JP2021089557A - Test system - Google Patents

Abstract

To provide a test system in which determining an integrity ratio between face data of a prior registered entrance exam candidate and face data photographed at a timing for taking a test can prevent fraud test or proxy test-taking.SOLUTION: There is provided a test system 10 which provides test contents to authorized users and makes it available for inputting answers against the test contents within a prescribed test time, wherein an integrity ratio between face data inputted by a camera of a communication terminal T110 and face image data stored in a user authentication database V150-7 is calculated, for executing fraud monitoring processing. The fraud monitoring processing is executed at monitoring timings in a plurality of times in accordance with a time elapsed within the test time, for determining whether or not the answers against the test contents are fraud based on the integrity ratio.SELECTED DRAWING: Figure 1

Description

Regarding the test system.

In qualification examinations, entrance examinations, university credit acquisition examinations, etc., identity verification is strictly required to prevent spoofing. Is visually confirmed by the examiner.

Japanese Patent Application Laid-Open No. 2003-150721

However, when a test tube uses a conventional technique to verify the identity of an examinee using a photo of the examinee's face, the larger the number of examinees, the longer it takes to verify the identity. There is a problem that the load becomes heavy. In addition, the examinee must always go to the designated examination venue to take the examination, and there is a problem that people in remote areas spend their efforts to go to the examination venue.

The present invention has been made in view of the above points, and an object of the present invention is a test system capable of conducting an online test, in which pre-registered test taker face data and images taken at the time of the test are taken. The purpose of the present invention is to provide a test system that can prevent fraudulent examinations and replacement ball examinations by determining the matching rate with the face data.

The present invention (1)
A test system that provides test content to authenticated users and enables input of answers to the test content within a predetermined test time.
Image data (for example, face data) input by an imaging means (for example, a camera of a communication terminal T110) and face image data (for example, a user authentication database) stored in advance corresponding to the authenticated user are stored. It is possible to execute a fraud monitoring process (for example, a process of comparative analysis by the fraud monitoring means V130-1) for calculating the matching rate with the (face image data).
It is possible to execute the fraud monitoring process at a plurality of monitoring timings according to the passage of time within the predetermined test time, and to determine whether or not the response to the test content is fraudulent based on the consistency rate. It is a test system characterized by.

The present invention (2)
The test system according to the invention (1), wherein it is possible to determine whether or not the response to the test content is incorrect by taking into consideration the information regarding the response to the test content in addition to the consistency rate.

The present invention (3)
The above-described invention (1), wherein the face image data is image data obtained by capturing a front face, and the matching rate is lowered even when the front face is not captured with respect to the image data input by the imaging means. Test system.

Since it is a system that enables online testing, it is possible to provide a testing system that enables people in remote areas to take the test without going to the testing site. In addition, by determining the matching rate between the face data of the examinee registered in advance and the face data taken during the test, it is possible to prevent fraudulent examinations and replacement ball examinations, which greatly reduces the burden on the examiner. It is possible to provide a test system that enables this.

It is an overall block diagram of the test system which concerns on this embodiment. It is a functional block diagram of the server which concerns on this embodiment. It is a system flow diagram of the process executed by the user information registration means which concerns on this embodiment. It is a system flow diagram of the process executed by the user login means which concerns on this embodiment. It is a system flow diagram of the process executed by the test execution means and fraud monitoring means which concerns on this Embodiment. It is a system flow diagram of the process executed by the test execution means and fraud monitoring means of the modified example which concerns on this embodiment. It is an overall block diagram of the learning system which concerns on this embodiment. It is a system flow diagram of the process executed by the search execution means which concerns on this embodiment. It is a figure which shows the example of the screen of the search result using the search keyword of the search execution means which concerns on this embodiment. It is a figure which shows the example of the screen of the detailed display request using the search keyword of the search execution means which concerns on this embodiment. It is a figure which shows the example of the screen of the full-text display executed by the full-text display means which concerns on this embodiment.

(Overall system configuration of test system 10)
First, the overall system configuration of the test system 10 according to the present embodiment will be described with reference to FIG. First, the test system 10 is a system that enables test management on the server V100 by operating the communication terminal T110. First, as shown in the figure, the test system 10 includes a communication terminal T110 and a server V100.

The communication terminal T110 includes a CPU (central processing device), ROM (read-only memory), RAM (random access memory), HDD (hard disk drive), I / F (communication interface device), operating means (keyboard, mouse, touch panel, etc.). (Operation device), personal computer equipped with display means (display device such as liquid crystal or touch panel), tablet computer, portable terminal device, and the like.

The server V100 includes a CPU (central processing device), a ROM (read-only memory), a RAM (random access memory), an HDD (hard disk drive), an I / F (communication interface device), and an operating means (keyboard, mouse, touch panel, etc.). Computers equipped with (operating devices), tablet computers, portable terminal devices, servers, and the like.

The server V100 is a device capable of various arithmetic processing and data processing, and communication processing with the communication terminal T110 via a network 820 such as the Internet.

The server V100 has a database for examination teaching materials, which is a database for specifying the teaching materials for the examination conducted by using the examination system 10 (see FIG. 1). In the test teaching material database, teaching material classification 1, teaching material classification 2, teaching material classification 3, teaching material name, and examination teaching material data, which are test teaching material identification information assigned to each test teaching material, are registered. The data registered in the test teaching material database is not limited to this information, and the test teaching material ID or the like associated with the test teaching material data may be registered.

Information (information on junior high school students, high school students, etc.) indicating the target of the test teaching material data used in the test is registered in the teaching material classification 1 of the test teaching material database. Information indicating test subjects such as social studies, science, national language, mathematics, and English is registered in the teaching material classification 2. In the teaching material classification 3, information in which the subjects of the examination are classified in detail is registered. For example, when the teaching material classification 2 is society, the information is history 1, history 2, history 3, geography 1, geography 2, and the like. As the teaching material name, the teaching material name of the examination teaching material data used in the examination is registered. Examination material data (examination question data) used in the examination is registered in the examination material data.

Further, the server V100 has a user authentication database V150-7, which is a database for uniquely identifying each user of the test system 10 (see FIG. 1). In the user authentication database V150-7, user ID, user password, user name, face image data, and voice data, which are user identification information assigned to each user, are registered. The data registered in the user authentication database V150-7 is not limited to these information, and fingerprint information, an email address, or the like may be registered as user identification information.

(System component / Server V100 / Block diagram)
Next, various functions of the server V100 according to the present embodiment will be described with reference to the block diagram of FIG. The server V100 registers the user ID, user password, user name, face image data, voice data, etc. required for user login when using the test system 10 and the education system 20 (details will be described later with reference to FIG. 7). User information registration means V110, test system 10, and education system 20, which will be described later, are means for logging in to the system by using a user ID or the like, and user login means V120, qualification test, which will be described later. It is a means to execute tests such as entrance examinations and university credit acquisition tests online, and is a means to perform a search using search keywords in the examination execution means V130 and the education system 20, which will be described later, and is a search described later. It has a data management means V150 that manages various data used in the execution means V140, the test system 10, and the education system 20.

Further, the test execution means V130 is a means for monitoring fraud in the test and has a fraud monitoring means V130-1 which will be described later.

Further, the data management means V150 is used in the format conversion means V150-1 and the education system 20 for converting the teaching material data (for example, teaching material video data) used in the education system 20 into the optimum video format when converting into text. Audio extraction means V150-2 that extracts audio from teaching material data (for example, teaching material video data), and text conversion means V150-3 that converts audio extracted by audio extraction means into text as data of a character string composed of character codes. , Data molding means V150-4 that molds the data of the character string converted into text by the text conversion means into a predetermined format, Indexing means V150-5 that converts the data molded by the data molding means into the format of the search database, Teaching material classification 1 ~ 3, Examination teaching material database V150-6 in which teaching material names, examination teaching material data, etc. are registered, User authentication database V150-6 in which user ID, user password, user name, face image data, voice data, etc. are registered 7. Teaching material classification 1-4, teaching material ID, teaching material name, teaching material data (for example, teaching material video data), teaching material length (for example, video length), etc. are registered in the learning teaching material database V150-8 (Fig. 7). It has a search database V150-9 (details will be described later using FIG. 7) in which teaching material IDs, conversion sentences, words in one sentence, playback start points, playback end points, etc. are registered. ing.

(System flow)
Here, with reference to FIG. 3, the processing executed by the user information registration means between the communication terminal T110 and the server V100 of the test system 10 and the search system 20 will be described in detail. This process is a system flow when the user registers the user's personal information using the communication terminal T110.

(System flow / User information registration means / Communication terminal T110 processing 1A)
First, the processing of the communication terminal T110 is executed. First, in step 102-S, the communication terminal T110 accesses the user information registration page based on the information (for example, URL) input by the user. Next, in step 103-S, the communication terminal T110 transmits a user information registration request for starting registration of user information to the server V100 side via the network 820.

(System flow / User information registration means / Server V100 process 1B)
Next, the process shifts to the processing of the server V100. First, in step 104-S, the user information registration means V110 receives the user information registration request transmitted from the communication terminal T110. Next, in step 106-S, the user information registration means V110 generates a user information registration request form (a form for inputting each item related to user information). Next, in step 108-S, the user information registration means V110 transmits a user information registration request form (a form for selecting each item related to user information) to the communication terminal T110 via the network 820.

