Scoring device, scoring method, and scoring program

Abstract

This allows for scoring based on multiple scoring criteria tailored to the learning objectives. [Solution] The scoring device 100 takes information about an English task and information about the answer to the task as input and generates one or more scoring results based on predetermined scoring criteria for the answer to the English task, based on a learning model that outputs scoring results based on predetermined scoring criteria. The scoring device 100 outputs one or more of the generated scoring results based on predetermined scoring criteria to the user.

Description

Technical Field

[0001] The present invention relates to a scoring device, a scoring method, and a scoring program.

Background Art

[0002] In recent years, in addition to school education, a method of learning English using English learning texts, applications, etc. according to the user's own English level has been known. For example, there is a conventional technique that presents English problems to a user via an application and scores the user's answers (see, for example, Non-Patent Document 1).

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, the prior art has a problem that it is difficult to perform scoring based on a plurality of scoring viewpoints according to the purpose of learning. For example, the prior art is a technique that automatically scores a user's answer to a presented problem and provides feedback to the user, but it is difficult for the user himself / herself to select and combine desired scoring viewpoints for feedback.

Means for Solving the Problems

[0005] Therefore, in order to solve the above-mentioned problems and achieve the objective, the scoring device of the present invention is characterized by comprising: a generation unit that takes information about an English task and information about the answer to said task as input and outputs a scoring result based on predetermined scoring criteria for the answer to said English task based on a learning model, and generates one or more scoring results based on predetermined scoring criteria; and an output unit that outputs one or more scoring results based on predetermined scoring criteria generated by the generation unit to the user. [Effects of the Invention]

[0006] The present invention has the effect of enabling scoring based on multiple scoring criteria that correspond to the learning objectives. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 is a diagram illustrating the overall process of the scoring device according to this embodiment. [Figure 2] Figure 2 shows the configuration of the scoring device according to this embodiment. [Figure 3] Figure 3 is a table diagram showing an example of problem information according to this embodiment. [Figure 4] Figure 4 shows an example of problem information according to this embodiment. [Figure 5] Figure 5 is a table diagram showing an example of the response information according to this embodiment. [Figure 6] Figure 6 shows an example of the response information according to this embodiment. [Figure 7] Figure 7 is a table diagram showing an example of scoring result information according to this embodiment. [Figure 8] Figure 8 shows an example of the scoring result output screen according to this embodiment. [Figure 9] Figure 9 shows an example of the scoring result output screen according to this embodiment. [Figure 10] Figure 10 shows an example of the scoring result output screen according to this embodiment. [Figure 11]Figure 11 shows an example of the scoring result output screen according to this embodiment. [Figure 12] Figure 12 is a flowchart showing the scoring process according to this embodiment. [Figure 13] Figure 13 shows an example of a computer that performs the scoring process according to this embodiment. [Modes for carrying out the invention]

[0008] The embodiments for carrying out the present invention (hereinafter referred to as "embodiments") will be described below with reference to the drawings. However, the embodiments are not limited to those described below.

[0009] <Overview> Figure 1 is a diagram illustrating the overall processing of the scoring device 100 according to this embodiment. The scoring device 100 shown in Figure 1 is an example of a computer that provides technology for scoring answers to input English tasks based on multiple scoring criteria, which are predetermined based on the task, answer conditions, etc.

[0010] (background) In recent years, users who engage in self-study of English in addition to school education may take language proficiency tests such as the Practical English Proficiency Test (registered trademark) (hereinafter sometimes referred to as "Eiken") to confirm their English ability. It is known that Eiken scores answers primarily based on the following criteria: grammar, structure, content, and vocabulary.

[0011] The above-mentioned "grammar" perspective is a perspective such as whether the user's answer contains grammar or spelling mistakes. Also, the "structure" perspective is a perspective such as whether the answer is structured according to a specified structure such as "introduction, main body, conclusion", and whether "claims and grounds" are appropriate. Also, the "content" perspective is a perspective such as whether an answer that conforms to the intention of the question has been given, that is, whether it conforms to the specified theme, and whether there is consistency, etc., whether an answer (hereinafter sometimes referred to as "answer conditions") has been given according to the specified answer method for the specified intention of the question. Also, the "vocabulary" perspective is a perspective such as whether the answer can be made using words appropriate for the level of taking the English proficiency test, whether the same expression is repeatedly used, whether there are abbreviated expressions, and whether there are more than a predetermined number of vocabulary words.

[0012] And users who wish to take the English proficiency test learn English through various methods such as commuting to a group that provides English learning guidance, doing problem sets, and doing exercises using applications. However, in the above-mentioned related art, it may be difficult to conduct exercises according to the purpose of English learning.

[0013] For example, there is a known related art that presents English tasks to a user and grades the user's answers in order to conduct exercises using an application. The above-mentioned related art is a technology that automatically grades the user's answer to the presented task and feeds back the results to the user, but even when it is desired to grade according to a plurality of grading perspectives according to the English proficiency test, it is difficult to execute grading based on a plurality of grading perspectives.

[0014] (Processing by the scoring device 100) Therefore, the scoring device 100 according to the present embodiment combines a plurality of scoring perspectives selected by the user and scores whether the user's answer is correct or not, whether it meets a predetermined task and answer conditions, and the like.

[0015] Now, let's return to FIG. 1 and explain the scoring process by the scoring device 100. The scoring device 100 outputs a scoring result based on a predetermined scoring perspective for an English task answer, based on a learning model that takes information about an English task and information about the answer to the task as input, and generates one or a combination of scoring results based on a predetermined scoring perspective (FIG. 1(1)).

[0016] For example, the scoring device 100 receives an input of information about an English task and information about the answer to the English task (FIG. 1(1-1)). For example, when a user conducts English learning related to the Eiken test, the scoring device 100 can receive information about scoring perspectives such as "grammar", "composition", "content", "vocabulary", etc. selected by the user.

[0017] Next, the scoring device 100 inputs the received information about the English task and the information about the answer to the English task into the trained learning model (FIG. 1(1-2)). Then, the scoring device 100 generates a scoring result obtained by scoring for each scoring perspective based on the trained learning model (FIG. 1(1-3)).

