Method for determining correctness, information processing device, and program

Abstract

To provide a correctness determination method that allows efficient learning of words.SOLUTION: In a correctness determination method, a computer performs: obtaining an image and a first word related to the image, and a second word that is an answer which is input by a user for the image displayed on a display unit; and executing a correctness determination for the second word, based on the first word and a third word which is determined based on aggregate results of words which are answered for the image by a plurality of other users.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an error determination method, an information processing apparatus, and a program.

Background Art

[0002] In recent years, there has been considered a word learning method of associating English words directly with an image like a native speaker instead of memorizing English words in Japanese translation. For example, there is a learning tool that directly associates an image and an English word for a user to memorize by displaying an image and the English word associated with the image.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, it is not very common that only one English word is associated with an image, and a plurality of words can be associated. Therefore, simply displaying a predetermined English word for one image is limited as an effect of the word learning method for associating an English word directly with an image and mastering it.

[0005] The present invention has been made in consideration of the above problems, and an object thereof is to provide an error determination method, an information processing apparatus, and a program capable of effectively performing word learning.

Means for Solving the Problems

[0006] To solve the above problems, the correct / incorrect determination method according to an embodiment of the present invention involves a computer acquiring an image, a first word related to the image, and a second word which is the answer entered by the user for the image displayed on the display unit, and then performing a correct / incorrect determination on the second word based on the first word and a third word determined based on the aggregated result of words answered by multiple other users for the image. death , The system acquires ranking data for words answered by multiple other users in response to the image. This ranking data is generated based on the aggregated results for each word answered by multiple other users and includes a ranking list corresponding to the pronunciation quality of the voice input responses to the image by multiple other users. When a user voice-inputs a second word in response to the image, the system determines the pronunciation quality of the input voice and displays the ranking list selected according to the determined pronunciation quality, along with the result of the correct / incorrect judgment for the second word, on the display unit. . [Effects of the Invention]

[0007] According to the present invention, it is possible to provide a method for determining correctness, an information processing device, and a program that enable effective word learning. [Brief explanation of the drawing]

[0008] [Figure 1] A functional block diagram showing the configuration of the electronic circuit of an information processing device according to an embodiment of the present invention. [Figure 2] A functional block diagram showing the configuration of the electronic circuit of a server according to an embodiment of the present invention. [Figure 3] A diagram showing an example of the data included in the user management data 32h in this embodiment. [Figure 4] A diagram illustrating the outline of the image word test in this embodiment. [Figure 5] A diagram illustrating the outline of the image word test in this embodiment. [Figure 6] A flowchart illustrating the image word test process of the information processing device in this embodiment. [Figure 7] A flowchart illustrating the image word test provision process of the server in this embodiment. [Figure 8] A flowchart showing the server's image word test data generation process (1) in this embodiment. [Figure 9] A flowchart showing the server's image word test data generation process (2) in this embodiment. [Figure 10] This figure shows an example of the screen displayed when a question is presented in the image-based word test according to this embodiment. [Figure 11] A diagram showing an example of a screen for the result of correct / incorrect determination. [Figure 12] A diagram showing an example of a screen for the result of correct / incorrect determination. [Figure 13] A diagram showing an example of a screen for the result of correct / incorrect determination. [Figure 14] A diagram showing an example of a screen for the result of correct / incorrect determination.

Embodiments for Carrying Out the Invention

[0009] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0010] FIG. 1 and FIG. 2 are block diagrams showing the configuration of a system according to an embodiment of the present invention. FIG. 1 shows the detailed configuration of the electronic circuit of the information processing apparatus 10 in the present embodiment, and FIG. 2 shows the detailed configuration of the electronic circuit of the server 30 in the present embodiment.

[0011] The system in the present embodiment enables a user who uses the information processing apparatus 10 to perform an image-word test provided by the server 30.

[0012] The image-word test is, for example, aimed at directly associating a word with an image (picture) for memorization. By displaying an image and having the user input (character input or pronunciation) the word associated (related) with the image by association, it is a word test using an image that presents the result of correct / incorrect determination for the answer.

[0013] The image includes, for example, illustrations, photos, graphics, characters, still images, moving images, animations, etc. Also, the word to be answered may be input in units of one word, or may be input in units of a sentence containing words or an idiom (phrase, set phrase, familiar phrase, etc.).

[0014] In this embodiment, an example in which the information processing apparatus 10 is configured as, for example, a smartphone will be described. Note that the information processing apparatus 10 can be realized by various electronic devices such as a personal computer, a tablet PC, a game apparatus, and an electronic dictionary.

[0015] The information processing apparatus 10 is connected to the server 30 through the network 8. Further, the server 30 is connected through the network 8 to, for example, an education server 40 provided with a function of providing educational content and an electronic dictionary 50 having a communication function equipped with a dictionary search function.

[0016] The network 8 includes various communication paths capable of wired or wireless communication, such as the Internet, a LAN (Local Area Network), and a public telephone line network.

[0017] As shown in FIG. 1, the information processing apparatus 10 has a configuration of a computer that reads a program recorded on various recording media or a transmitted program and controls its operation according to the read program, and a CPU (central processing unit) 11 is provided in the electronic circuit.

[0018] The CPU 11 functions as a control unit that controls the entire information processing apparatus 10. The CPU 11 controls the operation of each circuit unit according to a program stored in advance in the memory 12, a program read from a recording medium 13 such as a ROM card into the memory 12 via the recording medium reading unit 14, or a program downloaded from an external apparatus (such as a server) that provides a program through the network 8 such as the Internet and read into the memory 12.

[0019] The program stored in the memory 12 is activated according to an input signal corresponding to a user operation from the input unit 16 or a connection communication signal with a recording medium 13 such as an EEPROM (registered trademark), a RAM, or a ROM connected via the recording medium reading unit 14.

[0020] The CPU 11 is connected to memory 12, recording medium reading unit 14, communication unit 15, input unit 16, display unit 17, audio input unit (microphone) 18, audio output unit (speaker) 19, and other components.

[0021] The programs stored in memory 12 include the basic program (OS: Operating System) 12a that controls the entire device, as well as various application programs. The application programs include an image word test program 12b for performing the image word test provided by the information processing device 10. The CPU 11 executes the image word test processing described later using the image word test program 12b (see Figure 6).

[0022] The image word test program 12b includes a speech recognition program that performs speech recognition processing on the audio input from the audio input unit 18 to detect the spoken words (or sentences). Note that the speech recognition program may be separate from the image word test program 12b. In this case, the image word test program 12b obtains the processing results from the speech recognition program and uses them for the image word test processing.

[0023] Memory 12 stores image word test data 12c, response data 12d, user identification data 12e, and other data related to the execution of the image word test program 12b (image word test processing).

[0024] Image word test data 12c is data provided by server 30 for conducting the image word test. For each question, image word test data 12c includes, for example, image data 12c1 of the image used in the image word test, response ranking data 12c2 showing the response ranking obtained by aggregating response data (words) from multiple other users to the image and ranking them based on frequency of occurrence, number of responses, etc., and correct answer data 12c3 showing a pre-set correct word (first word) related to the image and a correct word (third word) determined based on the aggregated results of the response ranking. The frequency of occurrence will be described later.

