Incorrect behavior estimation device, incorrect behavior estimation method, and storage medium

Abstract

The invention provides an inappropriate behavior estimation device, an inappropriate behavior estimation method, a storage medium, and a technology capable of monitoring an inappropriate behavior using a Web browser or an inappropriate behavior using a stored file in CBT. The answer acquisition unit acquires, as an answer, a character string input to an answer column for a question. The operation history information acquisition unit acquires operation history information indicating an operation history for an input operation of an answer to the answer column. The estimation unit estimates, on the basis of the operation history information acquired by the operation history information acquisition unit, whether or not an improper behavior has been performed in the answer to the question.

Description

Technical Field

[0001] The present invention relates to a technology for conducting an experiment to judge and evaluate the knowledge or academic ability of a user using an information processing device, and more particularly to an inappropriate behavior presumption device, an inappropriate behavior presumption method, and a storage medium. Background Art

[0002] In the past, there was a computer-based test (CBT) as a method of allowing a user to operate an information processing device to judge the user's knowledge or academic ability and conduct an evaluation. In CBT, test questions are displayed on the screen of a display of an information processing device operated by a user as a subject. The user performs input operations in the information processing device, that is, inputs answers to the test questions displayed on the screen of the display. The user's knowledge or academic ability is judged based on the scoring results of the user's answers. The information processing device used in CBT is, for example, a personal computer or a tablet terminal.

[0003] Furthermore, a system for monitoring inappropriate behavior (cheating) of CBT subjects (users) has also been proposed (see Patent Document 1). The system of Patent Document 1 processes an image of a subject captured by a camera included in an information processing device, detects changes in the subject's line of sight or behavior of leaving the seat, and determines whether the subject has engaged in inappropriate behavior based on these detection results.

[0004] [Prior art literature]

[0005] [Patent Document]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2022-52144 Summary of the invention

[0007] [Problems to be solved by the invention]

[0008] However, the system described in Patent Document 1 does not monitor improper behavior using a Web browser installed in an information processing device or improper behavior using a file stored in the information processing device.

[0009] Inappropriate behavior using a web browser includes, for example, cheating by using keywords obtained from a given question as search terms to search the web and answer the questions by referring to the retrieved web pages. In addition, inappropriate behavior using a file stored in an information processing device includes, for example, cheating by answering the questions by referring to a file stored in advance in the information processing device (equivalent to a so-called cheat sheet).

[0010] Furthermore, research is being conducted to implement CBT in a form where the subject uses an information processing device owned by the subject. In this form, the CBT operator does not need to prepare an information processing device to be loaned to each subject, and the cost of implementation can be suppressed. Therefore, it is desirable to develop a system that can monitor the inappropriate behavior of the subject.

[0011] An object of the present invention is to provide a technology capable of monitoring improper behavior using a Web browser or improper behavior using a stored file.

[0012] [Technical means to solve the problem]

[0013] In order to achieve the above-mentioned object, the improper behavior estimation device of the present invention is constructed as follows.

[0014] The answer acquisition unit acquires a character string input into the answer column of the question as an answer. The operation history information acquisition unit acquires operation history information indicating the operation history of the input operation of the answer to the answer column. The estimation unit estimates whether an improper behavior was performed in answering the question based on the operation history information acquired by the operation history information acquisition unit.

[0015] For example, the operation history information acquisition unit acquires a history of establishing a correspondence between a character string input into the answer column and an input operation (a paste operation or an operation of a letter key) of the character string. The inference unit detects, for example, a character string input into the answer column by a paste operation, and infers whether the character string was input by an improper behavior. For example, the inference unit infers that an improper behavior was performed in answering the question when the character string input into the answer column by a paste operation is not a character string copied in the window prompting the question or is not a character string existing in the window prompting the question.

[0016] This makes it possible to monitor improper behavior of a user (subject) who pastes a character string copied from a web page retrieved by a web browser or a stored file (a cheating file) into an answer column to answer a question.

[0017] Furthermore, for example, the configuration may be as follows: the configuration includes a transition history information acquisition unit that acquires transition history information indicating a transition history of a window that presents a question between an active state in which the focus is on and an inactive state in which the focus is off, and the estimation unit estimates whether an improper behavior has been performed in answering the question using the transition history information acquired by the transition history information acquisition unit. For example, the estimation unit may compare the frequency of the window that presents a question from an active state to an inactive state, or from an inactive state to an active state, with a preset threshold value to estimate whether an improper behavior has been performed in answering the question.