(System flow / User information registration means / Communication terminal T110 processing 2A)
Next, the process shifts to the processing of the communication terminal T110. First, in step 110-S, the communication terminal T110 receives the user information registration request form transmitted from the server V100. Next, in step 112-S, the communication terminal T110 displays the user information registration request form on the display device of the communication terminal T110. Next, in step 114-S, the communication terminal T110 uses the personal information (name (user name of the user authentication database V150-7)) and face data (face image data of the user authentication database V150-7) input by the user. ), Voice data (information such as voice data name of user authentication database V150-7)) is set in the user information registration request form. Next, in step 116-S, the communication terminal T110 transmits the user information registration request form (completed) in which the personal information is input to the server V100 side via the network 820.

(System flow / User information registration means / Server V100 process 2B)
Next, the process shifts to the processing of the server V100. First, in step 118-S, the user information registration means V110 receives the user information registration request form (completed) transmitted from the communication terminal T110. Next, in step 120-S, the user information registration means V110 generates a user ID and a user password. Next, in step 122-S, the user information registration means V110 inputs the generated user ID and user password, and personal information (name, face data, voice data, etc.) entered in the user information registration request form. It is linked and stored in the user authentication database V150-7. Next, in step 124-S, the user information registration means V110 transmits the generated user ID and user password to the communication terminal T110 via the network 820.

Here, in the present embodiment, the user authentication database V150-7 is configured so that the user password and the user's personal information are managed correspondingly for each user ID. However, the user authentication database V150- The types of items 7 are not limited to the items shown in FIG. 1, and for example, the user's address, telephone number, date of birth, email address, school name, company name, department to which the company belongs, etc. should be managed. It may be configured.

(System flow / User information registration means / Communication terminal T110 processing 3A)
Next, the process shifts to the processing of the communication terminal T110. First, in step 126-S, the communication terminal T110 receives the user ID and user password transmitted from the server V100 via the network 820. Next, in step 128-S, the communication terminal T110 displays the received user ID and user password on the display device, and ends the process related to the user information registration means.

Next, with reference to FIG. 4, the processing executed by the user login means between the communication terminal T110 and the server V100 of the test system 10 and the education system 20 will be described in detail. This process is a system flow when the user logs in to the test system 10 and the education system 20 using the communication terminal T110.

(System flow / User login means / Communication terminal T110 processing 1C)
First, the processing of the communication terminal T110 is executed. First, in step 202-S, the communication terminal T110 accesses the user login page based on the information (for example, URL) input by the user. Next, in step 204-S, the communication terminal T110 transmits the login information input form request to the server V100 side via the network 820.

(System flow / User login means / Server V100 processing 1D)
Next, the process of the server V100 is executed. First, in step 206-S, the user login means V120 receives the login information input form request transmitted from the communication terminal T110. Next, in step 208-S, the user login means V120 generates a login information input form (a form for inputting each item related to user login). Next, in step 210-S, the user login means V120 transmits a login information input form (a form for inputting each item related to user login) to the communication terminal T110 via the network 820.

(System flow / User login means / Communication terminal T110 processing 2C)
Next, the processing of the communication terminal T110 is executed. First, in step 212-S, the communication terminal T110 receives the login information input form (form for inputting each item related to user login) transmitted from the server V100. Next, in step 214-S, the communication terminal T110 displays a login information input form (a form for inputting each item related to user login) on the display device. Next, in step 216-S, the communication terminal T110 sets the login information (user ID and user password) input by the user. Next, in step 218-S, the communication terminal T110 transmits the set login information (user ID and user password) to the server V100 via the network 820.

(System flow / User login means / Server V100 processing 2D)
Next, the process of the server V100 is executed. First, in step 220-S, the user login means V120 receives the user ID and user password transmitted from the communication terminal T110. Next, in step 222-S, the user login means V120 determines whether or not the received user ID and user password match the user ID and user password stored in the user authentication database V150-7. (Authenticate. Next, in step 224-S, the user login means V120 generates a main screen for the user. Next, in step 226-S, the user login means V120 transmits the generated main screen to the communication terminal T110 via the network 820.

(System flow / User login means / Communication terminal T110 processing 3C)
Next, the processing of the communication terminal T110 is executed. First, in step 228-S, the communication terminal T110 receives the main screen of the user transmitted from the server V100. Next, in step 230-S, the communication terminal T110 displays the received main screen on the display device. The main screen is a screen on which a test execution button or the like for starting the test content (test) described later is displayed.

By the process executed by the user login means, when accessing the server V100, the user ID and the user password entered by the communication terminal T110 are combined with the user ID and the user password managed by the server V100. Depending on whether or not they match, it is possible to determine whether or not the user of the communication terminal T110 is a legitimate user. Although not shown, when the authentication result is normal (regular user ID and user password), the functions of the test system 10 and the education system 20 can be used (the main screen of the user is the display device of the communication terminal T110). Is displayed in). On the other hand, if the authentication result is abnormal (illegal user ID or user password), the functions of the test system 10 and the education system 20 cannot be used (the main screen of the user is not displayed on the display device of the communication terminal T110). ..

Next, with reference to FIG. 5, the processing executed by the test execution means and the fraud monitoring means between the communication terminal T110 and the server V100 of the test system 10 will be described in detail. This process is a system flow when the examinee who is a user accesses the server V100 using the communication terminal T110 and logs in by the user login means shown in FIG.

(System flow / Test execution means / Communication terminal T110 processing 1E)
First, the processing of the communication terminal T110 is executed. First, in step 302-S, the communication terminal T110 determines that the test execution button has been selected because the test execution button for starting the test content (test) due to the examinee operating the communication terminal T110 is selected. To do. Next, in step 304-S, the communication terminal T110 acquires face data using the camera provided in the communication terminal T110 based on the selection of the test execution button. Next, in step 306-S, the communication terminal T110 transmits the face data acquired by the camera and the test question / answer form request to the server V100 side via the network 820. Here, the examination question / answer form is an examination question form used in the examination and an answer form for entering and selecting the answer of the examination question.

(System flow / test execution means, fraud monitoring means / server V100 processing 1F)
Next, the process of the server V100 is executed. First, in step 308-S, the test execution means V130 receives the face data and the test question / answer form request transmitted from the communication terminal 110. Next, in step 310-S, the fraud monitoring means V130-1 of the test execution means V130 compares the received face data with the face image data stored in the user authentication database V150-7 as fraud monitoring processing. To analyze. Next, in step 312-S, the fraud monitoring means V130-1 of the test execution means V130 combines the received face data and the face image data stored in the user authentication database V150-7 as a result of the comparative analysis. If the result of the comparative analysis is OK (no fraud), an exam question / answer form is generated. On the other hand, the fraud monitoring means V130-1 of the test execution means V130 cancels the test if the result of the comparative analysis is NG (with fraud). Next, in step 314-S, the test execution means V130 transmits the generated test question / answer form to the communication terminal T110 via the network 820.

Here, in the comparative analysis of the test execution means V130 by the fraud monitoring means V130-1, the received face data and the face image data stored in the user authentication database V150-7 are used for face recognition of artificial intelligence. For comparison, specifically, the positions of facial parts (eyes, nose, mouth, etc.) are determined, and the matching rate (also called recognition rate or similarity) of facial parts and contours is calculated, and the matching rate is determined. If the value is equal to or greater than the predetermined value (for example, 90% or more), the result of the comparative analysis is judged to be OK (no fraud), and if the matching rate is less than the predetermined value (for example, less than 90%), the result of the comparative analysis is determined to be NG (with fraud). to decide. In addition, it is conceivable that the examinee prepares his / her own face photograph and arranges it toward the camera, and performs an illegal examination such as cheating while maintaining the matching rate. In such a case, even if the matching rate is equal to or higher than a predetermined value (for example, 90% or higher), the face data received last time and the face data received this time are the same (for example, the face data received last time and this time). The similarity of the received face data may be determined (when the similarity is 98% or more), and the result of the comparative analysis may be determined to be NG (with fraud).

(System flow / Test execution means / Communication terminal T110 processing 2E)
Next, the processing of the communication terminal T110 is executed. First, in step 316-S, the communication terminal T110 receives the test question / answer form transmitted from the server V100. Next, in step 318-S, the communication terminal T110 displays the received test question / answer form on the display device. Next, in step 320-S, when the test start button of the communication terminal T110 is selected by the examinee, the communication terminal T110 determines that the test start button has been selected. Next, in step 322-S, the communication terminal T110 transmits the test start information to the server V100 via the network 820 based on the selection of the test start button.

