English learning support device, English learning support method, and English learning support program

Abstract

This enables appropriate support for English language learning. [Solution] The English learning support device identifies the attributes of words in an English sentence by inputting an English sentence into a model trained on a system diagram that defines the rules of sentence structure based on word order, tense, and verbs, which are parts of speech that define the actions or states of things in English. The English learning support device outputs correspondence information to the user that associates the words in the English sentence with the position of those words in the system diagram that shows the rules of sentence structure, using the attributes of the words.

Description

Technical Field

[0001] The present invention relates to an English learning support device, an English learning support method, and an English learning support program.

Background Art

[0002] Since the language systems of English and Japanese are completely different, in order for Japanese people whose native language is not English to master English as a foreign language, it is necessary to reduce the bias of Japanese as their native language as much as possible.

[0003] For example, as one of the differences between English and Japanese described above, there is a difference in word order. Therefore, there is a known prior art that realizes English education with awareness of word order by determining whether the number sequences corresponding to the three components (subject, tense, and sentence pattern) in an English sentence match the number sequence for each English sentence (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, the prior art has problems in appropriately supporting English learning. For example, the prior art can support the learning of the structure peculiar to English sentences. However, the prior art has problems in supporting the acquisition of accurate English after accurately understanding the role of word classes in English grammar and eliminating the bias of Japanese people whose native language is Japanese.

Means for Solving the Problems

[0006] Therefore, in order to solve the above problems and achieve the objective, the English learning support device of the present invention is characterized by comprising: an identification unit that identifies the attributes of words contained in an English sentence by inputting an English sentence to a model trained on a system diagram that defines the rules for constructing English sentences based on word order, tense, and verbs, which are parts of speech that define the actions or states of things in English; and an output unit that outputs correspondence information to the user, which associates the words contained in the English sentence with the position of the word in the system diagram that shows the rules for constructing the English sentence, using the attributes of the word. [Effects of the Invention]

[0007] This invention has the effect of enabling appropriate support for English language learning. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a diagram illustrating the concept of a grammar table. [Figure 2] Figure 2 is a diagram illustrating the overall overview of the English learning support according to the embodiment. [Figure 3] Figure 3 shows an example of the configuration of an English learning support device according to an embodiment. [Figure 4] Figure 4 is a table diagram showing an example of problem information according to the embodiment. [Figure 5] Figure 5 is a table diagram showing an example of user information according to the embodiment. [Figure 6] Figure 6 shows an example of a specific flow chart for the English structure according to this embodiment. [Figure 7] Figure 7 shows an example of a specific flow chart for the English structure according to this embodiment. [Figure 8] Figure 8 shows an example of a specific flow chart for the English structure according to this embodiment. [Figure 9] Figure 9 shows an example of a specific flow chart for the English structure according to this embodiment. [Figure 10] Figure 10 shows an example of a specific flow chart for the English structure according to this embodiment. [Figure 11] FIG. 11 is a diagram showing an example of a specific flow of an English composition according to an embodiment. [Figure 12] FIG. 12 is a diagram showing an example of a specific flow of an English composition according to an embodiment. [Figure 13] FIG. 13 is a diagram showing an example of a specific flow of an English composition according to an embodiment. [Figure 14] FIG. 14 is a diagram showing an example of a specific flow of an English composition according to an embodiment. [Figure 15] FIG. 15 is a diagram showing an example of an output command according to an embodiment. [Figure 16] FIG. 16 is a diagram showing an example of an output command according to an embodiment. [Figure 17] FIG. 17 is a diagram showing an example of an output command according to an embodiment. [Figure 18] FIG. 18 is a diagram showing an example of an output in JSON format according to an embodiment. [Figure 19] FIG. 19 is a diagram showing an example of an output of an English learning screen according to an embodiment. [Figure 20] FIG. 20 is a flowchart showing the processing of an English learning support device according to an embodiment. [Figure 21] FIG. 21 is a diagram showing an example of the hardware configuration of a computer that realizes an English learning support device according to an embodiment.

Embodiments of the Invention

[0009] Hereinafter, embodiments for implementing the present invention (hereinafter referred to as "embodiments") will be described with reference to the drawings. Note that each embodiment is not limited to the content described below.

[0010] <Introduction> (Background) English, which belongs to the Germanic branch within the Indo-European language family, and Japanese, which belongs to the Altaic language family, a minority group globally, have completely different language systems. Therefore, in order for Japanese people, whose mother tongue is not English, to master English as a foreign language, it is necessary to minimize the bias inherent in Japanese as their mother tongue.

[0011] One of the differences between English and Japanese mentioned above is the word order. For example, English has a word order of "subject → predicate → object, etc.", while Japanese has a significant difference in word order, such as "subject → object, etc. → predicate".

[0012] (English learning method based on grammar tables) Therefore, in order to understand English as it is, an English learning method using a grammar table (for example, refer to Reference 1) is known.

[0013] (Reference 1) Fujikawa Method Grammar Table, <URL:https: / / www.liberty-e.com / commitment / >, <Search date: August 28, 2025>

[0014] From here, the concept of the grammar table will be explained using Figure 1. Figure 1 is a diagram for explaining the concept of the grammar table. Figure 1 shows a grammar table indicating that the structure of an English sentence is determined centered around Verb.

[0015] Note that the concept of the grammar table explained using Figure 1 is a part of the content related to the grammar table shown in Reference 1 and the like. Also, in this embodiment, the grammar table means "a system diagram that determines the sentence structure rules of English based on the word order, tense, and the part of speech Verb that defines the action or state of things in English".

[0016] First, let me explain the challenges of English education in Japan. Current English education in Japan is said to have begun with its instruction in the old-style middle schools after the Meiji Restoration. The textbooks used at that time were simply translations of English grammar textbooks written in English into Japanese. And this practice of simply translating English grammar into Japanese continues to this day.

[0017] For example, in English education in Japan, the correspondence between English words is taught as follows: "Verb" is "verb," ​​"Noun" is "noun," "Adverb" is "adverb," ​​and "Adjective" is "adjective." Here, it can be said that the above correspondence holds true if all items have the same function in English and Japanese.

[0018] However, English and Japanese have fundamentally different linguistic systems, and even if we try to force English grammar to fit into Japanese, we cannot directly express the meanings of each part of speech in English grammar in Japanese grammar.

[0019] For example, the most important function in English, "Verb," ​​originally derives from the Latin word "Verbum," which originally meant "language" or "word." In other words, the English "Verb" has a function that goes beyond simply "expressing an action or state," as the Japanese verb does. Therefore, simply translating the English "Verb" into the Japanese "verb" does not accurately capture the original meaning of the verb in terms of Japanese concepts, leading to discrepancies in understanding English grammar.

[0020] Therefore, as mentioned above, in order to acquire English as a foreign language, it is effective to use grammar tables for English learning in order to eliminate as much as possible the filter of Japanese as one's native language and to understand English as English (that is, to understand the content in English without translating it from English to Japanese, etc.).

[0021] As shown in Figure 1, the grammar table indicates that "Verb," ​​which has the most important function in English, determines word order, tense, and action or state in English. First, we will explain in detail "Verb," ​​the most important part of speech in the grammar table.

[0022] The word "verb" originally means "language" or "word" (Etymology), and it exists as the core and center of English. Therefore, as shown in Figure 1 (1), a verb has four functions: "semantics," "syntax," "tense," and "verbal." In this respect, while the Japanese word "dōshi" (verb) shares the meaning of expressing action or state (semantics), the remaining three elements do not exist as concepts in Japanese verbs, and it can be said that more than 60% of the original function of a verb is different. In other words, while English verbs form a Golden Trinity, Japanese verbs are merely concepts that simply mean action or state.

[0023] "Semantics" is the only thing that Japanese verbs have in common with English, and it functions as a means. In other words, "semantics" can be said to express actions and states.