[0018] This makes it possible to monitor improper behavior of a user (subject) who enters characters into an answer column by operating the letter keys while referring to a Web page retrieved by a Web browser or a stored file (a cheating file) and answers the question.

[0019] [Effects of the Invention]

[0020] According to the present invention, it is possible to monitor improper behavior using a Web browser or improper behavior using a stored file. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram showing a network system for implementing CBT (Computer Based Testing).

[0022] Figure 2 (A) is a diagram showing an example of a screen in which a window for displaying a question sent from a server device is displayed on a user terminal. Figure 2 (B) is a diagram showing an example of a screen of a window in the middle of inputting an answer to a question distributed from a server device. Figure 2 (C) is a diagram showing an example of a window screen when the input of the answer to the question distributed from the server device is completed.

[0023] Figure 3 This is a block diagram showing the structure of the main parts of the server device.

[0024] Figure 4 This is a block diagram showing the structure of the main parts of the user terminal.

[0025] Figure 5 This is a graph showing operation history information.

[0026] Figure 6 This is a graph showing operation history information.

[0027] Figure 7 This is a graph showing operation history information.

[0028] Figure 8 This is a flowchart showing the operation of the user terminal in the above example.

[0029] Fig. 9 This is a flowchart showing the operation of the server device in the above example.

[0030] Fig.10 It is a flowchart which shows the estimation process.

[0031] Fig.11 This is a block diagram showing the structure of the main parts of the server device according to the second modification.

[0032] Fig.12This is a flowchart showing the estimation process of Modification Example 2.

[0033] Fig.13 This is a flowchart showing the estimation process of Modification 3.

[0034] [Explanation of Symbols]

[0035] 1.1A: Server device

[0036] 2: User terminal

[0037] 5: Network

[0038] 11, 11A: Control unit

[0039] 11a: Answer acquisition section

[0040] 11b: Operation history acquisition unit

[0041] 11c: Presumption

[0042] 11d: Transformation of historical information acquisition department

[0043] 12: Question Database (Question DB)

[0044] 13: Ministry of Communications

[0045] 21: Control unit

[0046] 21a: Answer determination section

[0047] 21b: Operation history information generation unit

[0048] 22: Display

[0049] 23: Operation Department

[0050] 24: Ministry of Communications DETAILED DESCRIPTION

[0051] Hereinafter, embodiments of the present invention will be described.

[0052] <1. Application examples>

[0053] Figure 1 This is a schematic diagram showing the network system for implementing CBT (Computer Based Testing). Figure 1 As shown, the network system of the example is configured such that a plurality of user terminals 2 are connected to a server device 1 via a network 5 so as to enable data communication. The example is an example in which the misconduct estimation device of the present invention is applied to the server device 1.

[0054] In addition, as described in the modification example below, the malpractice presumption device of the present invention can also be applied to the user terminal 2 instead of the server device 1.

[0055] The user terminal 2 is a personal computer or a tablet terminal operated by a user (subject) who receives a test implemented in the CBT method. In the above example, the user terminal 2 can be a terminal owned by the user, or a terminal lent by the CBT operator to the subject. In other words, in the above example, the user can use the user terminal 2 owned by himself / herself to receive CBT. Moreover, the user can also receive CBT in a personal space such as his / her home instead of at the test venue. When the CBT operator allows the use of the user terminal 2 owned by the user for the test, the CBT operator does not need to prepare a user terminal 2 to lend to the user as the subject. When the CBT operator allows the user to take the test in a personal space, the CBT operator does not need to prepare a test venue.

[0056] In the above example, the server device 1 distributes (sends) test questions to the user terminal 2 operated by the user. Moreover, the server device 1 scores the answers of the user sent from the user terminal 2, etc., and presumes whether the user has committed malpractice (cheated).

[0057] For example, the server device 1 distributes the following question to the user terminal 2:

[0058] "Translate the following Japanese into English.

[0059] ‘John, hello, how are you?’"