(System flow / test execution means, fraud monitoring means / server V100 processing 2F)
Next, the process of the server V100 is executed. First, in step 324-S, the test execution means V130 receives the test start information transmitted in step 322-S. Next, in step 326-S, the test execution means V130 transmits a face data request to the communication terminal T110 via the network 820 according to the lapse of a predetermined time. Here, an example in which 5 minutes have passed from the start of the test will be given as the predetermined time elapsed. Further, the timing of transmitting the face data request, which is the monitoring timing during the test, is configured to be performed a plurality of times within the test time (for example, 90 minutes), for example, when 5 minutes have passed from the start of the test. The first face data request is transmitted, and then the second and third face data requests are transmitted every 5 minutes. You may also send face data requests at random times during the test. Further, the face data request may be sent when the examinee's face is in front of the camera. In this case, the camera can shoot the front face, which is the optimum face data for determining the matching rate. Can be. It should be noted that although the face data request is configured to be transmitted every 5 minutes from the start of the test, the front face may not be photographed at that time, and in such a case, the matching rate is less than the predetermined value (for example, It is less than 90%), and there is a problem that it is judged to be fraudulent even though it has not been cheated. For such problems, a grace period of 1 hour (for example, 1 minute) is provided after 5 minutes have passed, and the face is set every second hour (for example, every 1 second) shorter than the 1st hour in this grace time. It may be configured to request data and acquire face data continuously. Then, in the comparative analysis of the test execution means V130 by the fraud monitoring means V130-1, if the face data of the front face exists in the face data acquired in this grace time, the face data of the front image and the user authentication database V150 It is preferable to configure the face image data stored in -7 to be compared. By increasing the period for acquiring the face data of the front face, which is optimal for determining the consistency rate, it is possible to solve the problem that it is determined that there is fraud even though no fraudulent activity has been performed. Become.

Further, the image of the camera may be monitored in real time, and the face data request may be transmitted at the timing when the camera confirms that the examinee's face has moved. Such a configuration is suitable when there is an examinee next to the examinee, and it is possible to take a picture of the timing when the examinee cheats the answer of the next person with a camera. Furthermore, when monitoring the image of the camera in real time, the artificial intelligence of facial expression recognition that reads the facial expression of the examinee can estimate the emotion of the examinee based on the image of the facial expression. With this artificial intelligence of facial expression recognition, it is possible to measure the emotion and concentration at that time from the facial expression of the examinee taken with the camera, and to know the timing when the concentration increases or conversely the timing when the concentration is interrupted. Because it is possible, it can be used as data to judge the fraud of the examinee. Furthermore, by analyzing not only the facial expressions of the examinees but also their movements, it is possible to know the degree of concentration on the examination as a whole, which can be used as data for judging the injustice of the examinees. Further, when monitoring the image of the camera in real time, the face data may be acquired by using the camera when the examinee's face is in front of the camera. In this case, the face data is in front. Since the data is for the face, it can be the most suitable face data for determining the consistency rate.

(System flow / Test execution means / Communication terminal T110 processing 2E)
Next, the processing of the communication terminal T110 is executed. First, in step 328-S, the communication terminal T110 receives the face data request transmitted in step 326-S. Next, in step 330-S, the communication terminal T110 photographs the examinee's face with a camera based on the face data request, and acquires the face data. Next, in step 332-S, the communication terminal T110 transmits the acquired face data to the server V100 via the network 820.

(System flow / test execution means, fraud monitoring means / server V100 processing 2F)
Next, the process of the server V100 is executed. First, in step 334-S, the test execution means V130 receives the face data transmitted in step 332-S. Next, in step 336-S, the fraud monitoring means V130-1 of the test execution means V130 compares the received face data with the face image data stored in the user authentication database V150-7 as the fraud monitoring process. To analyze. Next, in step 338-S, the fraud monitoring means V130-1 of the test execution means V130 stores the matching rate information in which the result of the comparative analysis (matching rate) and the reception time of the face data are associated with each other. The matching rate information is, for example, when the test is started at 9:00, the matching rate of the face data and the face image data received at 9:05, 9:10, 9:15, and so on. The data is stored in association with the reception time (shooting time) of the face data. As the matching rate information, three pieces of information such as the reception time of the face data, the face data, and the matching rate may be stored in association with each other.

(System flow / Test execution means / Communication terminal T110 processing 2E)
After the first face data is transmitted to the server V100 in step 332-S, and in step 340-S during the test, the communication terminal T110 is a test answer form in which the examinee answers the test during the test. Based on what you entered in, set the answer in the exam answer form. Next, after setting the test answer in the test answer form, the communication terminal T110 displays the test answer form (completed) on the network 820 based on the test taker selecting the test end button in step 342-S. Is transmitted to the server V100 via.

(System flow / Test execution means / Server V100 processing 3F)
Next, the process of the server V100 is executed. First, in step 344-S, the test execution means V130 receives the test response form (completed) transmitted from the communication terminal T110. Next, in step 346-S, the test execution means V130 stores the test response information in which the information on the test response form (input) and the consistency rate information are associated with each other.

The timing of transmitting the face data request, which is the monitoring timing during the test, is performed every 5 minutes within the test time (for example, 90 minutes), but the fraud monitoring means V130-1 in step 336-S is fraudulent. As a result of the comparative analysis of the monitoring process, when the matching rate is lower than the previous matching rate or when the matching rate is NG, the timing of sending the face data request is shortened to every 3 minutes or the like. You may. By doing so, it becomes possible to strengthen fraud monitoring during the test.

Further, when the matching rate decreases within the test time, the test execution means V130 may be configured to increase the number of facial parts used in the comparative analysis of fraud monitoring processing. For example, within the test time, the matching rate is calculated using the two parts of the face, the eyes and the mouth, and if the matching rate drops, the nose is added and the three parts of the eyes, mouth, and nose are used. Try to calculate the consistency rate. By doing so, it is possible to improve the accuracy of the matching rate and strengthen the fraud monitoring during the test.

Next, with reference to FIG. 6, the processing executed by the test execution means and the fraud monitoring means between the communication terminal T110 and the server V100, which is a modification of the test system 10, will be described in detail. In the process executed by the test executing means and the fraud monitoring means of FIG. 6, in step 326-S of the process executed by the test executing means and the fraud monitoring means of FIG. 5, the communication terminal T110 is based on the face data request of the server V100. Unlike acquiring face data with a camera, in step 340-S, the communication terminal T110 is configured to acquire face data with a camera based on the examinee inputting the test answer into the test answer form. Has been done. This process is a system flow when the examinee who is a user accesses the server V100 using the communication terminal T110 and logs in by the user login means shown in FIG.

(System flow / Test execution means / Communication terminal T110 processing 1E)
First, the processing of the communication terminal T110 is executed. First, in step 302-S, the communication terminal T110 determines that the test execution button has been selected because the test execution button for starting the test content (test) due to the examinee operating the communication terminal T110 is selected. To do. Next, in step 304-S, the communication terminal T110 acquires face data using the camera provided in the communication terminal T110 based on the selection of the test execution button. Next, in step 306-S, the communication terminal T110 transmits the face data acquired by the camera and the test question / answer form request to the server V100 side via the network 820. Here, the examination question / answer form is an examination question form used in the examination and an answer form for entering and selecting the answer of the examination question.

(System flow / test execution means, fraud monitoring means / server V100 processing 1F)
Next, the process of the server V100 is executed. First, in step 308-S, the test execution means V130 receives the face data and the test question / answer form request transmitted from the communication terminal 110. Next, in step 310-S, the fraud monitoring means V130-1 of the test execution means V130 compares the received face data with the face image data stored in the user authentication database V150-7 as fraud monitoring processing. To analyze. Next, in step 312-S, the fraud monitoring means V130-1 of the test execution means V130 combines the received face data and the face image data stored in the user authentication database V150-7 as a result of the comparative analysis. If the result of the comparative analysis is OK (no fraud), an exam question / answer form is generated. On the other hand, the fraud monitoring means V130-1 of the test execution means V130 cancels the test if the result of the comparative analysis is NG (with fraud). Next, in step 314-S, the test execution means V130 transmits the generated test question / answer form to the communication terminal T110 via the network 820.

(System flow / Test execution means / Communication terminal T110 processing 2E)
Next, the processing of the communication terminal T110 is executed. First, in step 316-S, the communication terminal T110 receives the test question / answer form transmitted from the server V100. Next, in step 318-S, the communication terminal T110 displays the received test question / answer form on the display device. Next, in step 320-S, when the test start button of the communication terminal T110 is selected by the examinee, the communication terminal T110 determines that the test start button has been selected. Next, in step 322-S, the communication terminal T110 transmits the test start information to the server V100 via the network 820 based on the selection of the test start button. Next, in step 340-S, the communication terminal T110 sets the answer in the test answer form based on the test taker inputting the test answer in the test answer form during the test. Next, in step 330-S, the communication terminal T110 photographs the examinee's face with a camera based on the test answer entered in the test answer form, and acquires face data. Next, in step 332-S, the communication terminal T110 transmits the acquired face data to the server V100 via the network 820. Next, after setting the test answer in the test answer form, the communication terminal T110 displays the test answer form (completed) on the network 820 based on the test taker selecting the test end button in step 342-S. Is transmitted to the server V100 via.

In step 330-S, the communication terminal T110 takes a picture of the examinee's face with a camera based on the test answer entered in the test answer form, and obtains the face data to obtain the test answer. When inputting, the input is performed in front of the display device of the communication terminal T110, and the examinee's face is in front of the camera. Therefore, the optimum front face data for determining the matching rate can be obtained. ..