[0025] Image data 12c1 may be data for a single image or data for multiple images. The answer ranking data 12c2 may be data for a single answer ranking for the image in question, or data for multiple answer rankings corresponding to multiple user classifications. The pre-set correct answer word (first word) indicated by the correct answer data 12c3 may be one or more, and the correct answer word (third word) determined based on the aggregated results of the answer rankings may also be one or more for each answer ranking. The answer ranking data includes at least the first word and the third word, and includes data in list (table) format (ranking list) representing rankings based on the frequency of occurrence or the number of answers for each word. The ranking list will be described later.

[0026] The pre-set correct answer words related to an image are, for example, the "manufacturer-designated correct answer" words set by the creator of the image word test. The correct answer words determined based on the aggregated results of the answer rankings ("the answer chosen by everyone") are, for example, words whose frequency of appearance among the answers (words) of multiple other users who took the image word test meets a predetermined condition.

[0027] The response data 12d is data that shows the content (words) of the answers to each question (image) entered by the user through user operation (key operation, voice input) in an image word test conducted based on the image word test data 12c. After the image word test is conducted, the response data 12d is sent to the server 30 and used for image word test data generation processing for aggregating response rankings and determining correct answers based on the aggregated response rankings. In addition, response time data indicating the time taken to answer, and voice quality data indicating the quality of the voice (pronunciation) when voice input is used can be added to the response data 12d. The response time data and voice quality data are used for classification when aggregating the response data 12d in the server 30.

[0028] User identification data 12e is data used to identify a user (or the information processing device 10) to use the information processing device 10 on the server 30, and includes, for example, a user ID and password that are pre-configured for each user.

[0029] The communication unit 15 performs communication control to communicate with other information processing devices such as a server 30 via a network 8 including the Internet and a LAN (Local Area Network), or to perform communication control to communicate with other information processing devices at close range using short-range wireless communication such as Bluetooth® or Wi-Fi®.

[0030] The input unit 16 is for inputting various data and instructions through user operations, and includes, for example, a keyboard, buttons, and a touch panel integrated with the display unit 17. With the touch panel, for example, data such as characters can be input by touching the software keyboard displayed on the display unit 17.

[0031] The display unit 17 is configured, for example, as a touch panel type display unit with stacked touch panels. The display unit 17 is realized by a display device such as a liquid crystal display (LCD) or an organic electroluminescent (EL) display.

[0032] The voice input unit 18 is a microphone used to input voices spoken by the user.

[0033] The audio output unit 19 is a speaker used to output voice, operation sounds, music, etc.

[0034] In this configuration, the information processing device 10, with its CPU 11 controlling the operation of each circuit component according to instructions written in various programs such as the image word test program 12b, achieves the functions described in the following operation description through the cooperation of software and hardware.

[0035] As shown in Figure 2, the server 30 has a computer configuration in which it reads programs recorded on various recording media or transmitted programs and its operation is controlled by the read programs, and its electronic circuit is equipped with a CPU (central processing unit) 31.

[0036] The CPU 31 functions as a control unit that controls the entire server 30. The CPU 31 controls the operation of each part of the circuit according to programs pre-stored in the memory 32, programs read into the memory 32 from a recording medium 33 such as a ROM card via the recording medium reading unit 34, or programs downloaded from an external device (such as a server) that provides programs via a network 8 such as the Internet and read into the memory 32.

[0037] The program stored in memory 32 is activated in response to an input signal corresponding to user operation from input unit 36, or to a connection communication signal with a recording medium 33 such as EEPROM®, RAM, or ROM connected via recording medium reading unit 34.

[0038] The CPU 31 is connected to a memory 32, a recording medium reading unit 34, a communication unit 35, an input unit 36, a display unit 37, and the like.

[0039] The programs stored in memory 32 include a basic program (OS: Operating System) 32a that controls the entire device, as well as various application programs. The application programs include an image word test management program 32b for providing image word tests to the information processing device 10, and an image word test data generation program 32c for generating image word test questions. The CPU 31 executes the image word test provision process described later using the image word test management program 32b (see Figure 7). The CPU 31 also executes the image word test data generation processes (1) and (2) described later using the image word test data generation program 32c (see Figures 8 and 9).

[0040] Memory 32 stores image word test data 32d, answer database 32e, collected word data 32f, image data 32g, user management data 32h, etc., which are related to the execution of the image word test management program 32b (image word test provision process) and the image word test data generation program 32c (image word test data generation process).

[0041] The image word test data 32d is data provided by the server 30 to the information processing device 10 for conducting the image word test. For each question, the image word test data 32d includes, for example, image data 32d1 of the image used in the image word test, answer ranking data 32d2 showing the answer ranking compiled by summarizing answer data (words) from multiple other users to the image, and correct answer data 32d3 showing a pre-set correct word (first word) related to the image and a correct word (third word) determined based on the summation results of the answer ranking. In other words, the image word test data 32d is data related to the questions for the image word test.

[0042] Image data 32d1 may be data for a single image or data for multiple images. The answer ranking data 32d2 may be data for a single answer ranking for the image in question, or it may be data for multiple answer rankings corresponding to multiple user classifications (such as classifications based on user attribute data described later, or classifications based on user levels indicated by user data). The correct answer data 32d3 may consist of one or more pre-set correct words (first word), and the correct answer words (third word) determined based on the aggregated results of the answer rankings may also consist of one or more for each answer ranking.

[0043] In the answer ranking, a ranking word data is generated for each answer (word), with a representative translation attached, and the word and translation at each rank are displayed. Representative translations can be added by searching for them in a dictionary data (not shown) in which words (headwords) and translations are registered in correspondence for each word.

[0044] The pre-set correct answer words related to an image are, for example, the "manufacturer-designated correct answer" words set by the creator of the image word test. The correct answer words determined based on the aggregated results of the answer rankings ("the answer chosen by everyone") are, for example, words whose frequency of appearance among the answers (words) of multiple other users who took the image word test meets a predetermined condition.

[0045] The answer database 32e is a database in which answer data, which shows the content (words) of answers to each question (image data 32d1) in the image word test conducted by the information processing device 10, is registered in association with the answer data, which is entered by user operation (key operation, voice input). The answer data is obtained from the information processing device 10 after the image word test is conducted by the information processing device 10 and registered in the answer database 32e. The answer data registered in the answer database 32e is used for the image word test data generation process for aggregating answer rankings and determining correct answers based on the aggregated answer ranking results.

[0046] The collected word data 32f is word data acquired from an external device in order to generate the image word test data 32d. Examples of external devices include an educational server 40 connected via network 8 and equipped with functions to provide educational content, and an electronic dictionary 50 with communication functions and a dictionary search function. For example, in the online dictionary search function 40a provided on the educational server 40 or in the electronic dictionary 50, the strings (words) targeted for dictionary search are collected as word data and used as words for the image word test questions.

[0047] Image data 32g is image data acquired to generate image word test data 32d. Image data 32g can be acquired from an image database stored in a server 30 (not shown) or an external device (not shown), which is pre-prepared for creating image word tests, based on the words indicated by the collected word data 32f, or from a group of images searched from a website via the internet, based on the words indicated by the collected word data 32f. The image database is assumed to have at least one keyword (word) associated with each image, and the image corresponding to the keyword can be identified based on the keyword. Each of the multiple collected word data 32f stored in memory 32 is associated with a representative translation of the collected word data 32f and at least one image data 32g that corresponds to (is related to) that translation. For example, if the translation is "wolf" (the recorded word data 32f is "wolf"), then the image 32g of a wolf is associated with keywords such as large canis, animal, and carnivore.