[0060] The user terminal 2 is a generally popular information processing terminal (personal computer or tablet terminal), and the display can display not only windows related to CBT, but also multiple windows. As is well known, a window is a display area on the screen generated by the application program that allocates the window. For example, there are windows related to CBT, windows related to web pages, windows related to saved files, etc. The user terminal 2 receives input operations of the user on the window (active window) on which the focus is focused.

[0061] Figure 2 (A) of is a diagram of an example of a screen showing a window for the user terminal to display the question distributed from the server device. Figure 2 (A) of Figure 2 (B) of Figure 2 (C) of shown are windows related to CBT. The user operates the input device (mouse, keyboard, touch screen, etc.) included in the user terminal 2 to answer the question distributed from the server device 1 (input the answer into the answer column).

[0062] Figure 2(B) is a diagram showing an example of a screen of a window in the middle of inputting an answer to a question distributed from a server device. Figure 2 (C) is a diagram showing an example of a window screen when the input of the answer to the question distributed from the server device is completed. Figure 2 (B) indicates the time when the user enters "Hello!" in the answer field. Figure 2 (C) indicates the time when the user enters "Hello! John, How are you?" in the answer column.

[0063] The user terminal 2 obtains input operation information indicating the content of the input operation performed by the user on the window related to the CBT, and generates operation history information in which the obtained input operation information is arranged in a time series. For example, when a character string is input in the answer column, the user terminal 2 obtains input operation information including the type of input operation at that time (operation of letter keys, paste operation, etc.) and the character string input in the answer column. Moreover, even when a character string is not input in the answer column, the user terminal 2 obtains input operation information indicating the operation, such as when performing a copy operation or an operation related to switching the focus of the window related to the CBT.

[0064] The user terminal 2 transmits the user's answer to the question (the character string input into the answer field) and operation history information when answering the question to the server device 1 .

[0065] The server device 1 scores the user's answer received from the user terminal 2. Furthermore, the server device 1 infers whether the user (subject) operating the user terminal 2 has performed improper behavior (cheating) based on the operation history information received from the user terminal 2. For example, the server device 1 infers whether the input of the string by the paste operation is improper behavior based on whether the string input by the paste operation into the answer column is the string copied in the window (window related to CBT) that prompts the question. By making the above inference, the server device 1 monitors whether the user has performed improper behavior in using the web browser or improper behavior in using the saved file.

[0066] <2. Construction example>

[0067] Figure 3 1 is a block diagram showing the configuration of the main parts of the server device of the above example. The server device 1 includes a control unit 11 , a question database 12 (question DB 12 ), and a communication unit 13 .

[0068] The control unit 11 controls each part of the server device 1. The control unit 11 includes an answer acquisition unit 11a, an operation history acquisition unit 11b, and an estimation unit 11c. The answer acquisition unit 11a, the operation history acquisition unit 11b, and the estimation unit 11c included in the control unit 11 are described below.

[0069] The question DB 12 is a database in which questions distributed to the user terminal 2 operated by the user are registered. In the above example, the correct answer (standard answer) of each question is registered in the question DB 12 in correspondence with each question.

[0070] The communication unit 13 performs data communication with the user terminal 2 operated by the user via the network 5. The communication unit 13 transmits (distributes) questions to the user terminal 2. The communication unit 13 also receives answers to the questions from the user terminal 2 and operation history information when answering the questions.

[0071] Next, the answer acquisition unit 11 a , the operation history acquisition unit 11 b , and the estimation unit 11 c included in the control unit 11 will be described.

[0072] The answer acquisition unit 11a acquires the answer to the question (the user's answer) received from the user terminal 2 in the communication unit 13. Then, the answer acquisition unit 11a scores the acquired user's answer.

[0073] The operation history acquisition unit 11b acquires the operation history information received from the user terminal 2 in the communication unit 13. The operation history information will be described below.

[0074] The estimation unit 11 c estimates whether there is a high possibility that the user has performed an improper behavior (cheating) when answering a question, based on the operation history information acquired by the operation history acquisition unit 11 b .

[0075] In the example, the control unit 11 of the server device 1 includes a hardware central processing unit (CPU), a memory, and other circuits. When the hardware CPU executes the misconduct inference program of the present invention, it operates as an answer acquisition unit 11a, an operation history acquisition unit 11b, and an inference unit 11c. Moreover, the memory includes an area for expanding the misconduct inference program of the present invention, or an area for temporarily storing data generated when executing the misconduct inference program. The control unit 11 can also be a large-scale integration (LSI) that integrates a hardware CPU, a memory, etc. Moreover, in the example, the hardware CPU is a computer that executes the misconduct inference method of the present invention.