[0024] "Syntax" defines word order (primarily verbs) and agreement (similarities), and is known as the Golden Rule. In other words, "Syntax" defines the order in which parts of speech appear (word order) in English sentences. Furthermore, "Syntax" defines the word order from the beginning of a sentence to the verb, the word order after the verb according to its pattern, the verb (V) changes according to the subject (S), and the agreement of tenses with adverbs. In this respect, the concept of "Syntax" does not exist in Japanese, and simply translating English grammar into Japanese results in a scotoma (blind spot).

[0025] "Tense" defines 12 regular tenses (past tense, past progressive, present perfect, and past perfect progressive) and 12 passive tenses (past tense, past progressive, present perfect, and past perfect progressive). However, since tenses in Japanese are expressed using only three tenses—past, present, and future—and the passive voice, it differs from the English "Tense," creating a scotoma (blind spot).

[0026] "Verbal" defines the function of "Noun," "Adjective," or "Adverb" that is artificially created by applying syntax centered on verbs. For example, "Verbal" forms the core, which is a "Clause" or "Phrase" that has the function of a "Noun," "Adjective," or "Adverb," ​​and defines the structure of English sentences. In this respect, as mentioned above, Japanese lacks the concept of syntax and the concept of verbal, resulting in a scotoma (blind spot).

[0027] Furthermore, due to the presence of the Verbal, English sentences may contain a main clause (first hierarchical structure) which is the primary sentence based on the Main Verb, and multiple sentences (second hierarchical structure) based on the Verbal as a Core contained within the main clause. In other words, English sentences can be said to have a three-dimensional (organic) hierarchical structure formed by multiple Cores created by the Verbal, which is one of the functions of the Verb.

[0028] Based on the four functions of a verb—"Semantics," "Syntax," "Tense," and "Verbal"—as described above, the structure of an English sentence is determined as shown in the grammar table in Figure 1.

[0029] As shown in Figure 1, the grammar table has a structure centered around the Verb (Figure 1 (2)), consisting of a "Subject Part (Figure 1 (3))" from the beginning of the sentence to the Verb, and a "Predicate Part (Figure 1 (4))" from the Verb to the end of the sentence.

[0030] The "Subject Part (Figure 1 (3))" is defined by the "Time Agreement (Figure 1 (3-1))" and the "S+V Agreement (Figure 1 (3-2))" based on the "Syntax" of the Verb, which determine the structure of the English sentence. In other words, the "Subject Part" is the word order determined by the Verb, and the word order is determined based on the power (centripetal force) that the Verb exerts on each part of speech.

[0031] The "Time Agreement (Figure 1 (3-1))" mentioned above is a rule that, in principle, the tense of the verb in an adverb closer to the beginning of a sentence than the main verb (the verb shown in Figure 1 (2)) will match the tense of the main verb. Furthermore, the "S+V Agreement (Figure 1 (3-2))" is a rule that the main verb (the verb shown in Figure 1 (2)) changes depending on the subject noun.

[0032] Furthermore, the "Adjective" and "Adverb" included in the "Subject Part (Figure 1 (3))" have the following modifier relationships. Note that the following is merely an example, and detailed explanations of other modifier relationships are omitted. Figure 1 (3-3): adverb modifies verb Figure 1 (3-4): adjective modifies noun. Figure 1 (3-5): Adverb modifies word adjective. Figure 1 (3-6): Adverbs modify word / clause / phrase adjectives.

[0033] The "Predicate Part (Figure 1 (4))" is defined by the "Time Agreement (Figure 1 (4-1))" and the "Verb Pattern Agreement (Figure 1 (4-2))" based on the "Syntax" of the Verb, which determine the structure of the English sentence. In other words, the "Predicate Part" is the word order determined by the Verb, and the word order is determined based on the force (centrifugal force) that the Verb exerts on each part of speech.

[0034] The "Time Agreement (Figure 1 (4-1))" mentioned above is a rule that, in principle, the tense of the verb in an adverb located closer to the end of a sentence than the main verb (the verb shown in Figure 1 (2)) will match the tense of the main verb.

[0035] Furthermore, the "Verb Pattern Agreement (Figure 1 (4-2))" is a rule that determines the word order from the Main Verb onward according to V1 to V5 described below. V1: Subject + Verb V2:Subject+Verb+Complement(noun / adjective) V3: Subject + Verb + Object (noun) V4: Subject + Verb + Object (person) + Object (thing) V5:Subject+Verb+Object+Complement(noun / adjective)

[0036] As mentioned above, the grammar table-based English learning method and the traditional English learning method, which has been in use since the Meiji era, differ significantly in their approach to understanding the structure of English sentences. In other words, the traditional English learning method, which focuses on the isolated understanding of Japanese nouns, adjectives, and adverbs, makes it difficult to gain a true understanding of English nouns, adjectives, and adverbs.

[0037] Furthermore, traditional English learning methods lack a core concept of syntax for Japanese nouns, adjectives, and adverbs, which weakens the understanding of English nouns, adjectives, and adverbs and creates blind spots.

[0038] In other words, the traditional method of learning English by translating English words and phrases into Japanese has limitations in truly understanding the English language. Therefore, by using a learning method based on grammar tables, misunderstandings caused by mere translation can be eliminated, allowing learners to understand English as English and achieve a true understanding of English, centered on its syntax. Furthermore, while the structure of English was previously explained in text, grammar tables represent the structure of English sentences as diagrams, making it easier for users to intuitively understand the structure of English through visual means.

[0039] However, learning English based on the grammar table described above is difficult for anyone other than an instructor who understands the grammar table, and there are challenges in properly supporting English learning. For example, there is a known reference technique that helps learners acquire English word order by determining whether the number sequence corresponding to the three components of an English sentence (subject, tense, and sentence pattern) matches the number sequence for each sentence (see, for example, Reference 2).

[0040] (Reference 2) Japanese Patent Publication No. 2014-081662

[0041] However, while the aforementioned reference techniques enable the learning of sentence structures specific to English, they do not support English learning based on grammar tables. Therefore, it is difficult to acquire accurate English by using these reference techniques, as is possible with English learning methods based on grammar tables, by accurately understanding the roles of parts of speech in English grammar and eliminating the biases that native Japanese speakers have.

[0042] (Processing by English learning support device 100) Therefore, the English learning support device 100 according to this embodiment assigns predetermined attributes to words contained in an English sentence, and highlights and displays the part of speech of the word contained in the English sentence on the grammar table based on the predetermined attributes assigned to the word, thereby supporting the user in learning to understand English as it is.

[0043] Here, we will explain the overall process performed by the English learning support device 100. Figure 2 is a diagram illustrating the overall overview of the English learning support according to this embodiment. The English learning support device 100 shown in Figure 2 is an example of a computer that realizes information processing to support English learning based on a grammar table. In this embodiment, the concept of "grammar table" may include all the contents of the Fujikawa Method disclosed in Reference 1, in addition to the contents described above.

[0044] First, the English learning support device 100 inputs the English sentence (Figure 2(2)) entered by the user to a large-scale language model 10 that has been pre-trained by a prompt (Figure 2(1)) which includes learning data for a grammar table and a command to perform a process of assigning labels indicating predetermined attributes to words contained in an English sentence based on the grammar table (S10). The English sentence entered by the user may be the entire sentence entered by the user, a sentence selected by the user, or a sentence with some blanks filled in.

[0045] The English learning support device 100 acquires label information (Figure 2(3)), which is the result of labeling words, output from the large-scale language model 10 (S11). For example, the label information may be "label of attribute a for word A", "label of attribute b for word B", "label of attribute c for word C", etc. (Figure 2(3)).

[0046] The English learning support device 100 uses the acquired label information to output correspondence information to a terminal device used by the user 20, which associates a predetermined table (Figure 2 (4-1)) showing where the words contained in the English sentence are located in the grammar table with the English sentence (Figure 2 (4-2)), and displays it on the terminal device (S12).

[0047] In this way, the English learning support device 100 according to this embodiment has the effect of enabling the user to efficiently learn English based on a grammar table by highlighting and displaying where the part of speech of a word contained in an English sentence is located on the grammar table.