[0048] User management data 32h is data for managing users (or information processing device 10) who use the image word test provided by server 30. For example, it records user attribute data that was registered in advance when using server 30, associated with user identification data (e.g., user ID and password) set in advance for each user, and user data that is updated when the image word test is conducted.

[0049] Figure 3 shows an example of the data included in the user management data 32h in this embodiment. The user management data 32h shown in Figure 3 includes user attribute data such as country (area), age (or date of birth), gender, and language used (Japanese, American English, British English, Chinese, etc.). User data is data that indicates the characteristics of the user, determined based on the user's answers when conducting an image word test. User data includes, for example, vocabulary level (or word ability) determined by analyzing the user's answers, such as the user's vocabulary level, which indicates the part of speech and field tendencies of the questions (words) that are answered correctly, and the number of words used in the answers, as well as the user level corresponding to the time taken to answer when conducting the image word test and the quality of the voice (pronunciation) when voice input is used when conducting the image word test. The time taken to answer in Figure 3 is, for example, the average time (seconds) that the user has taken to answer in image word tests that they have taken so far.

[0050] The vocabulary level and user level can be considered as separate user levels, or they can be combined to form a user level that includes at least two user levels (vocabulary level + time taken to answer, etc.).

[0051] The user attribute data and user data included in the user management data 32h are referenced to classify users (or the information processing device 10 used by the user) when aggregating user responses during the image word test. For example, responses can be aggregated within a range of levels close to the user's level, or by the same level, to generate a response ranking, and the correctness of the responses to the image word test can be determined based on the response ranking corresponding to the user's level. In other words, the correctness of the response can be determined based on a "commonly decided correct answer" determined based on the responses of other users at a similar level to the user's level, or other users at the same level.

[0052] Furthermore, it is not always necessary to determine correctness based on answer rankings corresponding to the user's level; for example, users could be allowed to specify answer rankings corresponding to users of any level. For instance, by having the system determine correctness based on answer rankings corresponding to users at a higher level than the user, it becomes possible to learn from the answers of higher-level users.

[0053] Furthermore, response rankings may be collected individually for each data point included in attribute data and user data, or response rankings may be aggregated for each user by combining multiple data points. Alternatively, the system may determine correctness based on multiple response rankings aggregated for each of several different classifications.

[0054] The communication unit 35 performs communication control to communicate with other information processing devices such as the information processing device 10, the educational server 40, and the electronic dictionary 50 via a network 8 including the Internet, LAN (Local Area Network), and LTE (Long Term Evolution) (registered trademark), or to perform communication control to communicate with other information processing devices in the nearby area using short-range wireless communication such as Bluetooth (registered trademark) and Wi-Fi (registered trademark).

[0055] The input unit 36 ​​is for inputting various data and instructions through user operation, and includes, for example, a keyboard, buttons, and a touch panel integrated with the display unit 37. With the touch panel, for example, data such as characters can be input by touching the software keyboard displayed on the display unit 17.

[0056] The display unit 37 is implemented by a display device such as a liquid crystal display (LCD) or an organic electroluminescence (EL) display.

[0057] In this configuration, the server 30 has a CPU 31 that controls the operation of each part of the circuit according to instructions written in various programs such as the image word test management program 32b and the image word test data generation program 32c. The software and hardware work together to realize the functions described in the following operation description.

[0058] Next, we will describe the outline of the image word test provided by the server 30 in this embodiment.

[0059] Figures 4 and 5 are diagrams illustrating the outline of the image word test in this embodiment.

[0060] The image word test uses image word test data provided by the server 30 to display images on the information processing device 10, as shown in Figure 5(A), and users answer the questions by inputting words corresponding to the images. Generally, there is rarely only one word associated with a single image, and when multiple users respond, multiple different words are entered as answers, as shown in Figure 5(B). In the example shown in Figure 5(B), for the image representing a "wolf" shown in Figure 5(A), the answers included "wolf," "howl," and "roar." As shown in Figure 5(C), the server 30 aggregates the words that multiple users have answered for a single image and compiles them as an answer ranking.

[0061] Server 30 manages response ranking data 32d2, which is a compilation of responses from multiple users for each word, associated with the image data 32d1 of the image shown in Figure 4(A) used for the image word test, and shows the response ranking (ranking list) shown in Figure 4(B).

[0062] The response ranking data 32d2 shown in Figure 4(B) shows that, after aggregating responses from multiple users for each word, the most frequent responses were, in order: "wolf," "howl," and "roar."

[0063] In this embodiment, the server 30 determines the correct word for one image (question) of the image word test as follows (1)(2).

[0064] For example, (1) a word designated as the "manufacturer's designated correct answer" by the creator (manufacturer) of the image word test, and (2) a word designated as the "commonly chosen correct answer" which is determined to meet certain conditions based on the aggregated results of the answer ranking, will be determined as the correct answer. The "commonly chosen correct answer" will be selected from words other than the "manufacturer's designated correct answer".

[0065] The predetermined conditions can be, for example, words whose frequency of appearance is determined to meet the predetermined conditions. The frequency of appearance for each word in the answer ranking is (number of times users answered that word (total count) / (total number of user answers)). For example, words whose frequency of appearance exceeds a predetermined threshold can be considered correct. However, this can be conditional on the total number of user answers being above a predetermined value. In other words, if the total number of user answers is small, the frequency of words that are clearly not correct may appear high even if only a small number of answers are given. By setting the condition that the total number of answers is above a predetermined value, the reliability of the "correct answer decided by everyone" can be increased.

[0066] Furthermore, for multiple words that can be considered correct answers, the number of responses can be added to the answer ranking as a sample beforehand, thereby ensuring that the total number of responses exceeds a certain value. This eliminates the need for the condition that the total number of responses must exceed a certain value.

[0067] Furthermore, the "correct answer chosen by everyone" can be determined not only by a word whose frequency of appearance exceeds a predetermined threshold, but also by other conditions. For example, among the multiple words tallied in the answer ranking, a word whose deviation (bias) in the total number of occurrences is above a threshold can be designated as the "correct answer chosen by everyone."

[0068] In addition, excluding the "manufacturer-designated correct answer," a predetermined number of words from the top of the total count, or a predetermined number of words from the top of the frequency of appearance, may be designated as the "correct answer chosen by everyone."

[0069] Additionally, a pre-set upper limit (e.g., 3 words) or lower limit (e.g., 1 word) for the number of words designated as "the correct answer by everyone" can be set, and the number of words must not exceed either limit.

[0070] Furthermore, as multiple users take the image-based word test, the threshold values ​​for frequency or frequency may be dynamically adjusted as the total number of user responses for a single image increases. For example, as the total number of user responses increases, the threshold value for frequency can be raised to limit the number of words that can be considered "the correct answer decided by everyone," or the threshold value for frequency can be lowered because responses may be distributed among multiple words, potentially lowering the frequency of each word.

[0071] Furthermore, under the conditions mentioned above, the words with the highest frequency or count are determined to be the "correct answer chosen by everyone," but it is also acceptable to choose a word that is not among the top words to be determined to be the "correct answer chosen by everyone."

[0072] For example, by setting at least one keyword related to an image, a user's response (word) to an image word test, or a translation of that word, can be determined as the "commonly chosen correct answer" if it matches (is included in) the keyword. In this case, a word that does not appear high in frequency or count may be determined as the "commonly chosen correct answer."