[0076] Figure 4 2 is a block diagram showing the configuration of the main parts of the user terminal of the above example. The user terminal 2 includes a control unit 21 , a display 22 , an operation unit 23 , and a communication unit 24 .

[0077] The control unit 21 controls the operation of each part of the user terminal 2. The control unit 21 includes an answer determination unit 21a and an operation history information generation unit 21b. The answer determination unit 21a and the operation history information generation unit 21b included in the control unit 21 are described below.

[0078] The display 22 displays questions distributed from the server device 1 , displays answers to questions input by the user, and the like.

[0079] The operation unit 23 detects the input operation of the user. The operation unit 23 may be configured to accept character input through a hardware keyboard or a software keyboard. Furthermore, the operation unit 23 may include a pointing device such as a mouse.

[0080] The communication unit 24 performs data communication with the server device 1 via the network 5 .

[0081] Next, the answer determination unit 21 a and the operation history information generation unit 21 b included in the control unit 21 will be described.

[0082] When an input operation related to completing the answer to the question displayed on the display 22 is performed, the answer determination unit 21 a determines the character string input into the answer field as the answer to the question.

[0083] The operation history information generating unit 21 b generates operation history information in which input operation information when answering questions is arranged in time series. Figure 5 , Figure 6 ,and Figure 7 This is a graph showing operation history information. Figure 5 to Figure 7 The problem number shown is the number of the specific problem (the identification number of the problem). Figure 5 to Figure 7 In order to explain that the generated operation history information is different depending on the user's input operation, the question No. is set to the same number.

[0084] Each behavior of the operation history information represents input operation information of the content of the input operation performed by the user on the window related to CBT. The input operation information includes the operation date and time, the operation category, and the input characters. The operation date and time represents the date and time when the input operation is performed. The operation category is, for example, an operation to deviate the focus from the window related to CBT (focus end), an operation to focus the focus on the window related to CBT (focus start), a copy operation to save a character string to the clipboard (copy), a paste operation to paste a character string saved to the clipboard (paste), and an operation of letter keys (key-down). The input characters are the characters (or character strings) entered into the answer column. There are operation categories where there are no characters entered into the answer column, and there are also operation categories where multiple characters (character strings) are entered into the answer column.

[0085] The control unit 21 of the user terminal 2 includes a hardware CPU, a memory, and other circuits. The hardware CPU operates as an answer determination unit 21a and an operation history information generation unit 21b. In addition, the memory includes an area for temporarily storing a character string input into the answer column and operation history information. The control unit 21 can be an LSI that integrates the hardware CPU, the memory, etc.

[0086] <3. Action example>

[0087] Figure 8 This is a flowchart showing the operation of the user terminal in the above example. Fig. 9 This is a flowchart showing the operation of the server device in the above example.

[0088] The server device 1 transmits the question to each user terminal 2. The user terminal 2 displays the question received from the server device 1 in a window related to CBT.

[0089] After the user terminal 2 detects that the user has performed an input operation on the window related to the CBT, the input operation information related to the input operation is obtained (s21, s22). As described above, the input operation is, for example, an operation to deviate the focus from the window related to the CBT (focus end), an operation to focus on the window related to the CBT (focus start), a copy operation to save a character string to the clipboard (copy), a paste operation to paste a character string saved to the clipboard (paste), and an operation of a letter key (key press).

[0090] If the window in which the user performs input operation is a window other than the window related to CBT, the user terminal 2 does not perform the processing after s22. Furthermore, if the window in which the user performs input operation is a window related to CBT and the input operation is to input characters related to the answer to the question, the user terminal 2 inputs the characters corresponding to the input operation into the answer column.

[0091] After the user terminal 2 acquires the input operation information related to the current input operation in s22, it updates the operation history information (s23). In s23, the input operation information related to the current input operation acquired in s22 is added to the operation history information.