[0048] (English learning support device 100) From here, the detailed functions of the English learning support device 100 will be explained using Figure 3. Figure 3 is a diagram showing an example of the configuration of the English learning support device 100 according to this embodiment. As shown in Figure 3, the English learning support device 100 has a communication unit 110, a storage unit 120, and a control unit 130.

[0049] Although not shown in Figure 3, the English learning support device 100 may be equipped with an input unit such as a keyboard or mouse to receive input from an administrator or other user. Furthermore, the English learning support device 100 may be equipped with a display unit such as a screen to show the administrator task information, user information, etc., stored in the memory unit 120.

[0050] (Communications Department 110) The communication unit 110 performs data communication related to inputting training data or prompts to a large-scale language model, inputting English text, and outputting correspondence information. The communication unit 110 is implemented, for example, by a NIC (Network Interface Card) or a network interface controller. The communication unit 110 is connected to a network (e.g., the Internet) by wire or wireless connection. The communication unit 110 then sends and receives information with external devices via the network.

[0051] The communication unit 110 may also transmit and receive information using any communication standard or technology, such as Wi-Fi (registered trademark), Bluetooth (registered trademark), SIM (Subscriber Identity Module), or LPWA (Low Power Wide Area).

[0052] (Storage unit 120) The storage 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 storage unit 120 is implemented by, for example, 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 3, the storage unit 120 also includes task information DB 121, user information DB 122, model DB 123, and learning data DB 124.

[0053] (Issue Information DB121) The task information DB121 is a database that stores tasks related to English learning provided to users who are learning English based on a grammar table. Here, an example of task information stored in the task information DB121 will be explained using Figure 4. Figure 4 is a table diagram showing an example of task information according to this embodiment.

[0054] The assignment information DB121 stores the "Assignment" item and related information in a table format, associating it with "No," which is information that identifies individual teaching materials. For example, as shown in Figure 4, the assignment information DB121 stores assignment "Eng-1," which is identified by No "1."

[0055] The "assignments" described above may include, for example, lecture data such as videos and texts related to English learning based on grammar tables, assignment sentences used as practice problems, or correct answers and explanations corresponding to the practice problems. The data included in the "assignments" may include, for example, text, images, videos, or data in other file formats.

[0056] (User Information DB122) Returning to Figure 3, let's continue the explanation. User Information DB122 is a database that stores information about users who are learning English based on the grammar table as user information. Here, an example of user information stored in User Information DB122 will be explained using Figure 5. Figure 5 is a table diagram showing an example of user information according to the embodiment.

[0057] The user information DB122 stores "user identification information" and "attribute information" items, as well as information related to those items, in a table format or similar, associating them with "No," which is information that identifies individual teaching material information. For example, as shown in Figure 5, the user information DB122 stores the user identification information "User-1," identified by No "1," and the attribute information "Att-1" in association with each other.

[0058] The "user identification information" mentioned above refers to information that identifies a user who is learning English based on a grammar table, and may be information expressed as a combination of text, numbers, or symbols, such as the user's name, nickname (handle name), or unique identifier. Furthermore, "attribute information" refers to information relating to the user's attributes, and may include, for example, English proficiency, learning status, qualifications held, psychographic data, or demographic data.

[0059] (Model DB123) Returning to Figure 3, let's continue the explanation. Model DB123 is a database that stores models used for processing to assign predetermined attributes to words contained in English texts. For example, Model DB123 can store large-scale language models as models. In this embodiment, a large-scale language model is given as an example, but Model DB123 is not limited to this and can store other models such as machine learning models based on other known technologies.

[0060] Specifically, model DB123 can store large-scale language models such as "ChatGPT®," which is a large-scale language model with general-purpose knowledge (see, for example, reference 3).

[0061] (Reference 3):ChatGPT(OpenAI),<URL:https: / / openai.com / chatgpt> ,<Searched on August 28, 2020>

[0062] (Training data DB124) The training data DB124 is a database that stores the training data to be input to the model. Specifically, the training data DB124 can store data as training data in which the specific steps of an English writing process, according to a grammar table that shows the structural rules of English writing, are represented as flowcharts. In addition, the training data DB124 can also use other data as training data, such as image data showing the grammar table, or system prompts that represent flowcharts in text.

[0063] (Control unit 130) Returning to Figure 3, let's continue the explanation. The control unit 130 is realized by a processor, MPU (Micro Processing Unit), CPU (Central Processing Unit), etc., executing various programs stored in the memory unit 120 using RAM as a working area. The control unit 130 is also realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). As shown in Figure 3, the control unit 130 has a receiving unit 131, a specifying unit 132, a generation unit 133, an output unit 134, and a learning unit 135.

[0064] (Reception desk 131) The reception unit 131 receives English text input from the user via the input unit or communication unit 110 described above. Specifically, the reception unit 131 receives English text input based on text input, voice input, or image recognition input.

[0065] Here, we will explain an example of the reception processing flow by the reception unit 131. First, the English learning support device 100 (for example, the output unit, etc.) displays to the user the English text to be used as an exercise problem, which has been pre-registered in the assignment information DB 121. Next, the reception unit 131 accepts the input of the English text selected by the user from the displayed assignment text. Then, the reception unit 131 transmits the received English text to the identification unit 132.

[0066] Furthermore, the reception unit 131 can also accept English text in formats other than the selected English text as described above. For example, the reception unit 131 can accept English text entered in text format via the user's terminal device, etc. The reception unit 131 can also accept English text entered by the user's voice via the user's terminal device or voice input device, etc. The reception unit 131 can also accept image data acquired by the user's terminal device, an optical reading device such as a scanner, or an imaging device such as a camera, after converting it into text based on technologies such as OCR (Optical Character Recognition / Reader).

[0067] (Specific Section 132) The identification unit 132 identifies the attributes of words contained in an English sentence by inputting the English sentence into a model trained on a system diagram (grammar table) that defines the rules of sentence structure based on word order, tense, and verbs, which are parts of speech that define the actions or states of things in English. In this embodiment, "word attributes" refer to items such as the following. Each item will be explained again in the section describing Figures 15 to 17, which will be described later. Category (part of speech / verb pattern) type (constituent unit) detail (function / modification relationship) tense (tense and voice) object / complement verbal chart category / analysis Punctuation

[0068] Specifically, the identification unit 132 identifies the attributes of words contained in an English sentence by inputting an English sentence to a model, which is a large-scale language model that assigns labels indicating the attributes of words identified according to a specific flow of English sentence structure shown by a grammar table stored in the learning data DB 124.

[0069] For example, the identification unit 132 inputs the English sentence selected by the user to a large-scale language model that is provided with learning data or system prompts indicating a specific flow of the English sentence structure described above as prior knowledge. Next, the identification unit 132 obtains label information from the large-scale language model, which is in which labels indicating predetermined attributes are assigned to the words contained in the English sentence selected by the user.

[0070] Furthermore, the identification unit 132 uses a large-scale language model trained to perform the above-mentioned label assignment process by assigning labels to words in the English sentence that correspond to attributes selected from among the attributes of parts of speech that may appear from the beginning of the English sentence up to the Verb, focusing on the Verb contained in the English sentence.

[0071] Furthermore, the large-scale language model used by the specific unit 132 may be tuned by pre-inputting training data, inputting system prompts, or inputting prompts along with English text each time.

[0072] For example, in the case of "pre-input of training data," the specific unit 132 can use a large-scale language model for which training data such as image data showing a grammar table or a flowchart showing a specific flow of English sentence construction based on the grammar table is provided as prior knowledge.

[0073] Furthermore, in the case of "inputting a system prompt," the specific unit 132 can use a large-scale language model in which system prompts such as text describing a grammar table or text describing a specific flow of English sentence structure based on a grammar table are set.