(System flow / test execution means, fraud monitoring means / server V100 processing 2F)
Next, the processing of the server V100 will be described. In step 324-S, the test execution means V130 receives the test start information transmitted in step 322-S. Next, in step 334-S, the test execution means V130 receives the face data transmitted in step 332-S. Next, in step 336-S, the fraud monitoring means V130-1 of the test execution means V130 compares the received face data with the face image data stored in the user authentication database V150-7 as the fraud monitoring process. To analyze. Next, in step 338-S, the fraud monitoring means V130-1 of the test execution means V130 obtains the matching rate information in which the result of the comparative analysis (matching rate) of the fraud monitoring process and the reception time of the face data are associated with each other. Remember. As for the consistency rate information, for example, when the test is started at 9:00, the answer to question 1 is input at 9:08, the answer to question 2 is input at 9:12, and the answer to question 3 is input. The data of the matching rate between the face data and the face image data received at 9:19 ... Is stored in association with the reception time (shooting time) of the face data. As the matching rate information, the reception time of the face data, the face data, and the matching rate may be stored in association with each other.

(System flow / Test execution means / Server V100 processing 3F)
Next, the process of the server V100 is executed. First, in step 344-S, the test execution means V130 receives the test response form (completed) transmitted from the communication terminal T110. Next, in step 346-S, the test execution means V130 stores the test response information in which the information on the test response form (input) and the consistency rate information are associated with each other.

In the test system 10 according to the present embodiment, the fraud monitoring means V130-1 of the test execution means V130 has face data input by the camera of the communication terminal T110 and a face image stored in advance in the user authentication database V150-7. Although the fraud monitoring process for calculating the consistency rate with the data is executed, when performing an English speaking test or the like, the fraud monitoring means V130-1 of the test execution means V130 is the communication terminal T110. It may be configured to calculate the matching rate between the voice data input by the microphone and the voice data stored in advance in the user authentication database V150-7.

When calculating the matching rate between voice data and voice data in an English speaking test or the like, steps 304-S, 306-S, 308-S, 310-S, and 312 of FIGS. 5 and 6 -S, Step 326-S (only in the case of FIG. 5), Step 328-S (only in the case of FIG. 5), Step 330-S, Step 332-S, Step 334-S, Step 336-S, Step 338- The processing of S is different from the case of calculating the consistency between the face data and the face image data.

Here, a process different from the process for calculating the consistency between the face data and the face image data will be described. Based on the fact that the test execution button is selected in step 304-S, the communication terminal T110 acquires face data using a camera and voice data using a microphone. Next, in step 306-S, the communication terminal T110 transmits the face data acquired by the camera, the voice data acquired by the microphone, and the test question / answer form request to the server V100 side via the network 820.

In step 308-S, the test execution means V130 receives the face data, voice data, and test question / answer form request transmitted from the communication terminal 110. Next, in step 310-S, the fraud monitoring means V130-1 of the test execution means V130 compares the received face data with the face image data stored in the user authentication database V150-7 as fraud monitoring processing. Along with the analysis, the received voice data and the voice data stored in the user authentication database V150-7 are compared and analyzed. Next, in step 312-S, the fraud monitoring means V130-1 of the test execution means V130 is OK with the result of comparative analysis between the received face data and the face image data stored in the user authentication database V150-7. If the result of comparison analysis between the received voice data and the voice data stored in the user authentication database V150-7 is OK (no fraud), an exam question / answer form is generated. To do. On the other hand, the fraud monitoring means V130-1 of the test execution means V130 cancels the test if the result of the comparative analysis is NG (with fraud).

Here, in the comparative analysis of the fraud monitoring process of the fraud monitoring means V130-1, the received face data, voice data and face image data stored in the user authentication database V150-7 are performed by using the face authentication of artificial intelligence. , Voice data is compared, and if the matching rate is equal to or more than the predetermined value (for example, 90% or more), the result of the comparison analysis is judged to be OK (no fraud), and the matching rate is less than the predetermined value (for example, less than 90%). The result of the analysis is judged to be NG (injustice).

In step 326-S, the face data request and the voice data request are transmitted according to the lapse of a predetermined time. Here, as the predetermined time lapse, an example in which 5 minutes have passed from the start of the test will be given. Further, the timing of transmitting the face data request and the voice data request, which is the monitoring timing during the test, is configured to be performed a plurality of times within the test time (for example, 90 minutes), for example, 5 minutes from the start of the test. It is configured to transmit the first face data request and voice data request when elapsed, and then transmit the second and third face data request and voice data request every 5 minutes. In addition, the face data request and the voice data request may be transmitted at random times during the test.

In step 328-S, the communication terminal T110 receives the face data request and the voice data request transmitted in step 326-S. Next, in step 330-S, the communication terminal T110 photographs the examinee's face with the camera based on the face data request and the voice data request, acquires the face data, collects the examinee's voice with the microphone, and voices. Get the data. Next, the communication terminal T110 transmits the acquired face data and voice data to the server V100 via the network 820 in step 332-S.

In step 334-S, the test execution means V130 receives the face data and voice data transmitted in step 332-S. Next, in step 336-S, the fraud monitoring means V130-1 of the test execution means V130 performs the received face data and voice data and the face image data stored in the user authentication database V150-7 as fraud monitoring processing. And the voice data are compared and analyzed. Next, in step 338-S, the fraud monitoring means V130-1 of the test execution means V130 matches the result (matching rate) of the comparative analysis of the fraud monitoring process with the reception time of the face data and the voice data. Store rate information. For example, when the test is started at 9:00, the matching rate information includes face data, voice data, face image data, and voice received at 9:05, 9:10, 9:15, and so on. The data of the consistency rate with the data is stored in association with the reception time (shooting time) of the face data and the voice data. As the matching rate information, the reception time of the face data and the voice data, the face data and the voice data, and the matching rate may be stored in association with each other.

In the English speaking test, etc., the face data and voice data transmitted from the communication terminal 110 are compared with the face image data and voice data stored in the user authentication database V150-7 as fraud monitoring processing. Although an example of analysis is shown, the data transmitted from the communication terminal 110 may be limited to voice data, and the data stored in the user authentication database V150-7 may be compared and analyzed.

When calculating the matching rate between voice data and voice data in an English speaking test, etc., steps 340-S, 330-S, and 332-S in FIG. 6 should be processed as follows. Is preferable. Step 340-S is a process in which the communication terminal T110 acquires voice data based on the speaking answer being input to the microphone. Step 330-S is a process in which the communication terminal T110 photographs the examinee's face with a camera and acquires face data based on the speaking answer being input to the microphone. Step 332-S is a process in which the communication terminal T110 transmits the acquired face data and voice data to the server V100 via the network 820. By acquiring face data based on the input of voice data in the speaking test in this way, it becomes possible to calculate the matching rate using the above-mentioned artificial intelligence of facial expression recognition including facial expressions during speaking. , It is possible to prevent fraudulent examinations and replacement ball examinations.

As the camera, an example of using a camera integrally provided in the communication terminal T110 or a camera provided separately from the communication terminal T110 has been shown, but the position of a person, the presence of a person, a face and a facial expression can be recognized. A camera capable of confirming the depth (for example, a scale for measuring from which position to which position the focus is in focus) may be used. With this configuration, even if the examinee prepares his / her own face photo and arranges it toward the camera to perform fraudulent examinations such as cheating while maintaining the consistency rate, the examinee can take the examination. Since it can be determined that the person is not in front of the camera and it is determined to be fraudulent, it is possible to further prevent fraudulent examinations and replacement ball examinations. Further, a plurality of cameras and microphones may be provided. In addition, a plurality of cameras may be provided, and by doing so, it is possible to recognize a person other than the examinee who is next to the examinee, so that it is possible to further prevent fraudulent examinations and replacement ball examinations. Furthermore, a plurality of microphones may be provided, and by doing so, the place where the voice (voice data) is emitted can be specified. Therefore, by using the camera at the same time as the camera provided with the depth sensor, You will be able to identify the position of a person more. Also, using a camera (or a normal camera) that can confirm the position of the person, the presence of the person, the depth at which the face and facial expression can be recognized (for example, a measure of how far the focus is from), the test time It may be configured to determine whether or not the examinee is in the seat within the period (during the examination time). With this configuration, if the presence of the examinee cannot be confirmed, it can be determined that some kind of fraud has been committed.

The test system (1) according to the present embodiment is a test system that provides test content to an authenticated user and enables input of an answer to the test content within a predetermined test time, and is an imaging means (imaging means (1). For example, the image data (for example, face data) input by the communication terminal T110) and the face image data (for example, the user authentication database V150-7) stored in advance corresponding to the authenticated user are stored. It is possible to execute the fraud monitoring means that calculates the matching rate with the existing face image data), and the fraud monitoring means is executed at multiple monitoring timings according to the passage of time within the predetermined test time to obtain the matching rate. It is a test system characterized in that it is possible to determine whether or not the answer to the test content is incorrect based on the test content.

In step 346-S, the test system (2) according to the present embodiment stores the test response information in which the information in the test response form (input) and the consistency rate information are associated with each other (for example, on the day of the test or on the day of the test). At a later date), the test system according to the present embodiment can determine whether or not the answer to the test is incorrect by taking into account the information about the answer to the test (for example, scoring information) in addition to the stored consistency rate. (1) The test system described. By constructing in this way, not only the memorized consistency rate but also the information about the test answer is added, and if the answer is similar to that of the next person, the cheat of the next person's answer can be cheated. It can be used as a basis for judgment.