[0092] The user terminal 2 determines whether the user has completed the answer to the question (s24). If the answer is not completed, it returns to s21. When the user has completed the answer to the question, the user terminal 2 sends the answer to the question and the operation history information related to the question to the server device 1 (s25), and returns to s21. When the user terminal 2 determines that the user has completed the answer to the question, it obtains the character string entered into the answer column of the question as the answer. In s25, the user terminal 2 sends the personal specific information that can identify the user who performed the operation to the server device 1. The personal specific information can be, for example, the test number assigned to the user in this CBT, or an Internet Protocol (IP) address.

[0093] Next, the operation of the server device 1 is described. The server device 1 waits to receive the answer to the question and the operation history information related to the question from the user terminal 2 using the communication unit 13 (s1). After receiving the answer to the question and the operation history information related to the question from the user terminal 2, the server device 1 obtains the answer to the question received by the answer acquisition unit 11a and scores the answer (s2, s3). The scoring of the answer in s3 can be performed, for example, by a deduction method or a point addition method, or by other methods, or by a scoring method using artificial intelligence (AI).

[0094] Furthermore, the operation history acquisition unit 11b acquires the operation history information received from the user terminal 2 (s4). The estimation unit 11c performs estimation processing based on the operation history information acquired by the operation history acquisition unit 11b through s4, that is, estimates whether the possibility that the user has performed inappropriate behavior when answering the question is high (s5). The estimation unit 11c associates the estimation result of s5 with the user (s6) and returns it to s1.

[0095] The estimation process related to s5 will be described. Fig.10is a flowchart showing the estimation process. The estimation unit 11c determines whether there is a character string input by a paste operation among the character strings input into the answer column (s51). The estimation unit 11c selects all the character strings input into the answer column and input by a paste operation. The estimation unit 11c refers to the operation history information and selects the input operation information whose operation type is paste.

[0096] If the character string input by the paste operation does not exist in the character strings input into the answer field, the estimation unit 11c estimates that there is no possibility that the user has performed inappropriate behavior when answering the question (s53), and ends the present process.

[0097] If there is a character string input by a paste operation among the character strings input into the answer column, the estimation unit 11c determines whether the character string (the character string input into the answer column by a paste operation) has been copied in the window related to the CBT (s52).

[0098] For example, in s52, the inference unit 11c refers to the operation history information and detects the operation (focus start) of focusing on the window related to the CBT that is before the operation of pasting the object and is closest in time to the operation of pasting the object. The inference unit 11c determines whether there is a copy operation between the detected operation of focusing on the window related to the CBT and the operation of pasting the object. If there is a copy operation between the detected operation of focusing on the window related to the CBT and the operation of pasting the object, the inference unit 11c infers that there is no possibility that the user has performed an inappropriate behavior (s53) and ends the present processing. On the other hand, if there is no copy operation between the detected operation of focusing on the window related to the CBT and the operation of pasting the object, the inference unit 11c infers that there is a high possibility that the user has performed an inappropriate behavior (s54) and ends the present processing.

[0099] When there is no copy operation between the detected operation of focusing on the window related to the CBT and the operation of pasting the object, it is highly likely that the character string pasted by the operation of pasting the object is a character string copied in a window other than the CBT, not a character string copied in the window related to the CBT. When it is highly likely that the character string pasted by the operation of pasting the object is a character string copied in a window other than the CBT, the inference unit 11c infers that it is highly likely that the user has performed an improper behavior.

[0100] In addition, the estimation unit 11c performs the following operations on the character string input into the answer column by the paste operation: Fig.10In the process shown, if it is estimated that the possibility of the user's improper behavior in s54 is high for at least one character string, it is estimated that the possibility of the user's improper behavior in the question is high. In other words, if there is no character string input into the answer column by pasting operation, or if it is estimated that there is no possibility of the user's improper behavior in s53 for all the character strings input into the answer column by pasting operation, the estimation unit 11c estimates that there is no possibility of the user's improper behavior in the question.

[0101] As described above, the server device 1 of the network system of the example monitors whether the user has performed inappropriate behavior based on whether the character string input into the answer column by the paste operation is the character string copied in the window related to CBT. Therefore, the network system of the example can monitor inappropriate behavior using the web browser installed in the user terminal 2 or inappropriate behavior using the file stored in the information processing device. Moreover, CBT can be implemented in a form that allows the subject to use the terminal (user terminal 2) owned by the individual, and the cost of implementing CBT can be suppressed.