[0018] The scoring device 100 outputs the generated scoring result based on one or a plurality of predetermined scoring perspectives to the user (FIG. 1(2)). For example, the scoring device 100 can output to a user who conducts English learning, including information such as the score, pointed-out areas, comments on the answer, etc. for each scoring perspective such as "grammar", "composition", "content", "vocabulary", etc. by the user.

[0019] In this way, the scoring device 100 according to the present embodiment is not limited to scoring based on one scoring perspective, but can execute scoring by combining a plurality of scoring perspectives according to the user's selection and set conditions, and output the generated scoring result to the user. That is, the scoring device 100 has the effect of enabling scoring based on a plurality of scoring perspectives according to the purpose of learning.

[0020] <Explanation of the Scoring Device 100> Next, the configuration of the scoring device 100 will be described. Figure 2 is a diagram showing the configuration of the scoring device 100 according to this embodiment. As shown in Figure 2, the scoring device 100 has a communication unit 110, a storage unit 120, and a control unit 130. Although not shown in Figure 2, the scoring device 100 may be equipped with an input unit such as a keyboard or mouse to receive input such as operations from an administrator. The scoring device 100 may also be equipped with a display unit such as a display to show information related to the scoring process set in the scoring device 100 to an administrator.

[0021] (Communications Department 110) The communication unit 110 performs data communication related to the input of information regarding English assignments and responses to those assignments. The communication unit 110 also performs data communication related to the output of information regarding scoring results based on predetermined scoring criteria. The communication unit 110 is implemented using a NIC (Network Interface Card), etc., and controls communication via telecommunication lines such as a LAN (Local Area Network) or the Internet. The communication unit 110 is connected to the network by wired or wireless connection as needed, and can send and receive information bidirectionally with terminal devices 200 etc. operated by the user.

[0022] (Storage unit 120) The memory unit 120 stores data and programs used for various processes by the control unit 130, as well as various data acquired through the operation of the control unit 130. The memory unit 120 is implemented using semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical discs. As shown in Figure 2, the memory unit 120 also includes a task information DB 121, an answer information DB 122, a scoring result information DB 123, and a learning model DB 124.

[0023] (Issue Information DB121) The Task Information DB121 is a database that stores information about tasks (task information) that are presented to the user. Specifically, the Task Information DB121 stores information such as the given task and the conditions for answering it. Here, an example of task information stored in the Task Information DB121 will be explained using a table diagram. Figure 3 is a table diagram showing an example of task information according to this embodiment.

[0024] As shown in Figure 3, the task information DB121 stores the following information, associated with "No," which is information that identifies an individual task: "RANK," which is the difficulty level of learning specified by the user, such as the Eiken grade; "QUESTION"; "TOPIC"; and "POINTS." For example, as shown in Figure 3, the task information DB121 stores "RANK" "2," "QUESTION" "A," "TOPIC" "B," and "POINTS" "C" associated with No "1."

[0025] From here, an example of the aforementioned QUESTION "A", TOPIC "B", and POINTS "C" will be explained using Figure 4. Figure 4 is a diagram showing an example of problem information according to this embodiment.

[0026] The QUESTION contains information about the conditions under which the user should create their answer. For example, QUESTION "A" includes information about the answer conditions such as "Write an essay on the given TOPIC," "Use TWO of the POINTS below to support your answer," "Structure: introduction, main body, and conclusion," and "Suggested length: 120-150 words" (Figure 4 (1)).

[0027] Furthermore, the TOPIC includes information such as the subject and title specified in the assignment. For example, TOPIC "B" includes information such as "Agree or Disagree: High school students should be able to choose to study a second foreign language in addition to English" (Figure 4 (2)).

[0028] Furthermore, POINTS includes information such as examples of perspectives that can be used as reference when writing the content. For example, POINTS "C" includes information such as "Communication," "Foreign tourists," "Globalization," and "Usefulness" (Figure 4 (3)).

[0029] (Answer information DB122) The Answer Information DB122 is a database that stores information (answer information) related to users' responses to English assignments. Specifically, the Answer Information DB122 stores information such as information identifying the user who submitted the answer and the content of the answer. Here, an example of the answer information stored in the Answer Information DB122 will be explained using a table diagram. Figure 5 is a table diagram showing an example of answer information according to this embodiment.

[0030] As shown in Figure 5, the response information DB122 stores the user and the response, associating them with "No," which is information that identifies individual response information. For example, the response information DB122 stores user "D" and response "E" as associated with No "1." The "user" mentioned above stores a unique number, text, etc., to identify the user. The "response" stores information about the user's response result.

[0031] From here, an example of the above-mentioned answer "E" will be explained using Figure 6. Figure 6 is a diagram showing an example of answer information according to this embodiment. For example, as shown in (1) of Figure 6, the answer information DB122 stores text information such as English as answer information, as an answer to a task entered by the user (an example of answer "E").

[0032] The response information DB122 can store text information, such as English sentences, for each input item when storing response information. For example, if the user inputs English sentences for each item in the "Introduction," "Body," and "Conclusion" sections, the response information DB122 can store the English text information for each of the aforementioned items (Figure 6 (1-1) to (1-3)).

[0033] (Scoring results information DB123) The scoring result information DB123 is a database that stores information (scoring result information) regarding the scoring results for user responses generated by the generation unit 132, which will be described later. Specifically, the scoring result information DB123 stores information such as information identifying the user who made the response, the date and time of the response, and the scoring results for each scoring criterion. Here, an example of the response information stored in the scoring result information DB123 will be explained using a table diagram. Figure 7 is a table diagram showing an example of scoring result information according to this embodiment.

[0034] As shown in Figure 7, the scoring result information DB123 stores the user and the scoring results for each scoring criterion, associated with "No," which is information that identifies individual scoring result information. For example, the scoring result information DB123 stores the user "D" and the scoring results "score" and "details" for each scoring criterion "grammar," "structure," "content," and "vocabulary," associated with No "1." Here, we will explain the scoring results "score" and "details" for each scoring criterion.