Here, as the scoring of the test, a process using an automatic scoring means may be adopted as an automatic scoring process in which the artificial intelligence automatically scores. In this case, the automatic scoring means divides the test answers into several patterns and scores them. When evaluating in stages rather than simple correct or incorrect answers, the answers in the same category are given uniform partial points by categorizing them according to the answer pattern. In addition, automatic scoring processing may be adopted in the scoring process when the answer is described in a format different from the selection format such as the mark sheet format (a format in which the examinee freely fills in the answer, etc.). In this case, after storing the test response information in which the information on the test response form (entered) and the consistency rate information are associated with each other (for example, on the day of the test or at a later date), the examinee in the automatic scoring process uses the stored consistency rate. The consistency rate of the results of the comparative analysis by the fraud monitoring means V130-1 is less than a predetermined value (for example, less than 90%), taking into account the similarity between the answers of the above and the answers of other examinees (for example, the next examinee). However, when the similarity of the answers is equal to or more than a predetermined value (for example, 80% or more), it may be configured so that it can be determined whether or not the answer to the test has a fraudulent canning. With such a configuration, it is possible to determine an illegal examination such as cheating in consideration of the scoring information of the automatic scoring process. Further, the scoring result may be used to provide the examinee with a service called recommendation used in the education system 20. For example, if the examinee makes a mistake in the problem of the Mesopotamian civilization, the information of the education ID VD_CSR1 or VD_KSBK1 may be provided to the examinee.

In the test system (3) according to the present embodiment, the face image data stored in the user authentication database V150-7 is image data obtained by capturing the front face of the examinee, and the photographing means (for example, the communication terminal T110). This is the test system according to the test system (1) according to the present embodiment, which determines that the matching rate is lowered when the face data captured by the camera) is not the one in which the front face is imaged. The image data of the front face and the face data can be the optimum data for determining the consistency rate.

(Overall system configuration of education system 20)
The overall system configuration of the education system 20 according to the present embodiment will be described with reference to FIG. 7. First, the education system 20 is a system that enables the management of tests on the server V100 by operating the communication terminal T110. First, as shown in the figure, the education system 20 includes a communication terminal T110 and a server V100.

The configuration of the communication terminal T110 and the server V100 is the same as that of the test system 10, but in the server V100 of the education system 20, the teaching material classifications 1 to 4, the teaching material ID, the teaching material name, the teaching material data (for example, the teaching material video data), and the teaching material It has a learning material database V150-8 in which lengths (for example, video lengths) and the like are registered (see FIG. 7). The data registered in the learning material database V150-8 is not limited to this information, and other information may be registered, and the teaching material data is via the network 820 in addition to the pre-registered data. It may be configured so that the downloaded teaching material data can be registered. Here, the teaching material data is preferably video teaching material video data for learning used in e-learning or the like (including PowerPoint data and PDF data in animation format), and the video includes video, board writing, and the like. Audio etc. are included. Note that data containing only audio may be the target of the teaching material data. Further, still image data including audio such as PDF data including audio may be used as the target of the teaching material data.

Information (information on junior high school students, high school students, etc.) indicating the target of the learner using the teaching material data (for example, teaching material video data) used in the education system 20 is registered in the teaching material classification 1 of the learning material database V150-8. ing. The teaching material classification 1 may be used as the subject information, and in that case, the subject is registered in the learning material database V150-8 as a subject for junior high school students and a subject for high school students, which is a subject different from the subject for junior high school students. In the teaching material classification 2, major classification subjects are registered as information indicating learning subjects such as society, geography and history, science, chemistry, physics, biology, earth science, national language, mathematics, and English. It should be noted that, depending on the major classification subject, the teaching material data (for example, teaching material video data) indicates the teaching material of a different subject in the teaching material classification 2. In the teaching material classification 3, the middle classification subject is registered as the information which classified the learning subject in detail. For example, when the teaching material classification 2 is society, the information is history 1, history 2, history 3, public citizen 1, public citizen 2, world history A, world history B, and the like. It should be noted that, depending on the middle classification subject, the teaching material data (for example, teaching material video data) indicates the teaching material of a different subject in the teaching material classification 3.

In the teaching material classification 4, sub-classification subjects are registered as information in which the learning subjects are further classified. It should be noted that, depending on the sub-category subject, the teaching material data (for example, teaching material video data) indicates the teaching material of a different subject in the teaching material classification 4. For example, when the teaching material classification 2 is geographic history and the teaching material classification 3 is world history B, information such as ancient Orient 1, ancient Orient 2, and ancient Orient 3 is registered. The teaching material ID is a number associated with teaching material data (for example, teaching material video data) which is a video used in learning. As the teaching material name, the teaching material name of the teaching material data (for example, teaching material video data) used in learning is registered. The teaching material video data as the teaching material data is a video used in learning, and the teaching material video data includes the teaching material name of the video of the teaching material video data (same as the teaching material name of the learning teaching material database), the explanation of the teaching material video data, and so on. The video length of the teaching material video data (same as the video length of the learning material database) is linked and stored. The teaching material length (for example, the video length) indicates the playback time of the teaching material data and the video time of the teaching material video data.

Further, the server V100 has a search database V150-9 which is used in the search execution means V140 of the education system 20 and in which the teaching material ID, the converted sentence, the word in the sentence, the reproduction start point, the reproduction end point, etc. are registered. (See Fig. 7). The data registered in the search database V150-9 is not limited to this information, and the teaching material categories 1 to 4, the teaching material names, and the teaching material data (for example, teaching materials) registered in the learning teaching material database V150-8 are not limited to these information. The video data), the length of the teaching material (for example, the length of the video) and the like may be registered so as to be associated with the teaching material ID of the same inspection database as the teaching material ID of the learning teaching material database V150-8.

The data registered in the search database V150-9 is registered as follows. The data registration process described below is preferably executed using artificial intelligence. First, the format conversion means V150-1 converts the teaching material data (for example, teaching material moving image data) used in the education system 20 into an optimal moving image format when converting into text. The format conversion means V150-1 converts a moving image that is not the A format, which is the optimum moving image format (for example, in the case of the moving image format of the B format), into the moving image format of the A format. Next, the audio extraction means V150-2 extracts audio from the teaching material data (for example, teaching material moving image data) converted into the optimum moving image format.

Next, the text conversion means V150-3 converts the voice extracted by the voice extraction means into text as data of a character string composed of a character code. Here, the teaching material ID stored in the learning teaching material database V150-8 of FIG. 7 is the teaching material data of VD_CSR1 (for example, teaching material video data) for junior high school students (teaching material classification 1), society (teaching material classification 2), and history 1. An example of the means for converting (teaching material classification 3) into text. In the means for converting the teaching material video data of the teaching material ID of VD_CSR1 into text, the words in one sentence separated by the breaks in the voice instead of the grammatical breaks and the words in the sentence as time stamp information are reproduced from the voice extracted by the voice extraction means. Extract the start point and playback end point. Then, a converted sentence composed of a plurality of words in one sentence is converted into text and processed. The converted sentence is a concatenation of all the words in the sentence, and the time stamp information is arranged in chronological order.

For example, the word in one sentence, the playback start point, and the playback end point extracted from the teaching material data (for example, teaching material video data) whose teaching material ID is VD_CSR1 are linked as follows.
"Word in one sentence =" first ", playback start point = 24.22, playback end point = 25.31"
"Word in one sentence =" Mesopotamian civilization ", playback start point = 25.83, playback end point = 27.79"
"Word in one sentence =" It is one of the four major civilizations "Playback start point = 28.42, playback end point = 29.96"
In addition, in the text conversion means, all the words in one sentence are concatenated to generate one conversion sentence, "One conversion sentence =" First, the Mesopotamian civilization is one of the four major civilizations. "" In addition, 24.22, 25.31 and the like indicate 24.22 seconds and 25.31 seconds, and the reproduction start point and the reproduction end point can be specified up to 1/100 second.

Next, the data molding means V150-4 molds the data of the character string converted into text by the text conversion means into a predetermined format. Here, the teaching material ID stored in the learning teaching material database V150-8 of FIG. 7 is the teaching material data of VD_CSR1 (for example, teaching material video data) for junior high school students (teaching material classification 1), society (teaching material classification 2), and history 1. The data molding means of (teaching material classification 3) is illustrated. In the data molding means of the teaching material video data whose teaching material ID is VD_CSR1, the time stamp information and the word in one sentence are obtained from the information converted into text by the text conversion means and the information (teaching material name, explanation, video length) of the teaching material video data. Extract the teaching material ID, description, and teaching material name.

For example, the data molding means of the teaching material data (for example, teaching material moving image data) whose teaching material ID is VD_CSR1 molds the data as follows.
"Playback start point = 24.22, playback end point = 25.31, word in one sentence =" first "
Playback start point = 25.83, playback end point = 27.79, word in one sentence = "Mesopotamian civilization"
Playback start point = 28.42, playback end point = 29.96, word in one sentence = "one of the four major civilizations"
Teaching material ID = VD_CSR1
Conversion 1 sentence = First, the Mesopotamian civilization is one of the four major civilizations Explanation = Learning the Mesopotamian civilization, which is the four major civilizations (for junior high school students)
Teaching material name = the origin of the four major civilizations "

Here, in the text conversion means, information is linked (associated) in the order of a word in one sentence, a playback start point, and a playback end point, but in the data molding means, the playback is in a different order from the text conversion means. Information is linked (corresponded) in the order of start point, playback end point, and word in one sentence, and processing is executed. The data molding means is a process of associating information in an order different from that of the text conversion means, and is configured to execute a process based on the playback start point. Therefore, the information is matched to the order of the playback start points. Since it is configured so that the data can be arranged side by side, data molding becomes easy.