[0102] <4. Modifications>

[0103] Modification 1

[0104] The configuration of the example is that the estimation unit 11 c estimates whether there is a possibility of improper behavior by the user based on whether there is a copy operation between the detected operation of focusing on the window related to the CBT and the operation of pasting the object.

[0105] In the modification example 1, if there is a character string input into the answer column by a paste operation, the inference unit 11c infers whether there is a possibility that the user has performed an inappropriate behavior based on whether the character string (the character string input into the answer column by a paste operation) is a character string displayed in the window related to CBT. For example, if there is a character string input into the answer column by a paste operation among the character strings displayed in the window related to CBT (for example, the character strings in the question text), the inference unit 11c infers that there is no possibility that the user has performed an inappropriate behavior. In other words, if there is no character string input into the answer column by a paste operation among the character strings displayed in the window related to CBT (for example, the character strings in the question text), the inference unit 11c infers that there is a high possibility that the user has performed an inappropriate behavior.

[0106] As described above, whether the user has performed inappropriate behavior may be monitored based on whether the character string input into the answer column by the paste operation is the character string displayed in the window related to CBT.

[0107] Modification 2

[0108] Fig.111 is a block diagram showing the structure of the main parts of the server device of the second variant. Fig.11 In, with Figure 3 The same reference numerals are used for the same structures. The network system to which the server device 1A of the modified example is applied is also the same as that of the above example. Figure 1 The user terminal 2 of the second modification has the same structure as that of the above-mentioned example.

[0109] The control unit 11A of the server device 1A of the modification 2 is different from the above example in that it includes a transition history information acquisition unit 11d, which selects input operation information that changes the window related to the CBT from the active state to the inactive state, or input operation information that changes the window related to the CBT from the inactive state to the active state, from the operation history information acquired by the operation history acquisition unit 11b.

[0110] Furthermore, the estimation unit 11c estimates whether the possibility that the user has performed an inappropriate behavior is high based on the number of input operation information selected by the transition history information acquisition unit 11d (input operation information that changes the window related to the CBT from the active state to the inactive state, or input operation information that changes the window related to the CBT from the inactive state to the active state). Specifically, if the number of input operation information selected by the transition history information acquisition unit 11d exceeds the threshold, that is, if the estimation unit 11c estimates that the frequency of setting windows other than the window related to the CBT to the active state is high, it is estimated that the possibility that the user has performed an inappropriate behavior is high.

[0111] In the second variant, the user terminal 2 also performs Figure 8 Furthermore, the server device 1 executes Fig. 9 The processing shown, but the presumed processing of s5 is Fig.12 The treatment shown is not Fig.10 The processing shown, the aspects of which differ from the example. Fig.12 This is a flowchart showing the estimation processing of the server device according to Modification 2.

[0112] The transition history information acquisition unit 11d refers to the operation history information acquired in s4 and selects the input operation information that changes the window related to the CBT from the active state to the inactive state (s61). In the above example, s61 is set as the process of selecting the input operation information that changes the window related to the CBT from the active state to the inactive state, but it can also be set as the process of selecting the input operation information that changes the window related to the CBT from the inactive state to the active state.

[0113] The estimation unit 11c counts the number of input operation information (input operation information that changes the window related to the CBT from the inactive state to the active state) selected by the transition history information acquisition unit 11d in s61 (s62). The estimation unit 11c determines whether the number of input operation information counted in s62 is lower than a preset threshold (s63).

[0114] If it is determined in s63 that it is lower than the threshold, the estimation unit 11c estimates that there is no possibility that the user has performed inappropriate behavior (s64), and ends this process. On the other hand, if it is determined in s63 that it is not lower than the threshold, the estimation unit 11c estimates that there is a high possibility that the user has performed inappropriate behavior (s65), and ends this process.

[0115] As described above, a user who frequently sets a window other than the window related to CBT to an active state is a user who frequently refers to a window other than the window related to CBT. That is, a user who frequently sets a window other than the window related to CBT to an active state is a user who is likely to refer to a Web page retrieved through a Web browser or a stored file (a file for cheating), etc. The server device 1A of the second modification example estimates that a user who is likely to refer to a Web page retrieved through a Web browser or a stored file (a file for cheating), etc. is likely to have committed an improper act.