[0035] For example, the scoring result information DB123 stores a score of "3" and a detail of "F" related to grammar (scoring result). The details related to grammar (scoring result), such as "F," include information such as the location of errors in the user's answer, the corrected text with the errors corrected, and comments on the user's answer.

[0036] For example, the scoring result information DB123 stores a score of "5" and a detail of "G" related to the structure (scoring result). The details related to the structure (scoring result), such as detail "G", include information such as parts of the user's answer that do not have the prescribed structure, the results of the structure check, and comments on the user's answer.

[0037] For example, the scoring result information DB123 stores a score of "5" and a detail of "H" related to the content (scoring result). The details related to the content (scoring result), such as the detail of "H", include information such as the result of checking whether the user's answer is in line with the intent of the question, the result of checking the content, and comments on the user's answer.

[0038] For example, the scoring result information DB123 stores the score "5" and the detail "I" related to the vocabulary (scoring result). The details related to the vocabulary (scoring result), such as detail "I", include information such as errors in the writing method, the results of the vocabulary check (results of checking the word or writing method), and comments on the user's answer.

[0039] (Learning model DB124) The learning model DB124 is a database that stores learning models for the generation unit 132 (described later) to generate scoring results based on predetermined scoring criteria. For example, the learning model DB124 can store generation models such as large-scale language models that generate output results in response to predetermined inputs.

[0040] It should be noted that the embodiment in which the learning model DB124 stores the learning model is merely an example and is not limited thereto. For example, the learning model may be incorporated into the generation unit 132 described later and used to generate scoring results based on scoring criteria.

[0041] (Control unit 130) Now, let's return to Figure 2 and continue the explanation. The control unit 130 has an internal memory for temporarily storing programs and processing data that define various processing procedures of the scoring device 100, and is realized by electronic circuits such as a CPU (Central Processing Unit) and an MPU (Micro Processing Unit), and integrated circuits such as an ASIC (Application Specific Integrated Circuit) and an FPGA (Field Programmable Gate Array). As shown in Figure 2, the control unit 130 has a receiving unit 131, a generation unit 132, and an output unit 133.

[0042] (Reception desk 131) The reception unit 131 receives predetermined information used by the scoring device 100 for scoring processing from the user's terminal device 200 or other information processing devices via the communication unit 110 described above. For example, the reception unit 131 receives information regarding the user's response to an English task, information regarding the task corresponding to that response, information regarding the response conditions, etc.

[0043] (Generation unit 132) The generation unit 132 inputs information about the English task to be input and information about the answer to the English task into a trained learning model to generate scoring results based on predetermined scoring criteria. The generation unit 132 then stores the generated scoring results in the scoring result information DB 123.

[0044] Specifically, the generation unit 132 inputs the following information into the trained learning model: the difficulty level (RANK) and response conditions (QUESTION, TOPIC, and POINTS) of the English learning task stored in the task information DB 121 as information related to the English task, and the responses to the task (answers) entered by the user in natural language sentences stored in the response information DB 122 as information related to the response to the task. The generation unit 132 then generates one or more scoring results based on predetermined scoring criteria selected by the user.

[0045] Here, as an example of the predetermined scoring criteria mentioned above, we will explain the criteria for whether the user's answer contains grammatical or vocabulary errors related to English (hereinafter sometimes simply referred to as the "grammatical criteria"), whether the user's answer has a predetermined structure (hereinafter sometimes simply referred to as the "structure criteria"), whether the user's answer is in line with the predetermined intent of the question (hereinafter sometimes simply referred to as the "content criteria"), and whether the number of words included in the user's answer or the way the words are written meets predetermined conditions (hereinafter sometimes simply referred to as the "vocabulary criteria").

[0046] The generation unit 132 generates scoring results using grammatical aspects as predetermined scoring criteria. Specifically, the generation unit 132 calculates a score based on whether the number of errors in the user's answer matches a predetermined standard category, such as "5" for 0 errors, "4" for 1-2 errors, and so on.

[0047] Furthermore, the generation unit 132 identifies the locations of errors in the user's response based on the learning model. The generation unit 132 also generates correct answers based on pre-prepared responses or correct answers generated based on the prompt text and surrounding context for the identified locations of errors. In addition, the generation unit 132 generates comments to provide feedback to the user based on the number of errors, the number of locations where errors occurred, the types of errors, the types of errors that occurred, and a comparison of the user's response results with those of other users.

[0048] The generation unit 132 generates scoring results using the configuration criteria as predetermined scoring criteria. Specifically, for parts of the user's response that are not answered according to the specified configuration, the generation unit 132 calculates a score based on pre-set criteria, such as "5" if all answers satisfy the specified configuration, and "1" if none do, etc.

[0049] Furthermore, the generation unit 132 identifies parts of the user's response that do not conform to the specified structure, based on the learning model. The generation unit 132 also generates correct answers based on pre-prepared answers or correct answers generated based on the prompt text and surrounding context as a result of the structure check for parts of the user's response that do not conform to the specified structure, based on the number of parts that do not conform to the specified structure, the types of errors, and comparisons with other users' response results, etc., to provide feedback to the user.

[0050] The generation unit 132 generates scoring results using the content criteria as predetermined scoring criteria. Specifically, the generation unit 132 calculates a score based on pre-set criteria, such as assigning a score of "5" if the user's answers fully conform to the intent of the question, and a score of "1" if the answers do not fully conform to the intent of the question, etc.

[0051] Furthermore, the generation unit 132 identifies parts of the user's answer that do not conform to the intent of the question, based on the learning model. Also, if the user's answer does not conform to the intent of the question, the generation unit 132 generates a correct answer based on pre-prepared answers or a correct answer generated based on the question text and surrounding context as a result of the content check. Furthermore, the output unit 133 generates comments to provide feedback to the user based on the number of answers that do not conform to the intent of the question, the types of errors, and a comparison of the answer results with other users.