[0073] Furthermore, in addition to corresponding keywords with images, words can also be designated as "the correct answer chosen by everyone" if their appearance rate or total count in the answer ranking is below a predetermined threshold. In other words, words that are related to images and are rare or uncommon in user responses can be designated as "the correct answer chosen by everyone." For example, by setting a condition that the appearance rate must be 1% or less, words that are rare in user responses can be designated as "the correct answer chosen by everyone."

[0074] This allows users to learn rare words associated with images by clearly indicating the "correct answer chosen by everyone" when displaying the answer ranking in the information processing device 10 during the image word test, thus enabling a learning effect (see Figure 14).

[0075] Thus, in the image vocabulary test conducted in the system of this embodiment, not only the word designated as the "manufacturer's correct answer" can be used, but the word determined as the "commonly decided correct answer" can also be determined based on the aggregated words that multiple users actually answered to the images when they took the image vocabulary test. In other words, by deliberately introducing "ambiguity" into the word designated as the correct answer related to the image in the image vocabulary test, the perspective of the user taking the image vocabulary test can be broadened, and effective vocabulary learning can be expected.

[0076] Next, the operation of the system in this embodiment will be described.

[0077] Figure 6 is a flowchart showing the image word test processing of the information processing device 10 in this embodiment. Figure 7 is a flowchart showing the image word test provision processing of the server 30 in this embodiment. Figures 8 and 9 are flowcharts showing the image word test data generation processing of the server 30 in this embodiment.

[0078] First, the image word test processing of the information processing device 10 and the image word test provision processing of the server 30 will be explained with reference to the flowcharts shown in Figures 6 and 7.

[0079] When a user of the information processing device 10 performs an image word test provided by the server 30, they start the image word test program 12b to begin the image word test process.

[0080] The CPU 11 of the information processing device 10 starts the image word test process and accesses the server 30 via the communication unit 15 (Figure 6, step A1). Here, the CPU 11 prompts the user to input user identification data (e.g., user ID and password) and sends it to the server 30. The CPU 11 also sends information to the server 30 via the communication unit 15, along with the user identification data, indicating that the image word test process has started in the information processing device 10.

[0081] When the CPU 31 of the server 30 receives an access request from the information processing device 10 via the communication unit 35 (Figure 7, step B1), it identifies the user (or the information processing device 10) based on the user identification data transmitted from the information processing device 10 and information indicating that the image word test processing has started in the information processing device 10, and then starts the image word test management program 32b based on the information indicating that the image word test processing has started in the information processing device 10 (step B2).

[0082] The CPU 31 selects an image word test to provide to the user (information processing device 10) based on the user identified using the user identification data (step B3). For example, the CPU 31 refers to user management data 32h based on the user identification data received from the information processing device 10, and obtains, for example, a level corresponding to the user identification data (e.g., vocabulary level), and selects answer ranking data 32d2 (ranking list) corresponding to this level. Alternatively, the CPU 31 can also refer to user management data 32h based on the user identification data, and obtain, for example, user attribute data (country (area), age, gender, etc.) corresponding to the user identification data, and select answer ranking data 32d2 (ranking list) corresponding to this attribute data.

[0083] As a result, the answer ranking data 32d2 corresponding to a single image (image data 32d1) differs depending on the user's level or attributes. Therefore, the word designated as the "correct answer chosen by everyone" in the correct answer data 32d3 determined based on the answer ranking data 32d2 also differs depending on the user's level and attributes. In other words, by using user-specific answer ranking data 32d2 in an image word test, it becomes possible to conduct an image word test using an answer ranking compiled from the answers of other users with the same level and attributes, and the word designated as the "correct answer chosen by everyone" based on that answer ranking.

[0084] In the explanation above, users are identified based on user identification data, but it is also possible to conduct the image word test on server 30 according to predetermined rules without identifying the user. Furthermore, in addition to providing image word test data according to the user's level, it is also possible to simply provide image word test data that the user has not previously taken.

[0085] The CPU 31 sends the image word test data 32d, which is the questions for the image word test, to the information processing device 10 (step B4). For example, the CPU 31 may acquire 10 sets of image word test data 32d and send them all to the information processing device 10 at once.

[0086] Meanwhile, when the CPU 11 of the information processing device 10 receives image word test data 32d from the server 30, it stores it in memory 12 (image word test data 12c) and starts an image word test based on the image word test data 12c (Figure 6, step A3).

[0087] First, the CPU 11 displays the image word test data 12c on the display unit 17.

[0088] Figure 10 shows an example of the screen displayed when a question is presented in this embodiment of the image word test. The screen displayed when a question is presented includes an image area 60 for displaying the image to be used as the question for the image word test, an input area 61 (voice input button 62) for entering the answer word, an answer button 64 for indicating completion of the answer, and a skip button 65 for indicating the transition to the next question (image display).

[0089] CPU 11 displays the image to be tested in image area 60 based on image data 12c1 of the first image word test data 12c (Step A4). At the same time as the image display begins, CPU 11 also starts measuring the response time (Step A5).

[0090] If an operation on the skip button 65 is detected, the CPU 11 similarly displays the image for the next question in the image area 60 based on the image data 12c1 of the image word test data 12c, and starts measuring the response time.

[0091] The CPU 11 enters a state of waiting for user input of a word (answer) into the input area 61. Based on the data input from the input unit 16 or the voice input unit 18, the CPU 11 obtains a word (answer) for the image (step A6).

[0092] For user operations to input words into the input area 61, the following methods can be used: (1) first input method, (2) second input method, and (3) third input method. For example, (1) first input method is character input using a software keyboard, (2) second input method is character input using voice input, and (3) third input method is an input method that combines character input using a software keyboard and voice input.

[0093] (1) In the first input method, for example, when a touch operation on the input area 61 is detected, the CPU 11 displays a software keyboard for character input that allows word input on the display unit 17. The CPU 11 displays the characters (words) entered by the touch operation on the software keyboard in the input area 61.

[0094] (2) In the second input method, for example, when a touch operation on the voice input button 62 is detected, the CPU 11 enters a voice input state from the voice input unit 18. When voice (voice reading a word) is input from the voice input unit 18, the CPU 11 performs voice recognition processing based on a voice recognition program on the input voice to detect the spoken word. The CPU 11 displays the word detected by voice recognition in the input area 61 as the word entered by the user operation.

[0095] (3) In the third input method, for example, after inputting a word by operating a software keyboard as in the first input method described above, the word is input by voice as in the second input method described above. Specifically, for example, after inputting "w", "o", "l", and "f" using the software keyboard, the word "wolf" is input by voice. Specifically, in cases like "cake shop" or "look forward to," which consist of multiple words, the user is first prompted to input "cake" or "look" using the software keyboard, and then the remaining words are input via voice.

[0096] Once the user has completed the input operation of a word (answer) for the image, they touch the answer button 64. When the CPU 11 detects the operation by the answer button 64, it stops measuring the response time and performs a correct / incorrect judgment on the word (answer) entered in the input area 61 based on the correct answer data 12c3 (step A7).

[0097] The CPU 11 stores the answer data 12d, which represents the word entered as the answer, in memory 12. It also adds answer time data, which is the result of measuring the response time, to the answer data 12d.