[0116] Therefore, the network system of the second modification can also monitor improper behavior using a web browser installed in the user terminal 2 or improper behavior using a file stored in the information processing device. Furthermore, CBT can be implemented in a form that allows the subject to use a terminal (user terminal 2) owned by the subject, and the cost of implementing CBT can be suppressed.

[0117] Modification 3

[0118] Moreover, you can also Fig.10 ,and Fig.12 The estimation processes shown are combined to estimate whether or not there is a high possibility that the user has performed an inappropriate behavior. Fig.13 FIG. 4 is a flowchart showing the estimation process of the modification example 3. Fig.13 In Fig.10 ,or Fig.12 In the estimation process of the third modification, if it is determined in s51 that there is no character string input by the paste operation or it is determined in s52 that the character string input by the paste operation is copied in the window related to the CBT, the processes of s61 to s65 are performed.

[0119] If it is determined in s52 that the character string input by the paste operation has not been copied in the window related to CBT, the processes of s61 to s63 are not performed, and it is estimated in s65 that there is a high possibility that the user has performed an improper behavior.

[0120] In addition, Fig.13 The example in the figure shows the case where the processes of s51 and s52 are performed before the processes of s61 to s63, but the processes of s61 to s63 are performed, and when it is determined that the value is below the threshold value in s63, the processes of s51 and s52 may be performed.

[0121] Modification 4

[0122] In the above example, it is assumed that server device 1 and server device 1A execute Fig.10 , Fig.12 ,or Fig.13 The estimation process shown in the figure may be performed by the user terminal 2. In this case, the estimation unit 11c or the transition history information acquisition unit 11d included in the server device 1 or the server device 1A may be provided in the control unit 21 of the user terminal 2. In this case, the operation history information generation unit 21b corresponds to the operation history information acquisition unit mentioned in the present invention.

[0123] In addition, the present invention is not directly limited to the embodiments, and the structural elements can be deformed and concretized in the implementation stage without departing from the main purpose. Moreover, various inventions can be formed by appropriate combinations of multiple structural elements disclosed in the embodiments. For example, some structural elements can be deleted from all the structural elements shown in the embodiments. Furthermore, the structural elements in different embodiments can be appropriately combined. Moreover, the order of each step in the flowchart shown in the description of all the examples is ultimately an example and can be appropriately changed within the possible range.

[0124] Furthermore, the correspondence between the structure of the present invention and the structure of the above-described embodiment can be described as in the following notes.

[0125] <Notes>

[0126] 1. A misconduct presumption device 1, comprising:

[0127] The answer acquisition unit 11a acquires the character string input into the answer column of the question as the answer;

[0128] The operation history information acquisition unit 11b acquires operation history information indicating the operation history of input operation of the answer to the answer column; and

[0129] The estimating unit 11c estimates whether an inappropriate behavior has been performed in answering the question based on the operation history information acquired by the operation history information acquiring unit 11b.

[0130] 2. The inappropriate behavior inference device 1 according to Note 1, wherein the inference unit 11c detects a character string input into the answer column by a paste operation based on the operation history information acquired by the operation history information acquisition unit 11b, and infers whether the input of the detected character string is through inappropriate behavior.

[0131] 3. The inappropriate behavior inference device 1 according to Supplementary Note 2, wherein if the character string input into the answer column by the paste operation is not the character string copied in the window presenting the question, the inference unit 11c infers that inappropriate behavior was performed in answering the question.

[0132] 4. The inappropriate behavior inference device 1 according to Note 2 or 3, wherein if the character string input into the answer column by the paste operation is not a character string existing in the window that prompts the question, the inference unit 11c infers that inappropriate behavior has been performed in answering the question.

[0133] 5. The misconduct inference device 1A according to any one of Notes 1 to 4, comprising a transition history information acquisition unit 11d, the transition history information acquisition unit 11d acquiring transition history information indicating a transition history of a window that prompts the question between an active state in which the window is focused and an inactive state in which the window is out of focus,

[0134] The estimating section 11 c further estimates whether or not an inappropriate behavior has been performed in answering the question using the transition history information acquired by the transition history information acquiring section 11 d .

[0135] 6. The inappropriate behavior inference device 1A according to Note 5, wherein the inference unit (11c) compares the frequency of the window prompting the question changing from an active state to an inactive state, or from an inactive state to an active state, with a preset threshold value to infer whether inappropriate behavior has been performed in answering the question.