[0052] The generation unit 132 generates scoring results using vocabulary as a predetermined scoring criterion. Specifically, if the user's response does not meet the pre-set conditions for the number of words or the way the words are written, the generation unit 132 calculates a score based on pre-set criteria, such as "5" if all of the response meets the pre-set conditions for the number of words or the way the words are written, and "1" if not, etc.

[0053] Furthermore, the generation unit 132 identifies locations that do not meet the pre-set conditions for how words should be written, based on the learning model. The generation unit 132 also generates correct answers based on pre-prepared responses or correct answers generated based on the prompt text and surrounding context as a result of the vocabulary check for locations that do not meet the pre-set conditions for how words should be written. In addition, the generation unit 132 generates comments to provide feedback to the user based on the number of words, the number of locations that do not meet the pre-set conditions for how words should be written, the types of errors, and a comparison of the user's response results with those of other users.

[0054] (Output section 133) The output unit 133 outputs to the user one or more predetermined scoring results generated by the generation unit 132. Specifically, the output unit 133 outputs to the user one or more of the scoring results related to grammar, composition, content, and vocabulary generated by the generation unit 132.

[0055] The output unit 133 outputs, from a grammatical perspective, a predetermined score determined based on the error rate, highlighting of the error locations in the user's response, a corrected version with the errors corrected, and comments on the user's response as the scoring result based on predetermined scoring criteria.

[0056] Specifically, the output unit 133 displays to the user the grammatical score calculated by the generation unit 132. The output unit 133 also highlights the locations of errors using red text, bold text, underlining, or background fill. The output unit 133 also displays corrected text related to the locations of errors. Furthermore, the output unit 133 displays comments to provide feedback to the user based on the number of errors, the number of locations where errors occurred, the types of errors, and a comparison of the user's response with other users.

[0057] The output unit 133 outputs, with regard to the configuration, a predetermined score determined based on the presence or absence of a predetermined configuration, highlighting of parts of the user's answer that do not have the predetermined configuration, the result of checking the configuration of the user's answer content, and comments on the user's answer content as the scoring result based on the predetermined scoring criteria.

[0058] Specifically, the output unit 133 displays to the user the score related to the compositional aspects calculated by the generation unit 132. The output unit 133 also highlights parts that do not conform to the specified compositional structure using red text, bold text, underlining, or background fill. Furthermore, the output unit 133 displays the compositional check results for parts that do not conform to the specified compositional structure, including correct answers based on pre-prepared answers or correct answers generated based on the problem statement and surrounding context. In addition, the output unit 133 displays comments to provide feedback to the user based on the number of parts that do not conform to the specified compositional structure, the types of errors, and a comparison of the user's answer with other users' answers.

[0059] The output unit 133 outputs, in terms of content, a predetermined score determined based on whether or not the user's answer conforms to the intent of the question, highlighting of parts of the user's answer that do not conform to the intent of the question, the results of checking the user's answer content, and comments on the user's answer content as the scoring result based on predetermined scoring criteria.

[0060] Specifically, the output unit 133 displays to the user the score related to the content aspects calculated by the generation unit 132. Furthermore, if the user's answer does not align with the intent of the question, the output unit 133 outputs a correct answer based on pre-prepared answers or a correct answer generated based on the question text and surrounding context as a result of the content check. The output unit 133 also outputs comments to provide feedback to the user regarding their answer, based on the number of incorrect answers, the types of errors, and comparisons with other users' answers.

[0061] The output unit 133 outputs, in terms of vocabulary, a predetermined score determined based on whether the number of words or the way words are written meets predetermined conditions, highlighting of errors in the writing method, the results of checking words or writing methods, and comments on the user's response as the scoring result based on predetermined scoring criteria.

[0062] Specifically, the output unit 133 displays to the user the score related to vocabulary calculated by the generation unit 132. The output unit 133 also highlights areas that do not meet the pre-set conditions for word formatting using red text, bold text, underlining, or background fill. Furthermore, for areas that do not meet the pre-set conditions for word formatting, the output unit 133 outputs correct answers based on pre-prepared responses or correct answers generated based on the prompt text and surrounding context as the vocabulary check result. In addition, the output unit 133 outputs comments to provide feedback to the user based on the number of words, the number of areas that do not meet the pre-set conditions for word formatting, the type of errors, and a comparison of the user's response with other users' responses.

[0063] Furthermore, the output unit 133 displays information such as "TOPIC" and "POINTS" stored in the task information DB 121 on the output screen related to the scoring results based on each scoring criterion. The output unit 133 also outputs to the user an output screen displaying information such as "Answer" stored in the answer information DB 122. Furthermore, the output unit 133 outputs to the user an output screen displaying information such as "Score" and "Details" for each scoring criterion stored in the scoring result information DB 123.

[0064] (Example of a scoring result output screen) From here, an example of an output screen related to the scoring results output by the scoring device 100 according to this embodiment will be described using Figures 8 to 11. Figures 8 to 11 are diagrams showing an example of a scoring result output screen according to this embodiment.

[0065] (An example of an output screen related to "grammatical perspectives") First, using Figure 8, we will explain an example of an output screen related to "whether or not it contains grammatical or vocabulary errors (grammatical perspective)."

[0066] As shown in Figure 8, the scoring device 100 outputs a screen displaying the scoring results from a grammatical perspective (Figure 8(1)), the scoring points (Figure 8(2)), the user's answer (Figure 8(3)), the corrected text (Figure 8(4)), and comments on the user's answer (Figure 8(5)).

[0067] As shown in Figure 8(1), the scoring device 100 can output a score for the answer based on the number of grammatical or vocabulary errors included in the user's answer as a scoring result. For example, the scoring device 100 can output a score of "Grammar: (out of 5)" as a scoring result related to grammar (Figure 8(1-1)).

[0068] As shown in Figure 8 (2), the scoring device 100 can output information on the points used for scoring, as well as an overview of the scoring results for the user. For example, the scoring device 100 can display text about the scoring points, such as "Is the correct grammar used?" or "Is the spelling correct?", in the area shown in Figure 8 (2-1). The scoring device 100 can also display text such as "There were ○ grammar and spelling mistakes. It's a passing grade, but let's try to reduce the number of mistakes even more!" in the area shown in Figure 8 (2-2).