Next, the indexing means V150-5 indexes the data molded by the data molding means into the format of the search database. As for the teaching material data (for example, teaching material video data) whose teaching material ID is VD_CSR1, the teaching material ID, conversion sentence, word in one sentence, playback start point, and playback end point are stored in the search database V150-9 as shown in FIG. ing. Similarly, as shown in FIG. 7, the teaching material video data having the teaching material ID of VD_KSBK1 stores the teaching material ID, the converted sentence, the word in one sentence, the reproduction start point, and the reproduction end point in the search database V150-9. In this way, the teaching material ID, conversion sentence, word in one sentence, playback start point, and playback end point of each teaching material video data are stored in the search database V150-9 in the order of the teaching material ID.

Next, with reference to FIGS. 8 to 10, the search execution means between the communication terminal T110 and the server V100 of the education system 20 will be described in detail. FIG. 8 shows a system flow in which a learner who is a user accesses the server V100 using the communication terminal T110 and logs in by the user login means shown in FIG. 4, wherein the user uses the communication terminal T110 to log in to the server. This is a system flow for accessing V100 and searching for teaching material data (for example, teaching material video data) that meets the conditions.

(System flow / Search execution means / Communication terminal T110 processing 1G)
First, the processing of the communication terminal T110 is executed. First, in step 502-S, the communication terminal T110 sets the keyword in the search form based on the fact that the learner operates the communication terminal T110 and inputs the search keyword according to the format of the search form. For example, as shown in FIG. 9, "Mesopotamia" is input as a search keyword. Next, in step 504-S, based on the fact that the search button shown in FIG. 9 is selected by the learner, the communication terminal T110 transmits the search form (completed) to the server V100 side via the network 820. To do.

(System flow / Search execution means / Server V100 processing 1H)
Next, the process of the server V100 is executed. First, in step 506-S, the search execution means V140 receives the search form (completed) transmitted from the communication terminal T110. Next, in step 508-S, the search execution means V140 uses the data of the conversion sentence stored in the search database V150-9 as the search keyword entered in the received search form (entered). Perform a based search. For example, from the data of the conversion sentence stored in the search database V150-9 of FIG. 7, the conversion sentence including the search keyword "Mesopotamia" is searched.

Next, in step 510-S, the search execution means V140 acquires the teaching material ID hit by the keyword search. When a plurality of teaching material IDs are hit, the search execution means V140 acquires all the teaching material IDs for all the hits. For example, when the search keyword "Mesopotamia" is searched from the data of the conversion sentence stored in the search database V150-9 of FIG. 7, "Mesopotamia" is added to the conversion sentence of the teaching material IDs VD_CSR1 and VD_KSBK1. Since it is included (see the search database in FIG. 7), VD_CSR1 and VD_KSBK1 are acquired as teaching material IDs. Information on VD_CSR1 and VD_KSBK1 which are different subjects in this way is acquired by searching.

Next, in step 512-S, the search execution means V140 uses the learning material database V150-8 corresponding to the acquired teaching material ID to classify 1 to 4, teaching material names, teaching material data (for example, teaching material video data), and teaching materials. Acquires length (for example, video length) information. For example, when the search keyword "mesopotamia" is searched from the data of the conversion sentence stored in the search database of FIG. 7 and VD_CSR1 and VD_KSBK1 are acquired as the teaching material IDs, the learning material shown in FIG. 7 is obtained. Information on teaching material classifications 1 to 4 having a teaching material ID of VD_CSR1, teaching material name, teaching material data (for example, teaching material video data), and teaching material length (for example, video length) is acquired from the database V150-8. Similarly, from the learning teaching material database V150-8 shown in FIG. 7, the teaching material ID is VD_KSBK1, the teaching material classifications 1 to 4, the teaching material name, the teaching material data (for example, the teaching material video data), and the teaching material length (for example, the video length). Information, the information of the conversion sentence is acquired from the search database V150-9 shown in FIG.

Next, in step 514-S, the search execution means V140 generates a search result screen based on the acquired information. For example, when the search keyword "Mesopotamia" is searched from the data of the conversion sentence stored in the search database V150-9 of FIG. 7, the search result screen as shown in FIG. 9 is generated. On the search result screen, score information, teaching material classifications 1 to 4, teaching material length (for example, video length) information, conversion one sentence information (one conversion one sentence, multiple conversion one sentence, all conversions) The information of one of the conversion sentences of one sentence) and the detail button (for example, the button image of detail 1 and detail 2 shown in FIG. 9) are displayed. As shown in FIG. 9, the search keyword "Mesopotamia" is displayed as bold or italic characters or highlighted so that the information of the converted sentence of the search result screen can be easily seen by the learner. It is preferable to do so. The underline between "What you study" and "The law of the Orient world" in Fig. 9 shows the division between words in one sentence and the video. Indicates a break in audio at. Next, in step 516-S, the search execution means V140 transmits the generated search result screen to the communication terminal T110 via the network 820. In addition, as the search result screen, a smaller image (a reduced image that allows you to check at a glance what kind of image it is and what kind of content it is) is displayed (thumbnail display). It may be configured to display as many as the number of search results (s).

Here, the score information is configured so that the score becomes higher as the number of search keywords is included, but the teaching material data already learned (for example, teaching material video data) based on the learner's past learning history. The score may be higher. Further, the score of the teaching material classification that the learner learned the most may be high based on the statistics of the teaching material classifications 1 to 4 associated with the already learned teaching material data (for example, the teaching material video data). By displaying the score information in this way, it is possible to provide a service called recommendation that presents the appropriate information that the learner wants, and if there is a homonym in the search keyword, the learner's learning history You can display the search results along.

(System flow / Search execution means / Communication terminal T110 processing 2G)
Next, the processing of the communication terminal T110 is executed. First, in step 518-S, the communication terminal T110 receives the screen of the search result transmitted from the server V100. Next, in step 520-S, the communication terminal T110 displays the received search result screen on the display device. For example, the search result screen as shown in FIG. 9 is displayed. Next, in step 522-S, the communication terminal T110 is detailed based on the fact that the learner operates the communication terminal T110 and the detail buttons (button images of details 1 and 2 shown in FIG. 9) are selected. The display request (including the teaching material ID) is transmitted to the server V100 via the network 820.

(System flow / Search execution means / Server V100 processing 2H)
Next, the process of the server V100 is executed. First, in step 524-S, the search execution means V140 receives the detailed display request (including the teaching material ID) transmitted in step 522-S. Next, in step 526-S, the search execution means V140 acquires teaching material data (for example, teaching material moving image data) of the learning teaching material database V150-8 based on the information of the teaching material ID of the received detailed display request. For example, when the button of detail 1 is selected in FIG. 9, since the teaching material ID is VD_KSBK1, the teaching material video data in which the teaching material ID of the learning teaching material database V150-8 is VD_KSBK1 is acquired.

Next, in step 528-S, the search execution means V140 provides information on the one-sentence word including the keyword and the one-sentence word before and after the one-sentence word containing the keyword among the one-sentence words in the information of the teaching material ID of the detailed display request. Is acquired from the search database V150-9 in order from the earliest playback start point. For example, when the search keyword is "Mesopotamia" and the teaching material ID is VD_KSBK1, the playback start point including "Mesopotamia" from the search database V150-9 is the first word in the sentence "Mesopotamia civilization". The playback start point is the word in the first sentence, "This is the Mesopotamian civilization," which is the word in the sentence before "The first civilization that emerged," and the word in the sentence after, "Here is." It's important. " Also, from the search database V150-9, "Mesopotamia", which has a playback start point containing "Mesopotamia" in the second sentence, and "Mesopotamia", which has a playback start point in the second sentence, Get information about the word "here" in the previous sentence and the word "in Greek" in the latter sentence.

Next, in step 530-S, the search execution means V140 searches the search database V150-9 for information on the reproduction start point of the word in the sentence including the keyword among the words in the sentence of the teaching material ID information of the detailed display request. Acquires the playback start points in the order of earliest. For example, when the search keyword is "Mesopotamia" and the teaching material ID is VD_KSBK1, the playback start point including "Mesopotamia" from the search database V150-9 is the first word in the sentence "Mesopotamia civilization". The information of 13.58, which is the playback start point of "is", is acquired. In addition, the information of 16.34, which is the reproduction start point of "Mesopotamia", which is the second word in one sentence, is acquired from the search database V150-9.

Next, in step 532-S, the search execution means V140 determines the teaching material data (video) with audio based on the acquired teaching material data (for example, teaching material video data) information, word information in one sentence, and playback start point information. Generate a screen. For example, a screen as shown in FIG. 10A is generated. On the screen of the teaching material data (video) with audio as the result screen of the detailed display request (display device of the communication terminal T110 as shown in FIG. 10A), the teaching material data with audio (video) is located at the upper position of the screen. ) Is displayed, and the play button, fast forward button, and play start point 13.58, which is the play start point of "Mesopotamian civilization", which is the first word in one sentence, are displayed at the middle position of the screen. If the playback button is selected in this state, the video is played from 13.58, which is the playback start point of the first word in the sentence, "Mesopotamian civilization." The character string data of the teaching material data (video) with audio is displayed at the lower position of the screen. The character string data includes a playback start point, a word in a sentence containing a search keyword, a word in a sentence before a word in a sentence containing a search keyword, and a sentence after a word in a sentence containing a search keyword. Display inside words.