[0074] Furthermore, in the case of "inputting prompts along with the English text each time," the specific unit 132 can input prompts such as text explaining the grammar table or text explaining the specific flow of the English sentence structure based on the grammar table, along with the target English text, into the large-scale language model and execute specific processing.

[0075] Here, using a large-scale language model in which a specific flow of English sentence structure based on a grammar table is set as prior knowledge, the processing flow executed by the identification unit 132 will be explained using Figures 6 to 14. Figures 6 to 14 show an example of a specific flow of English sentence structure according to the embodiment.

[0076] First, the "Subjective part" will be explained using Figure 6. When an English sentence is input, the identification unit 132 starts the "Subjective Part" (S101). Here, based on whether the beginning of the sentence is an Adverb related to a Verb, the identification unit 132 performs the following processing (S102).

[0077] If the beginning of the sentence is not an Adverb relating to a Verb (No. in S102), the identification unit 132 proceeds to the next step. On the other hand, if the beginning of the sentence is an Adverb relating to a Verb and the Adverb is "word" (Yes (word) in S102), the identification unit 132 identifies the Adverb as a word (S102a).

[0078] Furthermore, if the beginning of the sentence is an Adverb relating to a Verb, and the Adverb is a "clause / phrase" (Yes in S102), the identification unit 132 identifies the Adverb as a clause / phrase (S111). Note that the identification unit 132 executes a process relating to the "Verbal part" after the process in S111, and this process will be explained later using Figure 12.

[0079] Here, if the next Subject is not "Noun" (No in S103), the identification unit 132 proceeds to the next step. On the other hand, if the next Subject is "Noun" (Yes in S103), the identification unit 132 identifies the Noun as a clause / phrase (S121). After the step in S121, the identification unit 132 executes the steps related to the "Verbal part," which will be explained later using Figure 13.

[0080] The identification unit 132 repeatedly executes the processes described in S101 to S103 to search for the structure of the English text (S104). If the following Adjective is not an "inserted Adjective clause / phrase" (No. in S105), the identification unit 132 proceeds to the next step.

[0081] On the other hand, if the next Adjective is an "insertive Adjective clause / phrase" (Yes in S105), the identification unit 132 identifies it as an Adjective (clause / phrase) (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0082] If the next Adjective is not an "insertion Adjective clause" (No. in S106), the specific unit 132 starts the "Predicate Part" as the next step (S201).

[0083] On the other hand, if the next Adjective is an "insertion Adjective phrase" (Yes in S106), the identification unit 132 identifies it as Adjective(phrase) (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0084] Next, the "Predicate Part" will be explained using Figure 7. The specific unit 132 starts the "Predicate Part" when the processing of the "Subjective Part" is completed (S201).

[0085] Here, if the Verb is in the passive voice (Yes in S202), the specific part 132 refers to the Syntax(PV)-Predicate Part flow (S202a). Note that the Syntax(PV)-Predicate Part flow described above will not be explained in this embodiment.

[0086] On the other hand, if the Verb is not in the passive voice (No. in S202), the identification unit 132 identifies the pattern of the Verb (S203). Here, there are five patterns for the Verb: V1 (S211), V2 (S221), V3 (S231), V4 (S241), and V5 (S251). Therefore, the following sections will individually explain the flow for each pattern from V1 to V5.

[0087] If it is identified as V1(SV) (S211) and the end of the sentence is not an Adverb (No. in S204), the identification unit 132 terminates the English sentence structure identification flow.

[0088] On the other hand, if V1(SV) is identified (S211) and the end of the sentence is an Adverb (word) (Yes (word) in S204), the identification unit 132 identifies the Adverb as "word" (S204b). Then, the identification unit 132 terminates the English sentence structure identification flow.

[0089] Furthermore, if V1(SV) is identified (S211) and the end of the sentence is the Adverb "clause / phrase" (Yes (clause) in S204), the identification unit 132 identifies the Adverb as a clause / phrase (S111). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S111, and this process will be explained later using Figure 12.

[0090] The "V2 Part," which is executed when Verb is identified as V2(SVC), will be explained using Figure 8. First, the identification unit 132 identifies Verb as V2(SVC) (S221). Next, the identification unit 132 performs the identification of Complement (S222).

[0091] If Complement is an Adjective, the identification unit 132 identifies it as an Adjective (S223). Then, the identification unit 132 terminates Part V2.

[0092] On the other hand, if Complement is Noun, the identification unit 132 identifies it as Noun (S224). Here, if what follows Noun is not an "insertion Adjective clause / phrase" (No in S225), the identification unit 132 terminates V2 Part.

[0093] On the other hand, if what follows the Noun is an "insertive Adjective clause / phrase" (Yes in S225), the identification unit 132 identifies the Adjective as a clause / phrase (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0094] After completing Part V2, the specific unit 132 performs the same processing as after Part V1 is completed. Once the processing is complete, the specific unit 132 terminates the English structure specific flow.

[0095] Returning to Figure 7, we will continue the explanation. We will explain the "V3 Part" which is executed when Verb is identified as V3(SVO) using Figure 9. First, the identification unit 132 identifies Verb as V3(SVO) (S231).

[0096] Here, if the Object is Noun(clause / phrase) (Yes in S232), the identification unit 132 identifies it as Noun(clause / phrase) (S233). Then, the identification unit 132 terminates Part V3.

[0097] On the other hand, if the Object is not a Noun(clause / phrase) (No. in S232), the identification unit 132 identifies it as a Noun(word) (S234). Here, if what follows the Noun is not an "insertion Adjective clause / phrase" (No. in S235), the identification unit 132 terminates Part V3.

[0098] On the other hand, if what follows the Noun is an "insertive Adjective clause / phrase" (Yes in S235), the identification unit 132 identifies the Adjective as a clause / phrase (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0099] After completing Part V3, the specific unit 132 performs the same processing as after Part V1 is completed. Once the processing is complete, the specific unit 132 terminates the English structure specific flow.

[0100] Returning to Figure 7, we will continue the explanation. We will explain the "V4 Part" which is executed when Verb is identified as V4(SVOO), using Figure 10. First, the identification unit 132 identifies Verb as V4(SVOO) (S241).

[0101] Here, if the Object (person) is a Noun (clause / phrase) (Yes in S242), the identification unit 132 identifies it as a Noun (clause / phrase) (S243).

[0102] Here, if the Object (person) is not a Noun (clause / phrase) (No in S242), the identification unit 132 identifies it as a Noun (word) (S244). On the other hand, if what follows the Noun is an "inserted Adjective clause / phrase" (Yes in S245), the identification unit 132 identifies the Adjective as a clause / phrase (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, but this process will be explained later using Figure 14.

[0103] Then, if it is identified as Noun(clause / phrase) (S243), or if what follows Noun(word) is not an "inserted Adjective clause / phrase" (No. in S235), the identification unit 132 performs the following processing.

[0104] Here, if the Object is a Noun(clause / phrase) (Yes in S246), the identification unit 132 identifies it as a Noun(clause / phrase) (S247). Then, the identification unit 132 terminates Part V4.

[0105] On the other hand, if the Object is not a Noun (clause / phrase) (No in S246), the identification unit 132 identifies it as a Noun (word) (S248). Here, if what follows the Noun is an "inserted Adjective clause / phrase" (Yes in S249), the identification unit 132 identifies the Adjective as a clause / phrase (S131). The identification unit 132 then executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0106] Furthermore, if the Noun is not followed by an "insertive Adjective clause / phrase" (No. in S249), the specific section 132 terminates Part V4.

[0107] After completing Part V4, the specific unit 132 performs the same processing as the processing after Part V1 is completed. Once the processing is complete, the specific unit 132 terminates the specific flow for the English structure.

[0108] Returning to Figure 7, we will continue the explanation. We will explain the "V5 Part" which is executed when Verb is identified as V5(SVOC) using Figure 11. First, the identification unit 132 identifies Verb as V5(SVOC) (S251).

[0109] Here, if the Object is Noun(clause / phrase) (Yes in S252), the identification unit 132 identifies it as Noun(clause / phrase) (S253).