[0069] As shown in Figure 8 (3), the scoring device 100 displays the user's response to the English task. Furthermore, the scoring device 100 can highlight and display correction comments for parts of the user's response that contain grammatical or vocabulary errors.

[0070] For example, if the scoring device 100 finds that the user's answer contains tense or spelling errors, it can highlight the relevant words or expressions by changing their color and formatting (for example, they may be displayed in the area shown in (3-1) of Figure 8). Specifically, if the scoring device 100 finds that "the past tense or future tense is used in a place where the present tense should be used (for example, the area shown in (3-1) of Figure 8)," it can display "highlighting of the relevant section (Figure 8 (3-1))" and a correction comment such as "Let's use the present tense" (Figure 8 (3-2)) on the highlighted section, as shown in Figure 8 (3-1).

[0071] As shown in Figure 8 (4), the scoring device 100 can score the user's answer and then display a corrected version of the answer with the errors corrected to the correct answer. For example, the scoring device 100 can display a corrected version of the user's answer with "was" corrected to the correct answer "is" (for example, it may be displayed in the area shown in Figure 8 (4-1)).

[0072] As shown in Figure 8 (5), the scoring device 100 can display comments and other information regarding the user's response. For example, the scoring device 100 can display "TOPIC (Figure 8 (5-1))" and "POINTS (Figure 8 (5-2))" as conditions for the response set for the task.

[0073] Furthermore, for example, the scoring device 100 can display an example English answer and a Japanese translation of the example English answer as examples of answers to the English task presented to the user (Figure 8 (5-3)). The scoring device 100 can also display an explanation of the example answer to the English task presented to the user (Figure 8 (5-4)).

[0074] (An example of an output screen related to "Configuration Perspectives") Next, using Figure 9, we will explain an example of an output screen related to the "perspective of whether or not the user's response has a predetermined structure (perspective of structure)."

[0075] As shown in Figure 9, the scoring device 100 outputs a screen displaying the scoring results related to the configuration (Figure 9(1)), the scoring points (Figure 9(2)), the user's response (Figure 9(3)), the configuration check (Figure 9(4)), and comments on the user's response (Figure 9(5)). Note that the comments on the user's response shown in Figure 9(5) are the same as those in Figure 8(5), so an explanation is omitted in this section.

[0076] As shown in Figure 9(1), the scoring device 100 can output a score for the user's answer based on the composition of the answer as the scoring result. For example, the scoring device 100 can output a score of "Composition: (out of 5)" as the scoring result related to the composition aspect (Figure 9(1-1)).

[0077] As shown in Figure 9 (2), the scoring device 100 can output information on the points used for scoring, as well as an overview of the scoring results for the user. For example, the scoring device 100 can display text about the scoring points, such as "Is it written in the specified structure?" or "Are the claims and evidence linked together?", in the area shown in Figure 9 (2-1). The scoring device 100 can also display text such as "The structure and flow are clearly written. Keep up the good work!" in the area shown in Figure 9 (2-2).

[0078] As shown in Figure 9(3), the scoring device 100 displays the user's response to the English task.

[0079] As shown in Figure 9(4), the scoring device 100 can display the results of a structural check performed on the user's response. For example, the scoring device 100 displays the check results for the user's response regarding "claim," "reason," and "basis" (Figure 9(4-1)). Furthermore, the scoring device 100 displays a summary comment regarding the check results for the user's response in the area shown in Figure 9(4-2). The aforementioned "summary comment" may be, for example, "This paragraph contains one of your claims regarding the TOPIC. Your claim regarding the TOPIC is appropriately stated."

[0080] (An example of an output screen related to "Content Perspective") Next, using Figure 10, we will explain an example of an output screen related to the perspective of "whether the user's answer is in line with the intent of the question (perspective of content)."

[0081] As shown in Figure 10, the scoring device 100 outputs a screen displaying the scoring results related to the content (Figure 10(1)), the scoring points (Figure 10(2)), the user's response (Figure 10(3)), the content check (Figure 10(4)), and comments on the user's response (Figure 10(5)). Note that the comments on the user's response shown in Figure 10(5) are the same as those in Figure 8(5), so their explanation is omitted in this section.

[0082] As shown in Figure 10(1), the scoring device 100 can output a score for the user's response based on the content of the response as a scoring result. For example, the scoring device 100 can output a score of "Content: (out of 5)" as a scoring result related to the content (Figure 10(1-1)).

[0083] As shown in Figure 10 (2), the scoring device 100 can output information on the points used for scoring, as well as an overview of the scoring results for the user. For example, the scoring device 100 can display text in the area shown in Figure 10 (2-1) regarding scoring points related to whether the answer is in line with the intent of the question, such as "Is the writing consistent?", "Does the content address the question?", and "Are two points used?". The scoring device 100 can also display text in the area shown in Figure 10 (2-2) such as "The entire text is consistent and the content is well written, so keep up the good work!".

[0084] As shown in Figure 10 (3), the scoring device 100 displays the user's response to the English task.

[0085] As shown in Figure 10 (4), the scoring device 100 can display the results of a content check on the user's response. Specifically, the scoring device 100 displays comments on "consistency of writing" and "content of writing" as check results related to "your opinion," etc. (Figures 10 (4-1) and (4-2)). In the example shown in Figure 10, only the check results for "your opinion" are illustrated, but the device is not limited to this and can also display check results related to "reasons," etc.

[0086] For example, the scoring device 100 can display text such as "The argument for the QUESTION and the evidence for the reasoning are appropriately written" as a comment on "the consistency of the writing" in the area shown in (4-1) of Figure 10. In addition, the scoring device 100 can display text such as "The argument for the QUESTION and the evidence for the reasoning are appropriately written" as a comment on "the content of the writing" in the area shown in (4-2) of Figure 10.