Further, in the character string data, for example, when the search keyword is "Mesopotamia" and the teaching material ID is VD_KSBK1, the playback start point including "Mesopotamia" is the first sentence from the search database V150-9. "It's a Mesopotamian civilization", which is an internal word, and "The first civilization that emerged", which is the word before "It's a Mesopotamian civilization", which is the first word in the sentence, and the following sentence. If you get the information that the internal word "here is important", you may display "13.58 civilization is Mesopotamian civilization here is important". In other words, from the word "Mesopotamian civilization" before the word "Mesopotamian civilization" that includes the search keyword, "The first civilization that emerged" is changed to the word "This is important". Also, the word "first risen" should be omitted so that it will be shortened. In addition, the word in the sentence before "Mesopotamian civilization", which is a word in the sentence including the search keyword, is the word in the sentence with a fixed number of characters such as 3 characters, and the word in the sentence after is the word with a fixed number of characters such as 5 characters. It may be an internal word. By shortening (and a fixed number of characters) the word in the sentence that is located before the word in the sentence that includes the search keyword, the position of the word in the sentence that includes the search keyword can provide a screen that is easy for the learner to see. It will be possible.

In addition, the word "Mesopotamian civilization" before the word "Mesopotamian civilization" that includes the search keyword "The first civilization that emerged" and the word "This is important" after that. However, it is preferable to abbreviate the word so that it is shorter than the word "Mesopotamian civilization" that includes the search keyword. It is preferable to omit the preceding letter if it is a word in the preceding sentence, and omit the following letter if it is a word in the following sentence (for example, "13.58 civilization is a Mesopotamian civilization, this is important"). Is. With this configuration, it is possible to provide a screen in which the position of a word in a sentence including a search keyword is easy for the learner to see.

In addition, the search keyword of "Mesopotamian civilization" which is a word in one sentence including the search keyword, or the word "Mesopotamian civilization" itself which is a word in one sentence including the search keyword is bolded, italicized, and underlined. It is also preferable to emphasize by the characters used and highlighting. By doing so, it is possible to emphasize and display the words in one sentence including the search keyword, so that it is possible to provide a screen in which the learner can easily understand the search result.

Next, in step 534-S, the search execution means V140 transmits the screen of the generated teaching material data (video) with audio to the communication terminal T110 via the network 820.

(System flow / Search execution means / Communication terminal T110 processing 3G)
Next, the processing of the communication terminal T110 is executed. First, in step 536-S, the communication terminal T110 receives the screen of the teaching material data (video) transmitted in step 326-S. Next, in step 538-S, the communication terminal T110 displays the screen of the received teaching material data (video) with audio on the display device. For example, a screen as shown in FIG. 10A is displayed on the display device. Next, in step 540-S, the communication terminal T110 reproduces the teaching material data (video) with audio on the display device based on the reproduction operation in which the learner selects the triangular play button shown in FIG. 10 (a). indicate. For example, when the teaching material ID is VD_KSBK1, the playback start point including "Mesopotamia" and the first word in one sentence, "It is a Mesopotamian civilization", and the word in the previous sentence, "First rise". The play button is selected when "13.58 civilization is a Mesopotamian civilization is important" is displayed, which is the information "This is important", which is the word in the latter sentence. In that case, the playback start point that includes "Mesopotamia" is the first, as shown by the instructor's balloon in Fig. 10 (a), without being reproduced from the previous word in the sentence, "The first civilization that emerged." It is reproduced from the word "It is a Mesopotamian civilization" in one sentence.

When the audio-attached teaching material data (video) is reproduced and displayed based on the learner's playback operation, a conversion sentence (for example,) generated by the audio extraction means V150-2 and the text conversion means V150-3 from the audio of the audio-attached teaching material data , "The first civilization that emerged is the Mesopotamian civilization, which is important") may be displayed as a telop at a predetermined position on the screen of the display device of the communication terminal T110. In the case of displaying a telop of a converted single sentence in which the audio of the teaching material data with audio is converted into text in this way, and if there is a board A as shown in FIG. 10 (b), as shown in FIG. 10 (b). It is preferable to display the telop at a position that does not overlap with the board A. With this configuration, even in an environment where it is difficult to hear the voice, it is possible to check and learn the telop in which the voice of the teaching material data with voice is converted into text. Further, a button for outputting this telop as text data (“telop output” button shown in FIG. 10B) may be provided so that the text data can be output by the learner's selection operation. The learner can create a material with the content that suits him / her by collecting the text data of the key points. It is preferable that the character size of the telop is displayed by using a character whose character size is larger than the character size of the character string data displayed on the lower side of the screen of the communication terminal T110. Further, the character size of the telop may be set to the maximum character size among the characters displayed on the screen of the communication terminal T110.

Further, as shown in FIG. 10A, when there is a board writing A in the teaching material data (video) with audio, the content of the board writing A is converted into text to generate the board writing characters, and a predetermined screen of the display device of the communication terminal T110 is generated. The board writing B may be displayed at a position (a position that does not overlap with the board writing A). With this configuration, even if the characters written on the board A are difficult to see in an environment where the resolution of the image is low, it is possible to check and learn the text-written board B. .. It is preferable that the character size of the board writing B is displayed by using a character whose character size is larger than the character size of the board writing A. Further, a button for outputting the board writing B as text data (“board writing output” button shown in FIG. 10B) may be provided so that the text data can be output by the learner's selection operation. Learners can create materials with their own content by compiling the text data on the board at the key points.

Further, as shown in FIG. 10A, when there is a board writing A in the teaching material data (video) with audio and there is a lecturer in front of the board writing A, image processing for erasing the video of the lecturer is executed. The entire board may be confirmed as shown in FIG. 10 (b). In this image processing, not only the instructor is deleted, but also the characters on the board written behind the instructor are complemented based on the characters on the board displayed in the video data of other times. Is preferable. With such a configuration, the contents of the entire board can be checked at all times, so that learning can be performed efficiently. In addition, the contents of the board writing (board writing A, board writing B) may be searchable. Efficient learning can be achieved by searching the contents of the instructor's board in the teaching material data by keyword and learning.

When the teaching material data (video) with voice is played back and displayed based on the learner's playback operation, the voice of the instructor's dialogue of the teaching material data with voice is generated by the voice extraction means V150-2 and the text conversion means V150-3. All transformational sentences (for example, "The first civilization in the Near East was the Mesopotamian civilization, which is important here. Mesopotamia is the Greek word for the region between rivers ..."), as shown in Figure 11 ( As shown in a), the server V100 may be provided with a full-text display means for displaying a full-text display up to the number of characters that can be displayed at a predetermined position on the screen of the display device of the communication terminal T110.

As shown in FIGS. 11A and 11B, the full-text display means is a red frame (referred to as a red frame) indicated by a broken line, and a predetermined number of conversion sentences (one conversion) out of all conversion sentences. It is configured to enclose one sentence or a plurality of conversion sentences (three conversion sentences in FIGS. 11 (a) and 11 (b)). As shown in FIG. 11A, when the teaching material data (video) with audio is reproduced and displayed from the beginning, the reproduction time of the teaching material data (video) with audio is the beginning of the first predetermined number of conversion sentences. Regeneration of the first conversion sentence ("Mesopotamian civilization is the first civilization in the ancient Orient") and the last conversion sentence ("Mesopotamia means the area between rivers in Greek") When the full-text display means determines that it is the time between the end points, the full-text display means displays a red frame so as to surround the first predetermined number of conversion sentences.

The full-text display means is that the playback time of the teaching material data (video) with audio is the first predetermined number of conversion sentences and the last conversion sentence ("Semitic and Indo-European nomadic mountains due to the flat and open terrain." Beyond the time of the end point of regeneration of the hills and other areas where the war was constantly rising, the first conversion sentence of the second predetermined number of conversion sentences ("Mesopotamian civilization is the Tigris River Euphrates River basin" It is a city civilization created in the Mesopotamia region. ”) When it is determined that it has become the playback start point, the first predetermined number of conversions is scrolled up and the first predetermined number of conversions is performed from the display device screen. Hide one sentence (may be displayed so that part of it can be seen), and scroll up the second predetermined number of conversions, and the red frame indicates the second predetermined number of conversions. Enclose a sentence. At this time, the red frame is configured not to move, but the size of the red frame is changed according to the length of a predetermined number of conversion sentences.

And the full-text display means is that the playback time of the teaching material data (video) with audio is the last conversion sentence of the second predetermined number of conversion sentences ("Mesopotamia here means the area between rivers in Greek". ”) Beyond the time of the playback end point, it is judged that it became the playback start point of the first conversion sentence of the third predetermined number of conversion sentences (“Mesopotamian civilization was said to be a clay civilization”). If this is the case, scroll up the second predetermined number of conversion sentences to hide the second predetermined number of conversion sentences from the display device screen (even if a part of the conversion sentence is displayed so that it can be seen). (Good), scroll up the third predetermined number of conversion sentences so that the red frame surrounds the third predetermined number of conversion sentences. Although the full-text display means is configured not to move the red frame, it may be configured to move the red frame as shown in FIG. 11B.