[0098] Furthermore, if the response is given using the second input method described above, the CPU 11 may, in addition to determining whether the word (response) detected by the speech recognition process is correct or incorrect, also determine the pronunciation quality based on an indicator that shows the certainty of the sound in which the word was spoken, i.e., whether it is pronounced correctly. Data regarding the indicator (e.g., phonetic symbols) that shows the correctness of the pronunciation of the detected word (response) is obtained, for example, from phonetic symbol data that shows information on the phonetic symbols corresponding to the words (headwords) registered in the online dictionary function 40a of the educational server 40, via the communication unit 35 of the information processing device 10. The CPU 11 then determines the pronunciation quality by comparing the phonetic symbols of the word (response) detected by the speech recognition process with the phonetic symbol data obtained from the educational server 40. This determination result may be in three stages, for example, A to C, where A is a high evaluation and C is a low evaluation. The determination result is not limited to the above, and there are no particular restrictions as long as it can determine the quality of the pronunciation.

[0099] Alternatively, if the response is given using the third input method described above, the CPU 11 determines the pronunciation quality by comparing the phonetic symbols corresponding to a phrase consisting of multiple words, which are a combination of data representing phonetic symbols corresponding to a word (text) entered via the software keyboard and data representing phonetic symbols indicated by the audio data of the word entered via voice, with the phonetic symbols corresponding to that phrase obtained from the educational server 40 or the like.

[0100] Alternatively, if the response is given using the third input method described above, the CPU 11 obtains data representing phonetic symbols corresponding to the word (text) entered via the software keyboard and data representing phonetic symbols corresponding to the voice-inputted word from the educational server 40, etc., and compares them to determine the pronunciation quality. Furthermore, the method for determining pronunciation quality is not limited to the above, and other methods can be used.

[0101] If a response is received using the second or third input method, the CPU 11 adds pronunciation quality data, which indicates the pronunciation quality determined as described above, to the response data 12d. The pronunciation quality data and response time data added to the response data 12d are managed by the server 30 as user data referenced to classify the user's level.

[0102] In determining correctness, CPU11 checks whether the word entered as the answer to the image matches the word designated as the "manufacturer's correct answer" or the "commonly chosen correct answer" in the correct answer data 12c3. If the two match, CPU11 determines the entered word to be correct. If the word entered as the answer to the image does not match either the "manufacturer's designated correct answer" or the "commonly chosen correct answer," CPU11 determines it to be incorrect.

[0103] As explained above, the system uses user identification data to reference user management data 32h and selects response ranking data 32d2 according to the user level (e.g., vocabulary level) or user attributes (e.g., country, age) corresponding to the user identification data. However, it is also possible to select response ranking data 32d2 (ranking list) according to the time taken to answer (answer time slot) or pronunciation quality, which are determined when answering the image word test. Regarding pronunciation quality, this is effective when the answer entered in step A6 is voice input (second input method, third input method). If the ranking list is switched, the CPU 11 may display text or other information on the display unit 17 so that the user can recognize that a switch has occurred. For example, in a part of the answer ranking area 75 in Figure 11(A) (for example, the upper right corner), text or other information indicating the type of ranking list by time of response by other users (e.g., "Answer time 0 seconds or more to less than 2 seconds") may be displayed. The position and content of the string can be determined arbitrarily.

[0104] Similarly, for pronunciation quality, a string of characters indicating the quality of pronunciation (e.g., pronunciation quality level A) will be displayed in a part of the answer ranking area 75 (for example, the upper right corner). The levels indicating the quality of pronunciation can be defined in advance using numbers, letters, etc.

[0105] When the CPU 11 finishes determining correctness, it displays the screen for determining the correctness of the image word test on the display unit 17 (step A8).

[0106] Figure 11(A) shows an example of a screen for determining correctness. The screen for determining correctness includes an image area 70 for displaying the image used as the question for the image word test, an answer display area 71 for displaying the word entered as the answer, a judgment result area 74 for showing the result of the correctness determination ("correct" or "incorrect"), an answer ranking area 75 for displaying the answer ranking corresponding to the image (question) displayed in the image area 70, a stop button 76 for instructing the user to stop the image word test, and a next question button 78 for instructing the user to move on to the next question (image display).

[0107] In the answer ranking area 75, the words ranked 1st to 5th (by frequency or number of occurrences) and their translations are displayed as a compilation of words entered by other users as answers to the image displayed in image area 70. In addition, the words displayed in the answer ranking area 75 are accompanied by correct answer indicators C1 to indicate "manufacturer-designated correct answer," G1 and G2 to indicate "everyone's chosen correct answer," and B1 and B2 to indicate "incorrect answer," clearly indicating whether each word is a correct or incorrect answer.

[0108] In the example shown in Figure 11(A), five words are displayed on one screen in the answer ranking area 75. However, by scrolling the answer ranking displayed in the answer ranking area 75 in response to user operation via the input unit 15 using a scroll bar (not shown) in the answer ranking area 75, it is possible to display words from the 6th place onwards.

[0109] In the example shown in Figure 11(A), the word "wolf" is entered as the answer. Since the word "wolf" is a "manufacturer-specified correct answer," it is judged as correct, and "Correct" is displayed in the judgment result area 74.

[0110] On the screen displaying the correct / incorrect results, users can not only check the correctness of the word they entered as their answer, but also refer to multiple words displayed in the answer ranking, that is, words that other users answered as being related to the image. Therefore, for a single image displayed in image area 70, users can learn not only the word they answered, but also multiple words answered by other users, leading to improved learning efficiency.

[0111] Furthermore, the answer ranking allows users to see what words other users thought of in response to the images, which can pique their interest and lead to increased motivation to learn.

[0112] Furthermore, instead of having just one correct word for each image, there are "manufacturer-designated correct answers" and "community-determined correct answers" (each of which may have multiple options), introducing "ambiguity" into the correct answers and broadening the user's perspective. Therefore, compared to word learning where predetermined words are used as the correct answer for each image, this method allows for more effective and efficient word learning.

[0113] If the CPU 11 detects a user action on the next question button 78 (step A10, Yes), it processes the next question of the next image word test based on the image word test data 12c, in the same manner as described above (steps A4-A8).

[0114] Furthermore, if the CPU 11 detects a user action on the cancel button 76 (step A11, Yes), or if the processing of all problems has been completed (step A9, Yes), it sends the response data 12d for the image word test data 12c received from the server 30 to the server 30 and terminates the image word test processing (step A12).

[0115] Meanwhile, when the server 30 receives the response data from the information processing device 10 (step B5), it registers it in the response database 32e (step B6).

[0116] Furthermore, the pronunciation quality data and response time data attached to the response data are recorded in the user management data 32h in association with the user identification data of the user (information processing device 10) that is the source of the response data.

[0117] In this way, the server 30 can collect the words entered as answers to the images provided as questions for the image word test in conjunction with the execution of the image word test by the information processing device 10. These collected words are used in the image word test data generation process described later.

[0118] Here, we will explain an example of the screen showing the correct / incorrect result for other image-based word tests (images).

[0119] Figure 11(B) shows an example where the user entered the word "neigh" for an image displayed in image area 70. The word "neigh" does not match "horse," which is the number one ranked word in the answer ranking ("Manufacturer's designated correct answer"), but it matches the word that is the second-ranked word ("Everyone's chosen correct answer"), and is therefore judged as correct.

[0120] Figure 12(A) shows an example where the user entered the word "palm" in response to an image displayed in image area 70. The word "palm" is judged as incorrect because it does not match either the "manufacturer-designated correct answer" or the "commonly chosen correct answer" (ranked 4th). However, by referring to the answer ranking displayed in answer ranking area 75, users can learn words effectively by referring to the words that other users have entered.