[0110] Here, if the Object is not a Noun(clause / phrase) (No in S252), the identification unit 132 identifies it as a Noun(word) (S254). On the other hand, if what follows the Noun is an "insertion Adjective clause / phrase" (Yes in S255), the identification unit 132 identifies the Adjective as a clause / phrase (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, but this process will be explained later using Figure 14.

[0111] Then, if it is identified as Noun(clause / phrase) (S253), or if what follows Noun(word) is not an "insertive Adjective clause / phrase" (No. in S255), the identification unit 132 identifies Complement (S256).

[0112] Here, if Complement is an Adjective, the identification unit 132 identifies it as an Adjective (S257). Then, the identification unit 132 terminates Part V5.

[0113] On the other hand, if Complement is Noun, the identification unit 132 identifies it as Noun (S258). Here, if what follows Noun is not an "insertion Adjective clause / phrase" (No in S259), the identification unit 132 terminates Part V5.

[0114] Furthermore, if the Noun is followed by an "insertive Adjective clause / phrase" (Yes in S259), the identification unit 132 identifies the Adjective as a clause / phrase (S131). Note that the identification unit 132 executes a process related to the "Verbal part" after the process in S131, and this process will be explained later using Figure 14.

[0115] After completing Part V5, the specific unit 132 performs the same processing as the processing after Part V1 is completed. Once the processing is complete, the specific unit 132 terminates the English structure specific flow.

[0116] From here, we will explain the Verbal Part processes from "S111," "S121," and "S131" onward in Figures 6 to 11 mentioned above. First, we will explain the "Verbal Part (Adverb (clause / phrase))" after identifying the Adverb as clause / phrase using Figure 12.

[0117] The identification unit 132 identifies the Adverb as a clause / phrase (S111). Here, the identification unit 132 determines whether the identified Adverb is a clause or a phrase (S112).

[0118] If the identified Adverb is a clause (the clause in S112), the identification unit 132 identifies the type of clause (subordinate, conjunction, etc.) (S112a). Next, the identification unit 132 starts the Subjective Part (executing the steps from S101 onwards in Figure 6). Then, after the Subjective Part is completed, the identification unit 132 starts the Predicate Part (executing the steps from S201 onwards in Figure 7). Finally, after the Predicate Part is completed, the identification unit 132 terminates the English sentence structure identification flow.

[0119] On the other hand, if the identified Adverb is a phrase (the phrase in S112), the identification unit 132 identifies the type of phrase (infinitive, participle structure, preposition + noun, etc.) (S122b). Next, the identification unit 132 determines whether to execute the Subjective Part (S113).

[0120] If the Subjective Part is to be executed (Yes in S113), the specific unit 132 starts the Subjective Part (executes the steps from S101 onwards in Figure 6). Then, after the Subjective Part is completed, the specific unit 132 starts the Predicate Part (executes the steps from S201 onwards in Figure 7).

[0121] On the other hand, if the Subjective Part is not executed (No. in S113), the identification unit 132 skips the Subjective Part and starts the Predicate Part (executing the steps from S201 onwards in Figure 7). Then, after the Predicate Part is completed, the identification unit 132 terminates the English composition identification flow.

[0122] Next, using Figure 13, we will explain the "Verbal Part(Noun(clause / phrase))" after identifying the Noun as a clause / phrase.

[0123] The identification unit 132 identifies Noun as a clause / phrase (S121). Here, the identification unit 132 determines whether the identified Noun is a clause or a phrase (S122).

[0124] If the identified Noun is a clause (the clause in S122), the identification unit 132 identifies the type of clause (that, if / whether, 5W1H / question words, relative pronouns including antecedents, etc.) (S122a). Next, the identification unit 132 starts the Subjective Part (executing the steps from S101 onwards in Figure 6). Next, after the Subjective Part is completed, the identification unit 132 starts the Predicate Part (executing the steps from S201 onwards in Figure 7). Finally, after the Predicate Part is completed, the identification unit 132 terminates the English sentence structure identification flow.

[0125] On the other hand, if the identified Noun is a phrase (the phrase in S122), the identification unit 132 identifies the type of phrase (infinitive, particular gerund, 5W1H infinitive, etc.) (S122b). Next, the identification unit 132 determines whether to execute the Subjective Part (S123).

[0126] If the Subjective Part is to be executed (Yes in S123), the specific unit 132 starts the Subjective Part (executes the steps from S101 onwards in Figure 6). Then, after the Subjective Part is completed, the specific unit 132 starts the Predicate Part (executes the steps from S201 onwards in Figure 7).

[0127] Furthermore, if the Subjective Part is not executed (No. in S123), the identification unit 132 skips the Subjective Part and starts the Predicate Part (executing the steps from S201 onwards in Figure 7). After the Predicate Part is completed, the identification unit 132 terminates the English composition identification flow.

[0128] Next, using Figure 14, we will explain the "Verbal Part (Adjective (clause / phrase))" after identifying the Adjective as a clause / phrase.

[0129] The identification unit 132 identifies it as an Adjective (S131). Here, the identification unit 132 determines whether the identified Adjective is a clause or a phrase (S132).

[0130] If the identified Adjective is a clause (the clause in S132), the identification unit 132 identifies the type of clause (relative pronoun, relative Adverb, etc.) (S132a). Next, the identification unit 132 starts the Subjective Part (executing the steps from S101 onwards in Figure 6). After the Subject Part is completed, the identification unit 132 starts the Predicate Part (executing the steps from S201 onwards in Figure 7). Finally, after the Predicate Part is completed, the identification unit 132 terminates the English sentence structure identification flow.

[0131] On the other hand, if the identified Adjective is a phrase (phrase in S132), the identification unit 132 identifies the type of phrase (infinitive, participle (present / past), preposition + noun, apposition noun, etc.) (S132b). Next, the identification unit 132 determines whether to execute the Subjective Part (S133).

[0132] If the Subjective Part is to be executed (Yes in S133), the specific unit 132 starts the Subjective Part (executes the steps from S101 onwards in Figure 6). Then, after the Subjective Part is completed, the specific unit 132 starts the Predicate Part (executes the steps from S201 onwards in Figure 7).

[0133] Furthermore, if the Subjective Part is not executed (No. in S133), the identification unit 132 skips the Subjective Part and starts the Predicate Part (executing the steps from S201 onwards in Figure 7). After the Predicate Part is completed, the identification unit 132 terminates the English composition identification flow.

[0134] (Generation unit 133) Returning to Figure 3, the explanation continues. The generation unit 133 performs syntactic analysis of the input English sentence based on text input, voice input, or image recognition input, and generates tasks for English learning according to the grammar table.

[0135] For example, the generation unit 133 obtains label information assigned by the identification unit 132 to words contained in the English text entered by the user. Next, the generation unit 133 generates correspondence information that associates words with grammar tables based on the obtained label information. Then, the generation unit 133 generates tasks using a large-scale language model that generates predetermined tasks using the generated correspondence information. Finally, the generation unit 133 stores the generated tasks in the task information DB 121.

[0136] The above tasks may include, for example, a problem requiring students to specify where a given word is located in a grammar table.

[0137] (Output section 134) The output unit 134 outputs mapping information to the user, which associates the words contained in the English text with the position of those words in the grammar table, using the attributes of the words indicated by the label information.

[0138] Here, an example of an output command for realizing output processing by the output unit 134 will be explained using Figures 15 to 17. Figures 15 to 17 are diagrams showing an example of an output command according to the embodiment. In this embodiment, the JSON format is given as an example, but it is not limited to this, and output may be performed based on other known technologies.

[0139] The "Basic Structure (Figure 15 (1))" contains instructions to analyze the input English text, add grammatical information indicating predetermined attributes, and output it in JSON format.