In this way, when the teaching material data (video) with audio is continuously played on the screen of the display device of the communication terminal T110, the teaching material data (video) with audio is linked (corresponding) with the voice of the instructor's dialogue being played. The full text display is configured to automatically scroll up so that a predetermined number of conversion sentences are displayed in the red frame. With this configuration, even in an environment where it is difficult to hear the voice, it is possible to confirm and learn all the converted sentences in which the voice of the teaching material data with voice is converted into text. When a predetermined number of conversion sentences are one conversion sentence, the playback start point of the first conversion sentence and the playback end point of the last conversion sentence are targeted at the playback start point and the playback end point of one conversion sentence. To do.

Here, a predetermined number of converted single-sentence characters that cannot be displayed in the full-text display are configured to be displayed on the screen of the display device by the learner scrolling down the full-text display using a mouse or a touch panel. When the full-text display is scrolled down by the learner's operation, as shown in FIGS. 11 (a) and 11 (b), the full-text display means moves within the red frame to enter the red frame. Change the character of the conversion sentence displayed in. When the learner scrolls down the full-text display using the mouse or touch panel, the playback start point of the audio-attached teaching material data (video) is changed, and the full-text display means is the audio-attached teaching material data. It is determined at which position the changed playback start point of (video) is included between the playback start point and the playback end point of a predetermined number of conversion sentences, and the playback start point and the playback end point of the judgment result are determined. The red frame is moved and displayed so as to surround the predetermined number of conversion sentences indicated by.

In addition, by scrolling down the full-text display by the learner's operation, it is linked to the characters of the conversion sentence displayed in the red frame (for details, the playback start point indicated by the conversion sentence displayed in the red frame). The full-text display means changes the content of the video of the teaching material data with audio displayed at the upper position of the screen of the display device of the communication terminal T110 (in FIGS. 11A and 11B). The position of the instructor has been changed).

It should be noted that the audio of the teaching material data with audio may be converted into text by using a plurality of text conversion means V150-3. In the case of such a configuration, the first conversion one sentence (a word in one sentence is also possible) converted into text by the first first text conversion means V150-3 and the second second text conversion means V150-3. It may be different from the second conversion sentence (words in one sentence are also acceptable) converted into text in. For example, the first text conversion means V150-3 converted the text to "Ancient Near East", the second text conversion means V150-3 converted it to "Small Orient", and the first text conversion means V150-3 converted it to text. When displaying "Ancient Orient", enclose "Ancient Orient", which is a different conversion sentence (words in one sentence are also acceptable), in parentheses, such as "(Ancient Orient) first ...". It is preferable to indicate the possibility that there is an error in the textual content. In this way, there is a possibility that there is a conversion error in the conversion sentence generated by the text conversion means V150-3. Therefore, on the screen of the display device of the communication terminal T110 (for example, the lower right position), "automatic transcription sentence" is displayed. ] Is preferably displayed.

The education system (1) according to this embodiment is
Reproducible teaching material data A with audio (for example, teaching material video data of junior high school social history 1),
Subject data A (for example, information on teaching material classifications 1 to 4) indicating the subject of the teaching material data A with audio, and
The voice data extracted from the teaching material data A with voice is converted into character string data A (for example, the teaching material ID is VD_CSR1) as a conversion sentence of a predetermined conversion means (for example, voice extraction means, text conversion means). The Mesopotamian civilization is one of the four major civilizations. ")
Reproducible audio-attached teaching material data B (for example, high school world history B ancient 1 teaching material video data),
Subject data B (for example, information of teaching material classifications 1 to 4) indicating the subject of the teaching material data B with audio, and
The voice data extracted from the teaching material data B with voice is converted into character string data B (for example, the teaching material ID is VD_KSBK1) as a conversion sentence of a predetermined conversion means (for example, voice extraction means, text conversion means). The first civilization to emerge in the Orient was the Mesopotamian civilization. ")
An educational system capable of providing the audio-based teaching material data A and the audio-based teaching material data B as teaching materials for different subjects based on the subject data A and the subject data B.
It is possible to accept input of a search query (for example, "Mesopotamia" as input of a search keyword) and execute a search for both the character string data A and the character string data B.
The search query (eg, "Mesopotamia") is included in the string data A (eg, "First, the Mesopotamia civilization is one of the four major civilizations") and the search query is included in the string data B (eg, "Mesopotamia"). If it is determined that the civilization that first emerged in the ancient Orient is also included in the Mesopotamian civilization "), the search result data A for accessing the teaching material data A with audio (for example, VD_CSR1 as the teaching material ID, teaching material classification) 1 to 4, teaching material name, teaching material video data, video length, conversion sentence, thumbnail display) and search result data B for accessing the teaching material data B with audio (for example, VD_KSBK1 as teaching material ID, teaching material classification 1 to 4) 4. The educational system is characterized by outputting the teaching material name, teaching material video data, video length, conversion sentence, thumbnail display) as a series of search results.

The education system (2) according to this embodiment is
The character string data A (for example, "First, the Mesopotamia civilization is one of the four major civilizations") contains a plurality of spoken sentence data (for example, "First" as a word in one sentence, "Mesopotamia civilization is", and "Four major civilizations". If it is determined that the search query (for example, "mesopotamia") is included in a certain utterance sentence data in the character string data A, the voice data corresponding to the certain utterance sentence data Data that can identify which playback point of the audio-attached teaching material data A is attached (for example, a playback start point corresponding to VD_CSR1 as the teaching material ID) is output together with the search result data A.
The character string data B (for example, "the first civilization that emerged in the ancient Orient is the Mesopotamian civilization") is a plurality of spoken sentence data (for example, "in the ancient Orient" as a word in one sentence, "first emerged". If it is determined that the search query (for example, "Mesopotamia") is included in a certain utterance sentence data in the character string data B, the search query (for example, "Mesopotamia") is composed of "Civilization is" and "Mesopotamia civilization"). Data that can identify which playback point of the audio-attached teaching material data B the corresponding audio data is attached to (for example, the reproduction start point corresponding to VD_KSBK1 as the teaching material ID) is combined with the search result data B. This is the education system described in the education system (1) according to the present embodiment.

The education system (3) according to this embodiment is
The character string data A (for example, "First, the Mesopotamia civilization is one of the four major civilizations") contains a plurality of spoken sentence data (for example, "First" as a word in one sentence, "Mesopotamia civilization is", and "Four major civilizations". If it is determined that the search query (for example, "mesopotamia") is included in a certain utterance sentence data in the character string data A, the utterance sentence before the certain utterance sentence data is determined. At least one of the data (for example, "first" as a word in one sentence) and the utterance data after the certain utterance data (for example, "one of the four major civilizations" as a word in one sentence) It is added to the spoken sentence data (for example, "Mesopotamia civilization" as a word in one sentence) and output together with the search result data A.
The character string data B (for example, "the first civilization that emerged in the ancient Orient is the Mesopotamian civilization") is a plurality of spoken sentence data (for example, "in the ancient Orient" as a word in one sentence, "first emerged". If it is determined that the search query (for example, "mesopotamia") is included in a certain utterance sentence data in the character string data B, the search query (for example, "mesopotamia") is composed of "civilization is" and "is a mesopotamia civilization"). At least one of the previous utterance data (for example, "in the ancient Orient" as a word within a sentence) and the utterance data after the certain utterance data (for example, "Mesopotamian civilization" as a word within a sentence). Description of the education system (1) according to the present embodiment, which is added to the certain spoken sentence data (for example, "the first civilization that emerged" as a word in one sentence) and output together with the search result data B. Education system.

After learning in chapters using video as in the past, by performing a search using keywords from teaching materials of different subjects and learning the contents related to the contents learned in chapters, the degree of understanding is further improved. Can be deepened. In other words, the content learned systematically (learning along the vertical axis) can be further skewered and learned (learning along the horizontal axis), so relationships are built in the mind, deepening understanding, and memory. You can increase the number of withdrawals.

(Details of the present embodiment)
As mentioned above, the invention has been described in accordance with this embodiment, but the descriptions and drawings that form part of this disclosure should not be understood to limit the invention. As described above, it goes without saying that the present invention includes various embodiments not described here.

10 Test system 20 Education system T110 Communication terminal V100 Server 820 Network V110 User information registration means V120 User login means V130 Test execution means V140 Search execution means V150 Data management means V130-1 Fraud monitoring means V150-1 Format conversion means V150-2 Voice Extraction means V150-3 Text conversion means V150-4 Data molding means V150-5 Indexing means V150-6 Examination teaching material database V150-7 User authentication database V150-8 Learning teaching material database V150-9 Search database

Claims (3)

A test system that provides test content to authenticated users and enables input of answers to the test content within a predetermined test time.
It is possible to execute fraud monitoring processing that calculates the matching rate between the image data input by the imaging means and the face image data stored in advance corresponding to the authenticated user.
It is possible to execute the fraud monitoring process at a plurality of monitoring timings according to the passage of time within the predetermined test time, and to determine whether or not the response to the test content is fraudulent based on the consistency rate. A test system featuring. The test system according to claim 1, wherein it is possible to determine whether or not the response to the test content is incorrect by adding information on the response to the test content in addition to the consistency rate. The test according to claim 1, wherein the face image data is image data in which the front face is captured, and the matching rate is lowered even when the front face is not captured in the image data input by the imaging means. system.

Images