[0121] Figure 12(B) shows an example where the user entered the word "grand" in response to an image displayed in image area 70. The word "grand" matches the 5th ranked word in the answer ranking, but it is judged as correct because the 5th ranked word is designated as the "manufacturer's correct answer." In the case of the image shown in Figure 12(B), the answers are distributed among multiple words different from the "manufacturer's designated correct answer," meaning that multiple words were associated by the user, and therefore the words ranked 1st to 3rd are designated as the "correct answer chosen by everyone." This allows for effective vocabulary learning by referring to multiple words that differ from the "manufacturer's designated correct answer" answered by other users.

[0122] Figure 13(A) shows an example where the user enters the word "birthday" in response to multiple images displayed in image area 70. The image word test shown in Figure 13(A) is an example where the user is asked to enter one word as an answer based on associations with multiple images. The word "birthday" does not match "pastry chef," which is the third-ranked word in the answer ranking and corresponds to the "manufacturer-specified correct answer," but it matches the word that corresponds to the first-ranked word, "the correct answer chosen by everyone," and is therefore judged as correct. When displaying the images to be used in the image area 70 of the screen when the image word test question is presented, multiple images may be displayed side by side on one screen, or multiple images may be displayed one by one with a predetermined time interval between them.

[0123] Figure 13(B) shows an example where an image displayed in image area 70 is used as a "symbol," and an example where the user enters the word "river" for the image representing the "symbol." The word "river" is judged as correct because it matches the word "river" which is the number one ranked answer in the answer ranking, "Manufacturer-specified correct answer."

[0124] The example mentioned above shows a case where the words with the highest frequency or number of occurrences in the answer ranking are designated as the "correct answer chosen by everyone." However, Figure 14 shows an example where a word not among the top words is designated as the "correct answer chosen by everyone." In the example shown in Figure 14, an image representing "a cook peeling potatoes (vegetables)" is displayed in image area 70, and the user enters the word "spiralizer" (a tool for turning vegetables into noodles). The image in this image word test question is associated with broad conceptual keywords related to the image, such as "vegetables," "cooking," "gentleman," and "tool." Because the word "spiralizer" (a tool for turning vegetables into noodles), entered as an answer by another user, is determined to match the keyword "tool," the word "spiralizer" can become a candidate for the "correct answer chosen by everyone."

[0125] Furthermore, the word "spiralizer" appears in the answer ranking, for example, as the 98th most frequent answer with a frequency of 0.8%. This meets the condition pre-set by CPU 11 to designate answers with a frequency of 1% or less as rare "commonly chosen correct answers," and is therefore considered a rare "commonly chosen correct answer." In this case, a correct answer indicator R1, which indicates a "rare correct answer" and has a different display format than the correct answer indicators G1 and G2 that explicitly indicate "commonly chosen correct answers," is added to clearly indicate that the word is a rare correct answer. In addition, a "rare mark" 90 is added near the judgment result area 74 to clearly indicate that the correct answer was given by inputting a "rare correct answer." Note that the frequency value corresponding to a rare "commonly chosen correct answer" among the conditions pre-set by CPU 11 can be set to any value.

[0126] In this way, by not only designating the words with the highest frequency or total count in the answer ranking as "the correct answer chosen by everyone," but also by designating answers that meet certain conditions, even for words that are not among the top in frequency or total count, as "the correct answer chosen by everyone," a wide range of words related to images can be included as learning targets, thus enabling effective and efficient vocabulary learning.

[0127] Furthermore, if the condition is to consider only words with a high frequency of appearance or high count in the answer ranking as correct, rare words related to the image (words with few responses) may be clearly indicated as "incorrect but rare answers" rather than "correct answers chosen by everyone." Also, as shown in the example of incorrect answer display 91 in Figure 14, the display in the judgment result area 74 will be "incorrect," but a "rare mark" 90 will be added to clearly indicate that a word that is an "incorrect but rare answer" has been entered. As a result, as described above, these words will be included in the ranking as words related to the image, so effective and efficient word learning can be expected for the user.

[0128] The images displayed in image area 70 (image area 60) may include illustrations, photographs, shapes, text, still images, moving images, animations, etc.

[0129] Furthermore, the words used in the answer can be entered as single words, or as a single unit consisting of a sentence or idiom (phrase, set phrase, etc.) containing multiple words.

[0130] Furthermore, in the image word test of this embodiment, since multiple answers are set, such as "manufacturer-specified correct answer" and "commonly decided correct answer," it may also be possible to input multiple words as answers.

[0131] In the explanation above, for example, the server 30 sends a batch of image word test data 32d for 10 questions to the information processing device 10, and the information processing device 10 determines whether the input answers are correct or incorrect based on the correct answer data 12c3 contained in the image word test data 12c while the received image word test is being conducted. However, it is also possible for the server 30 to perform the correctness determination.

[0132] In this case, the server 30 sends only image data 32d1 representing the image in question to the information processing device 10. Based on the image data 32d1 from the server 30, the information processing device 10 displays the screen for the image word test (Figure 10) and prompts the user to input the answer word.

[0133] The information processing device 10 sends response data indicating the entered word to the server 30. If the word to be answered is entered via voice input, the voice data is also sent to the server 30.

[0134] The server 30 performs a correct / incorrect judgment on the response data (voice data) transmitted from the information processing device 10 in the same manner as described above, and sends the correct / incorrect judgment result and response ranking data 32d2 showing the response ranking to the information processing device 10, and displays the correct / incorrect judgment result screen in the same manner as described above.

[0135] This allows the server 30 to perform the main processing, thereby reducing the processing load on the information processing device 10.

[0136] Next, the image word test data generation process by the server 30 of this embodiment will be described.

[0137] First, we will explain the image word test data generation process (1) with reference to the flowchart shown in Figure 8. The image word test data generation process (1) generates image word test data based on response data collected from image word tests conducted by multiple users.

[0138] The CPU 31 selects one image word test data 32d from among multiple image word test data 32d stored in memory 32, obtains the image data 32d1 (the image in question) associated with the selected image word data 32d (step C1), and aggregates the answer data corresponding to this image data 32d1 (step C2). Specifically, the CPU 31 aggregates the multiple answer data (words) corresponding to the image data 32d1 obtained in step C1 in the answer database 32e, for each answer (word).

[0139] The CPU 31 obtains a representative translation for each of the collected response data (words) by searching, for example, dictionary data (not shown), and adds it to the response (word), thereby creating ranking word data that represents the words and translations displayed in the response ranking, and updating the response ranking data 32d2 (step C3). For example, for the first-place response displayed in the response ranking area 75 shown in Figure 11(A), the translation "Ōkami" is added to the word "wolf" to create ranking word data, and the response ranking data 32d2 is updated.

[0140] Next, CPU31 performs a correct / incorrect judgment on each of the answer data (words) included in the ranking word data. That is, it determines whether words other than those designated as "manufacturer-specified correct answers" are "correct answers chosen by everyone" or "incorrect answers." Here, as mentioned above, based on the aggregated content of each answer data (word) included in the ranking word data, for example, it determines whether the occurrence rate or aggregated number meets predetermined conditions, and determines whether it is "correct answer chosen by everyone" or "incorrect" (step C4).

[0141] CPU31 updates the correct answer data 32d3 to reflect the result of the judgment for each word, so that the answer data (word) that was determined to be the "correct answer decided by everyone" in step C4 is considered the correct answer (step C5).

[0142] Similarly, the CPU 31 obtains the image data 32d1 (the image in question) associated with the next image word test data 32d, and executes the processing of steps C2 to C5 described above.