[0140] Specifically, the English text to be analyzed is inserted into "sentence." Then, processing is performed on the inserted English text according to the contents described in "breakdown." Here, "breakdown" breaks down the English text to be analyzed into constituent elements such as words, phrases, and clauses, and assigns predetermined attributes (grammatical information) including "type," "detail," "category," "tense," "object / Complement," "verbal chart category / analysis," and "handling of punctuation."

[0141] Here, the aforementioned specified attributes will be explained using Figure 16. Figure 16 shows an example of a directive that defines "type (Figure 16 (1))", "detail (Figure 16 (2))", "category (Figure 16 (3))", "tense (Figure 16 (4))", "object / complement (Figure 16 (5))", "verbal chart category / analysis (Figure 16 (6))", and "Punctuation handling (Figure 16 (7))".

[0142] "Type (Figure 16 (1))" specifies the type of constituent element, and includes, for example, "word," "phrase," "clause," etc. Note that "words" applies to proper nouns and compound nouns.

[0143] "Detail (Figure 16 (2))" contains instructions that specify its role in the English text, such as "Subject," "modifies noun," "modifies verb," ​​or "sub-part (auxiliary verb or auxiliary verb within a verb phrase)."

[0144] "Category (Figure 16 (3))" defines specific parts of speech categories, including, for example, "noun," "adjective," "adverb," ​​"article / possessive adjective," "conjunction," "preposition," "pronoun (objective case)," and "punctuation." "Verbs" are represented in a format that identifies five patterns from V1 to V5 (for example, V3 is represented as "verb as v3").

[0145] "tense (Figure 16 (4))" contains instructions to specify the tense or voice of a verb, such as "past" and "present progressive voice." Additionally, for parts that are not main verbs, such as auxiliary verbs and other auxiliary verbs, there is an instruction to add "detail:sub-part" as "tense:(blank)."

[0146] The instruction "object / complement (Figure 16 (5))" includes a command to specify the object as "object" and the complement as "complement" when a verb takes an object or complement.

[0147] The "verbal chart category / analysis (Figure 16 (6))" command includes instructions for specifying the grammatical structure classification label (verbal chart category) and a concise functional description in Japanese (analysis). For example, the instruction for the grammatical structure classification label (verbal chart category) will result in the output data being written as "adverb phrase > preposition + noun," etc. Similarly, the instruction for a concise functional description in Japanese (analysis) will result in the output data being written as "Adverb phrase: M (modifier) ​​indicating time for the verb 'run'," etc.

[0148] "Handling of punctuation marks (Figure 16 (7))" includes instructions to output commas, periods, etc., as independent elements.

[0149] Next, we will explain the "separation process between verbs and adverbs" using Figure 17. Figure 17(1) shows the "separation process between verbs and adverbs," which, when a verb is separated by an adverb, instructs that the auxiliary verb part be written as detail:sub-part,tense:(blank), and that the complete tense information is written only for the main verb part.

[0150] In other words, the identification unit 132, when a verb in an English sentence is separated by an adverb, identifies the attributes of the words in the English sentence by assigning tense information to the main verb portion, with the portion corresponding to the auxiliary verb or auxiliary verb being a predetermined attribute. Then, the output unit 134 can generate output data with complete tense information only on the main verb portion, based on the identification result by the identification unit 132 and the command shown in (1) of Figure 17.

[0151] The output unit 134 inputs the labeled words generated by the identification unit 132 into a large-scale language model to which output commands (e.g., system prompts, etc.) shown in Figures 15 to 17 are applied, thereby generating JSON format data as shown in Figure 18. Figure 18 is a diagram showing an example of JSON format output according to the embodiment.

[0152] Then, using the JSON data described above, the output unit 134 displays the English learning screen on the user's terminal device. Specifically, the output unit 134 uses the generated JSON data to generate display data that includes instructions for displaying the English learning screen on the user's terminal device.

[0153] For example, the output unit 134 generates display data for displaying mapping information to the user based on known technologies such as JavaScript (registered trademark). The output unit 134 then outputs the generated display data to the user's terminal device, etc., so that the English learning screen can be displayed on the terminal device, etc. Furthermore, by utilizing the above-mentioned known technologies such as JavaScript (registered trademark), the output unit 134 can realize dynamic displays, such as switching the highlighting of the grammar table according to the user's selection.

[0154] Here, an example of the English learning screen display will be explained using Figure 19. Figure 19 is a diagram showing an example of the output of the English learning screen according to the embodiment. Figure 19 shows "grammar table and English sentence correspondence information (Figure 19 (1))", "reading aloud function (Figure 19 (2))", "explanation function (Figure 19 (3))", and "word / phrase list function (Figure 19 (4))".

[0155] First, let's explain the "correspondence information between the grammar table and the English text (Figure 19 (1))." As described above, the output unit 134 determines the position of the word in the English text in the grammar table according to the attributes of the identified word, as output to the user. The output unit 134 then displays correspondence information on the same screen, which associates a predetermined chart highlighting the determined position in the grammar table with the word in the English text.

[0156] For example, "Before the cafe's expansion" shown in (1-1a) of Figure 19 is positioned as an "adverb (Figure 19 (1-1b))" ​​in the grammar table. Also, "a small number" shown in (1-2a) of Figure 19 is positioned as an "adjective + subject (noun) + insertion (Figure 19 (1-2b))" in the grammar table.

[0157] Figure 19 (1-3a) shows that "could access" is located in the "Verb + V3 (Figure 19 (1-3b))" category on the grammar table. Also, Figure 19 (1-4a) shows that "the seating areas" is located in the "O (Figure 19 (1-4b))" category on the grammar table. Furthermore, Figure 19 (1-5a) shows that "at the same time" is located in the "Adverb (Figure 19 (1-5b))" category on the grammar table.

[0158] Note that the grammar table shown in Figure 19(1) is a simplified representation of the actual grammar table. Therefore, the functions of the grammar table shown in Figure 19(1) correspond to the functions of the grammar table shown in Figure 1, and each has the following meaning. Adverb (word) sub-phrase: adverb(phrase) Subclause: adverb(clause) Adjective (word) Subject (noun); Subject (noun) Insertion: parenthesis (modifies nouns, etc.) Verb + Vn: Verb and Verb Patterns C:Complement O:Object

[0159] English text may have a hierarchical structure based on a Main Verb and Verbal, with the Verb as its Core. In this case, if the output unit 134 selects a first hierarchical structure, which is determined based on Verbs, from among the hierarchical structures of English text based on combinations of main clauses, clauses, or phrases, it will display the first English text included in the first hierarchical structure and a predetermined diagram corresponding to the first English text in association with it. Furthermore, if the output unit 134 selects a second hierarchical structure, which is contained within the first hierarchical structure and whose structure is determined based on Verbs, it will display the second English text included in the second hierarchical structure and a predetermined diagram corresponding to the second English text in association with it.

[0160] For example, in the main clause (first hierarchical structure) shown in Figure 19 (1), "Before the cafe's expansion, a small number could access the seating areas at the same time," the adverbial phrase "Before the cafe's expansion" is included as the second hierarchical structure.

[0161] Therefore, if the entire main text (first hierarchical structure) is selected, the output unit 134 displays the correspondence with the grammar table for the entire main text, and if a second hierarchical structure included in the entire main text is selected, it switches to displaying the correspondence with the grammar table for the second hierarchical structure.

[0162] In this embodiment, we refer to the structure as a first hierarchical structure and a second hierarchical structure, but the first hierarchical structure may include multiple second hierarchical structures. Furthermore, the second hierarchical structure may include lower-level hierarchical structures such as a third hierarchical structure, a fourth hierarchical structure, and so on up to the nth hierarchical structure.

[0163] Next, the "read-aloud function (Figure 19 (2))" will be explained. For example, when the user presses the "read-aloud" button shown in Figure 19 (2-1), the output unit 134 displays the words contained in the English sentence in correspondence with their positions in the grammar table, by sequentially changing the display format of the text representing the English sentence from the beginning to the end of the sentence. The output unit 134 can also provide the user with the above-mentioned read-aloud function by arbitrarily changing the display speed, character color, character display format, etc.