[0087] (An example of an output screen related to the "vocabulary perspective") Next, using Figure 11, we will explain an example of an output screen related to "whether the number of words included in the user's response or the way the words are written meets predetermined conditions (vocabulary perspective)."

[0088] As shown in Figure 11, the scoring device 100 outputs a screen displaying the scoring results related to vocabulary (Figure 11(1)), scoring points (Figure 11(2)), the user's response (Figure 11(3)), a vocabulary check (Figure 11(4)), and comments on the user's response (Figure 11(5)). Note that the comments on the user's response shown in Figure 11(5) are the same as those in Figure 8(5), so an explanation is omitted in this section.

[0089] As shown in Figure 11(1), the scoring device 100 can output a score for the user's answer based on the content of the answer as a scoring result. For example, the scoring device 100 can output a score of "Vocabulary: (out of 5)" as a scoring result related to vocabulary (Figure 11(1-1)).

[0090] As shown in Figure 11 (2), the scoring device 100 can output information on the points used for scoring, as well as an overview of the scoring results for the user. For example, the scoring device 100 can display text about the scoring points, such as "Is the writing done using the specified vocabulary?", "Are abbreviated expressions used?", and "Are the same expressions used repeatedly?", in the area shown in Figure 11 (2-1). The scoring device 100 can also display text such as "Abbreviated and repeated expressions are used, so please check the 'Vocabulary Check' and review." in the area shown in Figure 11 (2-2).

[0091] As shown in Figure 11 (3), the scoring device 100 displays the user's response to the English task. For example, the scoring device 100 can highlight words or expressions in the user's response that are undesirable, such as "I'm," by changing the color tone (for example, they may be displayed in the area shown in Figure 11 (3-1)).

[0092] As shown in Figure 11 (4), the scoring device 100 can display the results of a vocabulary check on the user's answer. Specifically, the scoring device 100 displays comments on the check results related to "vocabulary," "abbreviated expressions," "repeated expressions," etc. (Figure 11 (4-1) to (4-3)).

[0093] For example, the scoring device 100 can display text such as, "The target word count for Grade 1 is 120 to 150 words. You have written well within the target word count range," as a comment about "vocabulary" in the area shown in (4-1) of Figure 11.

[0094] Furthermore, the scoring device 100 can display comments about "abbreviated expressions" such as, "The following expressions are used. In English composition, do not use abbreviated expressions such as 'I'm', but instead use 'I am'," in the area shown in (4-2) of Figure 11.

[0095] Furthermore, the scoring device 100 can display text such as "The repeated expression 'I think that' is used. When writing the same content, use different expressions." in the area shown in (4-3) of Figure 11 as a comment about "repeated expressions."

[0096] (Processing procedure) From here, the processing procedure related to the scoring device 100 according to this embodiment will be described. Figure 12 is a flowchart showing the scoring process according to this embodiment.

[0097] The reception unit 131 receives information regarding the English assignment and information regarding the response to the assignment (S101). Next, the generation unit 132 inputs the information regarding the English assignment and information regarding the response to the assignment into the learning model (S102). Next, the generation unit 132 generates scoring results based on predetermined scoring criteria (S103). Then, the output unit 133 outputs the scoring results based on predetermined scoring criteria (S104).

[0098] (effect) The effects of the scoring device 100 according to this embodiment will now be explained. In recent years, it has become known that users themselves use applications and the like to learn English, but it can be difficult to score the answer results according to the user's learning objectives.

[0099] For example, when a user studies English in preparation for taking the Eiken English proficiency test, it can be difficult to have an application or other system perform scoring based on criteria such as "grammar," "structure," "content," and "vocabulary," which are consistent with the scoring criteria of the Eiken test.

[0100] Therefore, the generation unit 132 of the scoring device 100 takes information about the English task and information about the answer to the task as input and generates one or more scoring results based on predetermined scoring criteria, based on a learning model that outputs scoring results based on predetermined scoring criteria for the answer to the English task. The output unit 133 of the scoring device 100 outputs one or more of the generated scoring results based on predetermined scoring criteria to the user.

[0101] Therefore, the scoring device 100 of this embodiment has the effect of enabling scoring based on multiple scoring criteria according to the learning objectives. In other words, when scoring answers made by a user, the scoring device 100 can output scoring results according to the user's selection of scoring criteria and the user's English learning objectives. As a result, the scoring device 100 has the effect of enabling effective English learning by the user.

[0102] Furthermore, in order to achieve the effects described above, the scoring device 100 performs the following specific processes.

[0103] The generation unit 132 inputs information about the English assignment, including the difficulty level of the assignment and the conditions for answering the assignment, and information about the answer to the assignment, including the answer to the assignment entered by the user in natural language, into a trained learning model. The generation unit 132 then generates one or more scoring results based on predetermined scoring criteria, based on the user's selection of predetermined scoring criteria.

[0104] As described above, the scoring device 100 can perform scoring based on multiple scoring criteria when scoring answers to tasks entered by the user, based on pre-set tasks and answer conditions for those tasks. Therefore, the scoring device 100 has the effect of enabling scoring based on scoring criteria such as grammar, structure, content, and vocabulary, depending on the user's selection and the level of the Eiken test or other exam the user is taking.

[0105] The generation unit 132 generates a scoring result by using at least one of the following as predetermined scoring criteria: grammar, structure, content, and vocabulary, based on the scoring criteria selected by the user or the scoring criteria corresponding to the level of the Eiken test being taken.

[0106] For example, the output unit 133 outputs, from a grammatical standpoint, at least one of the following as a scoring result based on predetermined scoring criteria: a predetermined score determined based on the error rate, highlighting of the locations where errors occurred in the user's answer, a corrected version with the errors corrected, and comments on the user's answer.

[0107] As described above, the scoring device 100 can perform scoring based on the "grammar" scoring criteria when the user selects a scoring criterion related to "grammar" or when the level of the Eiken test being taken includes a "grammar" scoring criterion. In other words, when the user desires to focus on grammar learning, the scoring device 100 enables more effective English learning by displaying the score related to the grammar scoring criterion, clearly indicating where errors occurred, and showing corrected versions of the answers and comments on the answers.