[0143] Once processing is complete for all image word test data 32d to be processed, the CPU 31 terminates the image word test data generation process (1) (Step C6, Yes).

[0144] The answer ranking data 32d2 is modified in accordance with fluctuations in the number of answer data collected as multiple users conduct the image word test. Therefore, since the "correct answer" or "incorrect answer" determined by everyone may change due to changes in the answer ranking, it is desirable to perform the image word test data generation process (1) for example, every certain period of time, when the change in the number of answer data collected exceeds a predetermined standard, and also at an appropriate time in accordance with instructions from the server 30 administrator.

[0145] Furthermore, the aggregation of response data (Step C2) can be performed for multiple user classifications. For example, it can be performed for user attributes (country (area), age, gender, language used, etc.), for user vocabulary level as indicated by user data, for the time taken to answer during the image word test (e.g., 2 seconds), or for the voice quality of the voice input (pronunciation) when the image word test was performed.

[0146] This allows for the generation of multiple categorized answer ranking data 32d2 (answer rankings) for users, enabling correct / incorrect judgments based on the answer rankings corresponding to the user's attributes and level when running the image word test.

[0147] In this way, by generating image word test data through the image word test data generation process (1), it is possible to conduct an image word test based on response data (response ranking) collected as multiple users take the image word test.

[0148] Next, the image word test data generation process (2) will be explained with reference to the flowchart shown in Figure 9. The image word test data generation process (2) generates new image word test data based on word data acquired from an external device. Examples of external devices include an educational server 40 (with an online dictionary search function 40a) or an electronic dictionary 50 connected via the network 8.

[0149] The educational server 40 or electronic dictionary 50 stores the string (word) that has been searched using dictionary search via user operation, and sends it to server 30, for example, when connecting to server 30.

[0150] The CPU 31 of server 30 records the string (word) sent from the educational server 40 or electronic dictionary 50 as the search target for dictionary search as collected word data 32f (step D1). Note that the string (word) here is not recorded (registered) in the collected word data 32f. In addition, the collected word data 32f stores the string (word) in association with transmission count data indicating the number of times the same string (word) has been sent from the educational server 40 or electronic dictionary 50. Hereafter, in the explanation of the flowchart in Figure 9, the above string (word) will be referred to as the search target word data.

[0151] The CPU 31 starts generating image word test data at a predetermined timing. For example, if, among the multiple search target word data newly stored as collected word data 32f, there is a search target word data for which the transmission count data recorded in association with that search target word data exceeds a predetermined threshold value, the CPU 31 determines to generate new image word test data. In other words, search target word data for which the transmission count data exceeds the threshold value is a word that has been frequently searched by multiple users using the education server 40 or electronic dictionary 50, and effective word learning can be expected by using this word as the target of word learning (image word test).

[0152] Here, it is not necessary to target only words (headwords) registered in the dictionary data of the online dictionary search function 40a or the electronic dictionary 50. For example, new words and coined words not registered in the dictionary data can be targeted. Even if a new word or coined word is not registered in the dictionary data, if multiple users have attempted to search for that word (new word or coined word), the number of submissions will exceed the threshold value and it may be included in the image word test data.

[0153] If CPU31 determines that it is necessary to generate new image word test data (Step D2, Yes), it retrieves any search target word data from among the multiple search target word data newly stored in the collected word data 32f (Step D3), and determines the image corresponding to this word data (Step D4).

[0154] In step D4, for example, the CPU 31 obtains representative translations of dictionary search word data from the online dictionary function 40a of the educational server 40 or from an electronic dictionary 50, etc. Then, among the multiple image data 32g recorded in memory 32, if there is an image that corresponds to (associates) the obtained translation for at least one keyword associated with the image data 32g, the CPU 31 determines that the image is the image that corresponds to (associates) the search target word data for the translation.

[0155] The CPU 31 creates image word test data (step D5) using representative translations of dictionary search word data as correct answer data 32d3, and images corresponding to (associated with) the dictionary search word data determined in step D4 as image data 32d1, which are used as questions for the image word test, and saves this data to memory 32 (step D6).

[0156] Similarly, the CPU 31 can create image word test data based on word data collected from an external device (educational server 40 or electronic dictionary 50) (steps D3 to D6). When the CPU 31 runs out of collected word data 32f to process (step D7, Yes), it terminates the image word test data generation process (2).

[0157] In this way, the image word test data generation process (2) creates image word test questions (image word test data) based on word data collected from an external device (educational server 40 or electronic dictionary 50) that is the target of dictionary searches, making it possible to conduct image word tests on words that are expected to be effective for word learning for multiple users.

[0158] While the explanation above focuses on English vocabulary, it is also possible to target other languages ​​(such as Chinese, Korean, or Arabic).

[0159] Alternatively, the information processing device 10 may obtain aggregate data for each word in the answer ranking along with the answer ranking data from the server 30, and based on the aggregate data and predetermined conditions (number of words counted, frequency of appearance, rarity, etc.), determine the "correct answer chosen by everyone" (third word), and perform a correct / incorrect judgment on the word answered by the user.

[0160] Furthermore, the methods described in the embodiments, i.e., each method such as the processing shown in the flowchart, can be stored and distributed as a program that can be executed by a computer on a recording medium such as a memory card (ROM card, RAM card, etc.), magnetic disk (flexible disk, hard disk, etc.), optical disk (CD-ROM, DVD, etc.), or semiconductor memory. The computer can then read the program recorded on the external recording medium, and its operation will be controlled by this program, thereby realizing the same processing as described in the embodiments.

[0161] Furthermore, the program data required to implement each method can be transmitted over a network (the internet) in the form of program code, and the same functions as those of the embodiments described above can be obtained from computers connected to this network.

[0162] Furthermore, the present invention is not limited to the embodiments, and can be modified in various ways during implementation without departing from its essence. Moreover, the embodiments include inventions at various stages, and various inventions can be extracted by appropriate combinations of the multiple constituent elements disclosed. For example, even if some constituent elements are deleted from all the constituent elements shown in the embodiments, or if some constituent elements are combined, if the problem described in the section on the problem the invention aims to solve is solved and the effect described in the section on the effect of the invention is obtained, then the configuration in which these constituent elements are deleted or combined can be extracted as an invention.

[0163] The invention described in the original claims of this application is listed below.

[0164] [1] Computers The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. A method for determining whether a second word is correct or incorrect, based on the first word and a third word determined based on the aggregated results of words answered by multiple other users in response to the image.

[0165] [2] Among the multiple words that multiple other users have responded to the image, the word whose frequency of occurrence satisfies a predetermined condition is determined to be the third word. The method for determining correctness of the user's answer as correct, as described in [1], wherein the second word is the same as the first word or the third word.

[0166] [3] Obtain ranking data of the words that multiple other users have answered in response to the image, The correctness determination method according to [1] or [2], which displays a ranking list generated based on the aggregated results for each word answered by multiple other users, which are included in the answer ranking data, and the result of the correctness determination for the second word on the display unit.

[0167] [4] The method for determining correctness according to [3], wherein the third word and information indicating that it has been determined to be the third word are displayed side by side on the display unit from among the words included in the ranking list.

[0168] [5] The aforementioned response ranking data includes the aforementioned ranking list according to the time taken for multiple other users to respond to the aforementioned image, The time from displaying the image on the display unit until obtaining the second word, which is the user's response to the image, is measured. The method for determining correctness according to [3] or [4], wherein the ranking list selected according to the measured time and the result of the correctness determination for the second word are displayed on the display unit.