[0164] For example, the output unit 134 can sequentially change the text display format from the beginning of the sentence to the end of the sentence for the English sentence shown in Figure 19 (2), "Before the cafe's expansion, a small number could access the seating areas at the same time.", so that the user can grasp the speed at which the sentence is being read aloud.

[0165] Furthermore, as described above, when sequentially changing the display format of the text representing the English sentence from the beginning to the end, the output unit 134 may also simultaneously display the correspondence with the grammar table. That is, the output unit 134 can display the location of the next word to be read aloud in the grammar table in accordance with the change in character color based on the reading aloud function.

[0166] Furthermore, when the reception unit 131 receives spoken English text input by the user, the output unit 134 can display to the user the difference between the user's spoken text and the changes made to the text display format by the output unit 134. For example, the output unit 134 can simultaneously display the changes to the display format of the first text based on user input and the changes to the display format of the second text made by the output unit 134 in different display formats to clearly show the difference to the user.

[0167] Next, we will explain the "explanation function (Figure 19 (3))." For example, when the user presses the "explanation" button shown in Figure 19 (3-1), the output unit 134 displays the explanation from the grammar table superimposed on the correspondence information. For example, as shown in Figure 19 (3), the output unit 134 can display an explanation based on the grammar table for the displayed English text to the user.

[0168] Next, we will explain the "Word / Phrase List Function (Figure 19 (4))." For example, when the user presses the "Word" button shown in Figure 19 (4-1), the output unit 134 overlays the classification of the word, the English-Japanese explanation of the word, or the English-English explanation of the word on the correspondence information. For example, as shown in Figure 19 (4), the output unit 134 can display a grammar table-based explanation of the word for the word contained in the displayed English sentence to the user.

[0169] (Learning Section 135) The learning unit 135 learns the large-scale language model by providing it with prior knowledge using training data or system prompts. In this embodiment, the learning unit 135 does not particularly limit the learning method and can learn various models, including the large-scale language model, based on known model learning means.

[0170] (Processing procedure according to the embodiment) From here, a series of processing steps implemented by the English learning support device 100 according to this embodiment will be described. Figure 20 is a flowchart of the processing of the English learning support device 100 according to this embodiment.

[0171] The English learning support device 100 waits for an English sentence to be entered (No in S301). Then, when an English sentence is entered (Yes in S301), the English learning support device 100 starts processing.

[0172] The English learning support device 100 inputs English text into a large-scale language model (S302). Next, the English learning support device 100 assigns labels indicating predetermined attributes to the words contained in the English text (S303). Then, the English learning support device 100 outputs correspondence information that associates the labeled words with the grammar table (S304).

[0173] If supplementary information is to be displayed (Yes in S305), the English learning support device 100 displays the supplementary information superimposed on the correspondence information (S306). In this embodiment, "supplementary information" may include, for example, a "reading aloud function (Figure 19 (2))", an "explanation function (Figure 19 (3))", or a "word / phrase list function (Figure 19 (4))".

[0174] On the other hand, if no supplementary information is displayed (No. in S305), the English learning support device 100 skips the process in S106. Then, the English learning support device 100 terminates the process.

[0175] (effect) From here, we will explain the effects of the English learning support device 100 according to this embodiment. Conventionally, there are challenges in appropriately supporting learning to acquire accurate English by accurately understanding the roles of parts of speech in English grammar, such as in English learning methods based on grammar tables, and eliminating the biases that Japanese native speakers have.

[0176] Therefore, the identification unit 132 of the English learning support device 100 according to the embodiment identifies the attributes of the words contained in the English sentence by inputting the English sentence into a model trained based on a grammar table. The output unit 134 of the English learning support device 100 outputs correspondence information to the user, which associates the words contained in the English sentence with the position of those words in the grammar table using the attributes of the words.

[0177] Therefore, the English learning support device 100 has the effect of enabling appropriate support for English learning. Furthermore, the English learning support device 100 achieves predetermined effects by performing the processes described below.

[0178] The identification unit 132 takes the input English text as input to a model, which is a large-scale language model that assigns labels indicating the attributes of words identified in accordance with the specific flow of English sentence structure shown by the grammar table to the words contained in the English text, and identifies the attributes of the words contained in the English text.

[0179] Through the process described above, the English learning support device 100 can appropriately display where each word in an English sentence is located in the grammar table by assigning a label based on the grammar table. As a result, the English learning support device 100 effectively assists the user in acquiring the ability to appropriately interpret English sentences according to the grammar table.

[0180] The specific unit 132 uses a large-scale language model trained to perform the following actions: assign labels to words in an English sentence that correspond to attributes selected from among the attributes of parts of speech that may appear from the beginning of the sentence to the position up to the verb, and assign labels to words in an English sentence that correspond to attributes selected from among the attributes of parts of speech that may appear from the verb to the end of the sentence, with the verb being the central focus.

[0181] Through the processing described above, the English learning support device 100 can appropriately perform the identification of verb-centered sentence structure on a large-scale language model based on a grammar table, which differs significantly from conventional English education. As a result, by performing appropriate processing on the large-scale language model, the English learning support device 100 can perform grammar table-based sentence interpretation for any English text.

[0182] The output unit 134 determines the position of the word in the English text in a grammar table according to the attributes of the identified word, as output to the user. The output unit 134 then displays correspondence information on the same screen, which associates the determined position in the grammar table with the word in the English text, in a predetermined diagram that highlights the word in the grammar table.

[0183] Through the processing described above, the English learning support device 100 can use words to which labels indicating predetermined attributes have been assigned by a large-scale language model to highlight and display the corresponding parts of the grammar table for those words. Furthermore, when the user changes the selection of English sentences included in the main text, the English learning support device 100 can dynamically switch the highlighting display in accordance with that change. In other words, the English learning support device 100 can clearly display to the user the correspondence between the words included in the English sentence selected by the user and the grammar table. As a result, the English learning support device 100 has the effect of enabling the user to acquire the ability to appropriately and easily understand English sentence structure in accordance with the grammar table.

[0184] The output unit 134, as output to the user, displays, in association with a predetermined diagram or chart corresponding to the first English sentence included in the first hierarchical structure, if a first hierarchical structure whose structure is determined based on verbs is selected from among the hierarchical structures of English sentences based on combinations of main clauses, clauses, or phrases. Furthermore, if a second hierarchical structure, which is contained within the first hierarchical structure and whose structure is determined based on verbs, is selected, the output unit 134 displays, in association with a predetermined diagram or chart corresponding to the second English sentence included in the second hierarchical structure.

[0185] Through the processing described above, the English learning support device 100 can appropriately display the hierarchical structure of English sentences that have a hierarchical structure consisting of a main verb and a core formed by other verbs, based on a grammar table. That is, when a main verb is selected, the English learning support device 100 displays the structure of the main verb based on a grammar table, and when a lower-level English sentence included in the main verb is selected, it displays the structure of the lower-level English sentence based on a grammar table. As a result, the English learning support device 100 has the effect of enabling the user to acquire the ability to accurately understand the hierarchical structure of English sentences based on a grammar table.

[0186] The reception unit 131 accepts English text input based on text input, voice input, or image recognition input. Through the above-described processing, the English learning support device 100 enables English text input by various means. For example, even when the user wishes to perform English text analysis based on a grammar table on an English text they have written themselves, the English learning support device 100 can accept the English text by various means such as text, voice, or image recognition, thereby improving user convenience.

[0187] The generation unit 133 performs syntactic analysis of the input English text based on text input, voice input, or image recognition input, and generates tasks for English learning according to a grammar table.

[0188] The above-described process enables the English learning support device 100 to freely create assignments using any English sentence other than those pre-registered. As a result, the English learning support device 100 allows users to learn with a high degree of flexibility.

[0189] The identification unit 132 identifies the attributes of the words in the English sentence by assigning tense information to the main verb portion, with the portion corresponding to the auxiliary verb or auxiliary verb being a predetermined attribute, when the verb in the English sentence is separated by an adverb.