[0108] For example, the output unit 133 outputs, as a scoring result based on a predetermined scoring criterion, at least one of the following: a predetermined score determined based on the presence or absence of a predetermined configuration; highlighting of parts of the user's answer that do not have the predetermined configuration; the result of checking the configuration of the user's answer content; and comments on the user's answer content.

[0109] As described above, the scoring device 100 can perform scoring based on the "structure" scoring criterion when the user selects a scoring criterion related to "structure" or when the level of the Eiken test being taken includes the "structure" scoring criterion. In other words, when the user desires to focus on learning about structure, the scoring device 100 can enable more effective English learning by displaying the score related to the structure scoring criterion, the results of the structure check of the answer, and comments on the content of the answer to the user.

[0110] For example, the output unit 133 outputs, in terms of content, a predetermined score determined based on whether or not the user's answer conforms to the intent of the question, highlighting of parts of the user's answer that do not conform to the intent of the question, the results of checking the user's answer content, and comments on the user's answer content as a scoring result based on predetermined scoring criteria.

[0111] As described above, the scoring device 100 can perform scoring based on the scoring criteria for "content" when the user selects a scoring criterion related to "content" or when the level of the Eiken test being taken includes a scoring criterion for "content". In other words, when the user desires to focus on studying content, the scoring device 100 displays the user a score related to the scoring criterion for that content, the results of the content check for the answer, and comments on the answer, thereby enabling more effective English learning.

[0112] For example, the output unit 133 outputs, in terms of vocabulary, at least one of the following as a scoring result based on predetermined scoring criteria: a predetermined score determined based on whether the number of words or the way words are written meets predetermined conditions; highlighting of errors in the writing method; the result of checking the words or writing method; and comments on the user's response.

[0113] As described above, the scoring device 100 can perform scoring based on the "vocabulary" scoring criteria when the user selects a scoring criterion related to "vocabulary" or when the level of the Eiken test being taken includes a "vocabulary" scoring criterion. In other words, when the user desires to focus on vocabulary learning, the scoring device 100 enables more effective English learning by displaying the score related to the vocabulary scoring criterion, indicating any errors in the writing method, showing the results of word and writing method checks, and providing comments on the answers.

[0114] With the functions described above, the scoring device 100 can automatically score and correct the answers that the user provides when they are studying English independently. Therefore, the scoring device 100 enables the user to practice English repeatedly on their own.

[0115] Furthermore, the scoring device 100 makes it possible to streamline the scoring work not only for the user but also for the instructors who guide the user. Specifically, before the user submits the answers they have created in their English learning to the instructor, the scoring device 100 can automatically score and correct the answers the user has made. The scoring device 100 also allows the user to correct any mistakes in the answers beforehand before submitting them to the instructor. As a result, the scoring device 100 makes it possible to reduce the workload of instructors in scoring work compared to before.

[0116] <Variation> The following describes modifications that can be implemented by the scoring device 100 according to this embodiment.

[0117] (Applicable areas) In this embodiment, an example of how scoring processing for the "Eiken" (English Proficiency Test) is carried out has been described, but it is not limited to this. However, the scoring device 100 according to this embodiment can realize scoring processing for purposes other than the Eiken, such as learning instruction in educational institutions, or learning according to a predetermined difficulty level set by the user.

[0118] (Data, etc.) The information regarding the English task, the information regarding the response to the task, the scoring criteria (grammar, structure, content, vocabulary), the scoring results, the names of the functional parts of the scoring device 100, steps, processes, and the names of steps or processes used in the description of the above embodiment are merely examples and can be changed at will.

[0119] For example, while it was explained that the issue information DB121 stores "RANK," "QUESTION," "TOPIC," and "POINTS" associated with "No," which is information that identifies individual issues, it is not limited to this.

[0120] Furthermore, for example, while it was explained that the response information DB122 stores the user and the response in association with "No," which is information that identifies individual response information, it is not limited to this. Also, for example, while it was explained that the scoring result information DB123 stores the user and the scoring result for each scoring criterion in association with "No," which is information that identifies individual scoring result information, it is not limited to this.

[0121] (Regarding the output screen) The format of the output screen in this embodiment is not particularly limited. For example, as shown in Figures 8 to 11, the scoring device 100 displays an output screen to the user relating to the scoring results based on predetermined scoring criteria such as grammar, structure, content, and vocabulary. However, the target scoring criteria, the content to be displayed, the display position of the content, the format when displaying the content, etc., are not particularly limited. In other words, the scoring device 100 can generate an output screen relating to the scoring results based on predetermined scoring criteria by arbitrarily combining predetermined formats, display positions, layout configurations, etc., and display it to the user.

[0122] (Flowcharts, etc.) In flowcharts, each step may be rearranged as long as it does not create inconsistencies, and some steps may be omitted. Furthermore, conjunctions such as "next," "continue," "in addition," "at this time," and "on this occasion" in flowchart descriptions do not limit the order or timing of the processes in the flowchart.

[0123] <Hardware Configuration> Each component of the illustrated device is a functional concept and does not necessarily have to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions. Furthermore, each processing function performed by each device can be implemented, all or any part of it, by a CPU and the program that is analyzed and executed by that CPU, or by hardware using wired logic.

[0124] Furthermore, among the processes described in this embodiment, all or part of those described as being performed automatically can be performed manually using known methods. In addition, the processing procedures, control procedures, specific names, and information including various data and parameters shown in the drawings can be arbitrarily changed unless otherwise specified.

[0125] <Program> In one embodiment, the various devices constituting the scoring device 100 can be implemented by installing a scoring program as packaged software or online software on a desired computer. For example, by having the above-mentioned scoring program run on an information processing device, it can function as various devices constituting the scoring device 100. The information processing device referred to here includes desktop or notebook personal computers. In addition, the information processing device also includes mobile communication terminals such as smartphones and mobile phones, and slate terminals such as PDAs (Personal Digital Assistants).