[0136] 7. A misconduct presumption device 1A, comprising:

[0137] The answer acquisition unit 11a acquires the character string input into the answer column of the question as the answer;

[0138] a transition history information acquisition unit 11d for acquiring transition history information indicating a transition history of a window prompting the question between an active state in which the window is focused and an inactive state in which the window is out of focus; and

[0139] The estimating unit 11 c estimates whether or not an inappropriate behavior has been performed in answering the question based on the transition history information acquired by the transition history information acquiring unit.

Claims

1. A misconduct presumption device, comprising: An answer acquisition unit acquires a character string input into the answer column of the question as an answer; an operation history information acquisition unit that acquires operation history information indicating an operation history of an input operation of an answer to the answer column; and The estimating unit estimates whether an improper behavior has been performed in answering the question based on the operation history information acquired by the operation history information acquiring unit.

2. The inappropriate behavior inference device according to claim 1, wherein the inference unit detects a character string input into the answer column by a paste operation based on the operation history information acquired by the operation history information acquisition unit, and infers whether the input of the detected character string is through inappropriate behavior.

3. The inappropriate behavior inference device according to claim 2, wherein if the character string input into the answer column by the paste operation is not the character string copied in the window presenting the question, the inference unit infers that inappropriate behavior was performed in answering the question.

4. The inappropriate behavior inference device according to claim 2, wherein if the character string input to the answer column by the paste operation is not a character string existing in the window presenting the question, the inference unit infers that inappropriate behavior was performed in answering the question.

5. The misconduct presumption device according to any one of claims 1 to 4, comprising: a transition history information acquisition unit that acquires transition history information indicating a transition history of a window that prompts the question between an active state in which the window is focused and an inactive state in which the window is out of focus; The estimating section further estimates whether or not an improper behavior has been performed in answering the question using the transition history information acquired by the transition history information acquiring section.

6. The inappropriate behavior inference device according to claim 5, wherein the inference unit compares the frequency of the window prompting the question changing from an active state to an inactive state, or from an inactive state to an active state, with a preset threshold value to infer whether inappropriate behavior has been performed in answering the question.

7. A misconduct presumption device, comprising: An answer acquisition unit acquires a character string input into the answer column of the question as an answer; a transition history information acquisition unit that acquires transition history information, wherein the transition history information indicates a transition history of a window that prompts the question between an active state in which the window is focused and an inactive state in which the window is out of focus; and The estimating unit estimates whether or not an improper behavior has been performed in answering the question based on the transition history information acquired by the transition history information acquiring unit.

8. A method for presumption of misconduct, wherein a computer executes the following steps: The answer acquisition step acquires the character string input into the answer column of the question as the answer; an operation history information acquisition step of acquiring operation history information indicating an operation history of an input operation for an answer to the answer column; and The inference step is to infer whether or not an improper behavior has been performed in answering the question based on the operation history information acquired by the operation history information acquisition step.

9. A method for presumption of misconduct, wherein a computer executes the following steps: The answer acquisition step acquires the character string input into the answer column of the question as the answer; a transition history information acquisition step of acquiring transition history information, wherein the transition history information indicates a transition history of the window prompting the question between an active state in which the window is focused and an inactive state in which the window is out of focus; and An inference step of inferring whether or not an improper behavior has been performed in answering the question based on the transition history information acquired by the transition history information acquisition step.

10. A computer-readable storage medium having a misconduct presumption program stored thereon, which causes a computer to execute the following steps: The answer acquisition step acquires the character string input into the answer column of the question as the answer; an operation history information acquisition step of acquiring operation history information indicating an operation history of an input operation for an answer to the answer column; and The inference step is to infer whether or not an improper behavior has been performed in answering the question based on the operation history information acquired by the operation history information acquisition step.

11. A computer-readable storage medium having a misconduct presumption program stored thereon, which causes a computer to execute the following steps: The answer acquisition step acquires the character string input into the answer column of the question as the answer; a transition history information acquisition step of acquiring transition history information, wherein the transition history information indicates a transition history of the window prompting the question between an active state in which the window is focused and an inactive state in which the window is out of focus; and An inference step of inferring whether or not an improper behavior has been performed in answering the question based on the transition history information acquired by the transition history information acquisition step.

Images