, whether the sum of the points allocated to each question constituting the test matches the total score of the test, and outputs an alert if it determines that they do not match. A target question information acquisition unit that constitutes the aforementioned test and acquires target question information regarding target questions for which scoring is to be set. A similar information acquisition unit acquires similar information about similar problems similar to each target problem from the storage unit based on the aforementioned target problem information. A scoring calculation unit calculates the points allocated to each target problem based on the aforementioned target problem information, similarity information for each target problem, and a function used for calculating the points allocated to each target problem. Output unit that outputs the aforementioned scoring, A program that causes the aforementioned computer to function.

Images