[0126] Figure 13 shows an example of a computer that performs scoring processing according to this embodiment. Computer 1000 has, for example, memory 1010 and CPU 1020. Computer 1000 also has a hard disk drive interface 1030, a disk drive interface 1040, a serial port interface 1050, a video adapter 1060, and a network interface 1070. These components are connected by a bus 1080.

[0127] Memory 1010 includes ROM (Read Only Memory) 1011 and RAM 1012. ROM 1011 stores, for example, a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 1030 is connected to the hard disk drive 1090. The disk drive interface 1040 is connected to the disk drive 1100. For example, a removable storage medium such as a magnetic disk or optical disk is inserted into the disk drive 1100. The serial port interface 1050 is connected to, for example, a mouse 1110 and a keyboard 1120. The video adapter 1060 is connected to, for example, a display 1130.

[0128] The hard disk drive 1090 stores, for example, an OS (Operating System) 1091, an application program 1092, a program module 1093, and program data 1094. That is, the program that defines each process of the various devices constituting the scoring device 100 is implemented as a program module 1093 in which code executable by a computer is written. The program module 1093 is stored, for example, in the hard disk drive 1090. For example, a program module 1093 for performing the same processes as the functional configuration of the various devices constituting the scoring device 100 is stored in the hard disk drive 1090. Note that the hard disk drive 1090 may be replaced by an SSD (Solid State Drive).

[0129] Furthermore, the configuration data used in the processing of the embodiment described above is stored as program data 1094 in, for example, memory 1010 or hard disk drive 1090. The CPU 1020 then reads the program module 1093 and program data 1094 stored in memory 1010 or hard disk drive 1090 into RAM 1012 as needed and executes the processing of the embodiment described above.

[0130] Furthermore, the program module 1093 and program data 1094 are not limited to being stored in the hard disk drive 1090; for example, they may be stored in a removable storage medium and read by the CPU 1020 via a disk drive 1100 or the like. Alternatively, the program module 1093 and program data 1094 may be stored in another computer connected via a network (LAN, WAN (Wide Area Network), etc.). The program module 1093 and program data 1094 may then be read from the other computer by the CPU 1020 via a network interface 1070.

[0131] <Other> Although this embodiment has been described above, this embodiment is not limited by the description and drawings that constitute part of the disclosure. That is, all other embodiments, examples, and operational techniques made by those skilled in the art based on this embodiment are included in the scope of this embodiment. [Explanation of Symbols]

[0132] 100 scoring devices 110 Communications Department 120 Storage section 121 Issue Information Database 122 Answer information DB 123 Scoring Result Information Database 124 Learning Model DB 130 Control Unit 131 Reception Department 132 Generation part 133 Output section

Claims

1. A generation unit generates one or more scoring results based on predetermined scoring criteria, based on a learning model that takes information about an English assignment and information about the answer to said assignment as input and outputs scoring results based on predetermined scoring criteria for the answer to said English assignment, An output unit that outputs to the user a scoring result based on one or more predetermined scoring criteria generated by the generation unit, A scoring device characterized by having the following features.

2. The generating unit is Regarding the aforementioned English assignment, the difficulty level of the assignment related to learning English and the conditions for answering the assignment, As information regarding the answer to the said problem, the answer to the said problem, which is entered by the user in natural language, Input into the previously trained model, Based on the user's selection of predetermined scoring criteria, the system generates one or more scoring results based on the predetermined scoring criteria. The scoring device according to feature 1.

3. The generating unit is At least one of the following perspectives is considered: whether the user's answer contains grammatical or vocabulary errors related to English; whether the user's answer has a predetermined structure; whether the user's answer is in line with the intent of the question; and whether the number of words or the way the words are written in the user's answer satisfies predetermined conditions. The above predetermined scoring criteria are used to generate the scoring results. The scoring device according to claim 1 or 2.

4. The output unit is, Regarding whether the user's response contains grammatical or vocabulary errors related to English, A predetermined score determined based on the aforementioned error rate, Highlighting of the location where the error occurred in the user's response, The corrected version with the aforementioned error corrected, Comments on the user's response, At least one of the above is output as the scoring result based on the predetermined scoring criteria. The scoring device according to feature 3.

5. The output unit is, Regarding the perspective of whether the user's response has a predetermined configuration, A predetermined score determined based on the presence or absence of the aforementioned predetermined configuration, Highlighting of the parts of the user's response that do not have the predetermined configuration, The results of checking the structure of the user's response, Comments on the user's response, At least one of the above is output as the scoring result based on the predetermined scoring criteria. The scoring device according to feature 3.

6. The output unit is, Regarding whether the user's answer is in line with the intent of the question, A predetermined score is determined based on whether the user's answer is in line with the intent of the question, Highlighting of parts of the user's response that do not align with the intent of the question, The results of checking the user's response, Comments on the user's response, At least one of the above is output as the scoring result based on the predetermined scoring criteria. The scoring device according to feature 3.

7. The output unit is, Regarding whether the number of words or the way the words are written in the user's response meets the prescribed conditions, A predetermined score determined based on whether the number of the aforementioned words or the way the aforementioned words are written satisfies predetermined conditions, The error in the aforementioned description method is highlighted, The result of checking the aforementioned word or the aforementioned method of description, Comments on the user's response, At least one of the above is output as the scoring result based on the predetermined scoring criteria. The scoring device according to feature 3.

8. A scoring method to be executed by a scoring device, A generation process that generates one or more combinations of scoring results based on predetermined scoring criteria, based on a learning model that takes information about an English assignment and information about the answer to said assignment as input and outputs scoring results based on predetermined scoring criteria for the answer to said English assignment, Output step for outputting to the user one or more scoring results based on the predetermined scoring criteria generated in the generation step, A scoring method characterized by including [a specific element].

9. A generation step that generates one or more combinations of scoring results based on predetermined scoring criteria, based on a learning model that takes information about an English assignment and information about the answer to said assignment as input and outputs scoring results based on predetermined scoring criteria for the answer to said English assignment, Output step: Outputting to the user one or more scoring results based on the predetermined scoring criteria generated in the generation step, A scoring program that causes a computer to execute a scoring command.

Images