[0169] [6] The aforementioned response ranking data includes the aforementioned ranking list according to the pronunciation quality of responses voice-input by multiple other users for the aforementioned image, When the user inputs the second word by voice in relation to the image, the pronunciation quality of the input voice is determined. The method for determining correctness according to [3] or [4], wherein the ranking list selected according to the determined pronunciation quality and the result of the correctness determination for the second word are displayed on the display unit.

[0170] [7] Computers The image and a first word associated with the image are provided to multiple information processing devices. Obtain a second word, which is the answer to the image, from multiple information processing devices. A method for determining correctness, which determines a third word, determined based on the aggregated results of the second word obtained from the plurality of information processing devices, as the correct answer to the image.

[0171] [8] The method for determining correctness according to [7], wherein, among multiple second words obtained from multiple information processing devices, a word whose occurrence rate of the same second word satisfies a predetermined condition is determined as the third word.

[0172] [9] Equipped with a control unit, The control unit, The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. An information processing device that performs a correct / incorrect determination on the second word based on the first word and a third word determined based on the aggregated results of words answered by multiple other users in response to the image.

[0173]

[10] On the computer, Obtain the image and the first word associated with the image, The image is displayed on the display unit, and a second word is obtained as the user's response to the image. The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. A program that performs a correct / incorrect determination on the second word based on the first word and a third word determined based on the aggregated results of words answered by multiple other users in response to the image.

[0174]

[11] On the computer, The image and a first word associated with the image are provided to multiple information processing devices. Obtain a second word, which is the answer to the image, from multiple information processing devices. A program that causes a third word determined based on the aggregated results of the second word obtained from the plurality of information processing devices to be determined as the correct answer to the image. [Explanation of Symbols]

[0175] 8...Network, 10...Information processing device, 11,31...CPU, 12,32...Memory, 12a,32a...Basic program, 12b...Image word test program, 12c,32d...Image word test data, 12c1,32d1...Image data, 12c2,32d2...Answer ranking data, 12c3,32d3...Correct answer data, 12d...Answer data, 12e...User identification data, 13,33...Recording medium, 14,34...Recording medium reading unit, 15,35...Communication unit, 16,36...Input unit, 17,37...Display unit, 18...Voice input unit, 19...Voice output unit, 30...Server, 32b...Image word test management program, 32c...Image word test data generation program, 32e...Answer database, 32f...Collected word data, 32g...Image data, 32h...User management data 3, 40...Educational server, 50...Electronic dictionary.

Claims

1. Computers The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. Based on the first word and the third word determined based on the aggregated results of words answered by multiple other users in response to the image, a correct / incorrect judgment is performed on the second word. Obtain ranking data of the words that multiple other users have answered in response to the image, The aforementioned answer ranking data is generated based on the aggregated results for each word answered by multiple other users, and includes a ranking list according to the pronunciation quality of the answers voice-input by multiple other users for the image. When the user inputs the second word by voice in relation to the image, the pronunciation quality of the input voice is determined. The ranking list selected according to the determined pronunciation quality and the result of the correct / incorrect judgment for the second word are displayed on the display unit. Method for determining correctness.

2. Among the multiple words that multiple other users have responded to the image, the word whose frequency of occurrence satisfies a predetermined condition is determined to be the third word. If the second word is the same as the first word or the third word, the user's answer is determined to be correct. The method for determining correctness according to claim 1.

3. The ranking list and the result of the correct / incorrect judgment for the second word are displayed on the display unit. The method for determining correctness according to claim 1 or claim 2.

4. The third word from the ranking list and information indicating that it has been determined to be the third word are displayed side by side on the display unit. The method for determining correctness according to claim 3.

5. The aforementioned response ranking data includes the aforementioned ranking list according to the time taken for multiple other users to respond to the aforementioned image, The time from displaying the image on the display unit until obtaining the second word, which is the user's response to the image, is measured. The ranking list selected according to the measured time and the result of the correct / incorrect judgment for the second word are displayed on the display unit. The method for determining correctness according to claim 3 or claim 4.

6. Computers The image and a first word related to the image are provided to multiple information processing devices. A second word, which is the answer to the image, is obtained from the aforementioned plurality of information processing devices. The third word determined based on the aggregated results of the second word obtained from the aforementioned multiple information processing devices is determined to be the correct answer to the image. The ranking data of the words answered to the images acquired from the aforementioned multiple information processing devices is obtained, The aforementioned response ranking data is generated based on the aggregated results for each word answered in each of the multiple information processing devices, and includes a ranking list according to the pronunciation quality of the responses voice-input for the image in the multiple information processing devices. When the second word is input to the image via voice into the information processing device, the pronunciation quality of the input voice is determined. The ranking list selected according to the determined pronunciation quality and the result of the correct / incorrect judgment for the second word are displayed on the display unit of the information processing device. Method for determining correctness.

7. Among multiple second words obtained from multiple information processing devices, the word whose occurrence rate of the same second word satisfies a predetermined condition is determined to be the third word. The method for determining correctness according to claim 6.

8. Equipped with a control unit, The control unit, The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. Based on the first word and the third word determined based on the aggregated results of words answered by multiple other users in response to the image, a correct / incorrect judgment is performed on the second word. Obtain ranking data of the words that multiple other users have answered in response to the image, The aforementioned answer ranking data is generated based on the aggregated results for each word answered by multiple other users, and includes a ranking list according to the pronunciation quality of the answers voice-input by multiple other users for the image. When the user inputs the second word by voice in relation to the image, the pronunciation quality of the input voice is determined. The ranking list selected according to the determined pronunciation quality and the result of the correct / incorrect judgment for the second word are displayed on the display unit. Information processing device.

9. Computers, Obtain the image and the first word associated with the image, The image is displayed on the display unit, and a second word is obtained as the user's response to the image. The system obtains an image, a first word related to the image, and a second word which is the user's response to the image displayed on the display unit. Based on the first word and the third word determined based on the aggregated results of words answered by multiple other users in response to the image, a correct / incorrect judgment is performed on the second word. Obtain ranking data of the words that multiple other users have answered in response to the image, The aforementioned answer ranking data is generated based on the aggregated results for each word answered by multiple other users, and includes a ranking list according to the pronunciation quality of the answers voice-input by multiple other users for the image. When the user inputs the second word by voice in relation to the image, the pronunciation quality of the input voice is determined. The process involves displaying the ranking list selected according to the determined pronunciation quality, and the result of the correct / incorrect judgment for the second word, on the display unit. A program designed to make it work in that way.

10. On the computer, The image and a first word related to the image are provided to multiple information processing devices. Obtain a second word, which is the answer to the image, from multiple information processing devices. The third word determined based on the aggregated results of the second word obtained from the plurality of information processing devices is determined to be the correct answer to the image. The ranking data of the words answered to the images acquired from the aforementioned multiple information processing devices is obtained, The aforementioned response ranking data is generated based on the aggregated results for each word answered in each of the multiple information processing devices, and includes a ranking list according to the pronunciation quality of the responses voice-input for the image in the multiple information processing devices. When the second word is input to the image via voice into the information processing device, the pronunciation quality of the input voice is determined. The ranking list selected according to the determined pronunciation quality and the result of the correct / incorrect judgment for the second word are displayed on the display unit of the information processing device. A program designed to make it work in that way.

Images