[0190] The English learning support device 100, through the processing described above, has the effect of mechanically and consistently converting the sentence structure of English texts into JSON format, which can then be used for analysis, searching, or generating educational content.

[0191] When the output unit 134 displays words in an English sentence in correspondence with their positions in the grammar table, it sequentially changes the display format of the text representing the English sentence from the beginning to the end of the sentence.

[0192] Through the processing described above, the English learning support device 100 enables the user to practice reading aloud English sentences at an appropriate speed. Furthermore, during reading practice, the English learning support device 100 can display the location of the next word to be read aloud in the grammar table. As a result, the English learning support device 100 has the effect of enabling the user to learn English while being aware of the grammar table when practicing reading aloud.

[0193] The output unit 134 displays the grammar table explanations superimposed on the correspondence information. Through the above-described process, the English learning support device 100 enables users to efficiently learn English based on the grammar table by superimposing explanations of the grammar table and other information when the user learns English based on the correspondence display between words contained in English sentences and the grammar table.

[0194] The output unit 134 overlays the classification of the word contained in the English text, the English-Japanese explanation of the word, or the English-English explanation of the word onto the correspondence information. Through the above-described process, the English learning support device 100 enables the user to efficiently learn English based on the grammar table by overlaying explanations of English words based on the grammar table when the user learns English based on the correspondence display between the words contained in the English text and the grammar table.

[0195] <Variation> The following describes some modifications implemented by the English learning support device 100 according to this embodiment.

[0196] (Changing the highlighting method according to learning progress) The English learning support device 100 (output unit) changes the highlighting method in a predetermined chart where the position in the grammar table is highlighted, according to the user's learning progress. Specifically, based on the user's past learning progress, the English learning support device 100 (output unit) can perform controls such as highlighting with a different color, changing the size, displaying additional speech bubbles, or emitting sounds for areas where the user's incorrect answer rate exceeds a threshold.

[0197] For example, the English learning support device 100 (output unit) can change the highlighting method to one suitable for the user according to their English proficiency, learning status, qualifications, psychographic data, or demographic data. In order to realize the above function, the English learning support device 100 (output unit) may use, for example, a model (large-scale language model) that has been trained to identify a highlighting method appropriate for a given user when information such as the target user's English proficiency, learning status, qualifications, psychographic data, or demographic data is input.

[0198] Through the processing described above, the English learning support device 100 enables users to easily identify their weak points by highlighting and displaying the areas where they struggle when they are learning English based on the correspondence display between words contained in English sentences and grammar tables. As a result, the English learning support device 100 has the effect of enabling users to efficiently learn English based on grammar tables.

[0199] (Data, etc.) The terms used in the description of the above embodiment, such as grammar table, Verb, Syntax, Verbal, and the names of the functional parts of the English learning support device 100, steps, processes, and names of steps or processes, are merely examples and can be changed at will.

[0200] (Regarding the use of the model) In this embodiment, the model (large-scale language model) used by the English learning support device 100 is described as being stored in the model DB 123 of the storage unit 120, but this is not limited to this. For example, the English learning support device 100 can access an external information processing device (server, etc.) and use a predetermined model.

[0201] (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.

[0202] (Systems, etc.) Of the processes described in the embodiments and modifications described above, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0203] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need 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, usage conditions, etc.

[0204] The aforementioned components include those that are easily conceivable by those skilled in the art, those that are substantially identical, and those that fall within the so-called equivalent range. Furthermore, the embodiments and modifications described above can be combined as appropriate, as long as the processing content is not contradictory.

[0205] Furthermore, the terms "section," "module," and "unit" mentioned above can be replaced with "means" or "circuit," etc. For example, a control unit can be replaced with a control means or a control circuit.

[0206] Although several embodiments have been described in detail above with reference to the drawings, these are merely examples, and it is possible to implement these embodiments in various modified and improved forms based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0207] <Hardware Configuration> The device included in the English learning support device 100 according to this embodiment is implemented by a computer 1000 having a configuration as shown in Figure 21. Figure 21 is a diagram showing an example of the hardware configuration of a computer that implements the English learning support device 100 according to this embodiment.

[0208] The computer 1000 has a configuration in which a CPU 1100, memory 1200, auxiliary storage device 1300, input interface 1400, output interface 1500, and communication interface 1600 are connected by a bus 1700.

[0209] The CPU 1100 operates based on programs stored in the memory 1200 or auxiliary storage device 1300, and controls each functional unit. The memory 1200 consists of, for example, RAM (Random Access Memory) or ROM (Read Only Memory), and stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0210] For example, when the computer 1000 functions as the English learning support device 100 according to this embodiment, the CPU 1100 of the computer 1000 can realize the functions of the control unit 130 by executing a program loaded on the memory 1200.

[0211] The auxiliary storage device 1300 stores programs executed by the CPU 1100, as well as data used by such programs. The CPU 1100 controls input devices 1410, such as keyboards and mice, via the input interface 1400. The CPU 1100 also acquires data from the input devices 1410 via the input interface 1400.

[0212] The CPU 1100 controls output devices 1510, such as displays and printers, via the output interface 1500. The CPU 1100 also outputs generated data to the output devices 1510 via the output interface 1500.

[0213] The communication interface 1600 receives data from other devices via a predetermined network NW and sends it to the CPU 1100, and the CPU 1100 transmits the generated data to other devices via the predetermined network NW. [Explanation of Symbols]

[0214] 100 English Learning Support Devices 110 Communications Department 120 Storage section 121 Issue Information Database 122 User Information Database 123 Model DB 124 Training Data Database 130 Control Unit 131 Reception Department 132 Specific part 133 Generation part 134 Output section 135 Learning Department

Claims

1. A specification unit that identifies the attributes of words contained in an English sentence by inputting an English sentence into a model trained on a system diagram that defines the rules of sentence structure based on verbs, which are parts of speech that define word order, tense, and the action or state of things in English; An output unit that outputs correspondence information to the user, which associates the words contained in the aforementioned English text with the position of the said words in a system diagram showing the structural rules of the aforementioned English text, using the attributes of the said words. An English learning support device characterized by having the following features.

2. The specified part is, The model is a large-scale language model that assigns labels indicating the attributes of words identified in accordance with a specific flow of English sentence structure shown by a system diagram illustrating the rules for constructing the aforementioned English sentence, to the words contained in the English sentence. The input English text is used to identify the attributes of the words contained in the English text. The English learning support device according to feature 1.

3. The specified part is, The process involves assigning labels to words in the English sentence that correspond to attributes selected from among the attributes of parts of speech that may appear from the beginning of the sentence up to the Verb contained in the English sentence. And, Centered around the Verb contained in the English sentence, the label is assigned to the words in the English sentence that correspond to an attribute selected from among the attributes of the part of speech determined by the Verb, and that can appear from the Verb to the end of the English sentence. Using the large-scale language model trained to perform the following: The English learning support device according to feature 2.

4. The output unit is, The output to the aforementioned user is as follows: According to the attributes of the identified word, the position of the word contained in the English sentence in the diagram showing the rules for the structure of the English sentence is determined. A predetermined diagram showing the position of a specific element in a system diagram illustrating the determined structural rules of the English sentence, and the corresponding information relating the words contained in the English sentence are displayed on the same screen. An English learning support device according to any one of features 1 to 3.

5. The output unit is, The output to the aforementioned user is as follows: Of the hierarchical structures of the aforementioned English sentences based on combinations of main clauses, clauses, or phrases, If a first hierarchical structure whose configuration is determined based on the Verb is selected, the first English sentence included in the first hierarchical structure and the predetermined diagram corresponding to the first English sentence are displayed in association with each other. If a second hierarchical structure, which is contained within the first hierarchical structure and whose configuration is determined based on the Verb, is selected, the second English text included in the second hierarchical structure and the predetermined diagram corresponding to the second English text are displayed in association. The English learning support device according to feature 4.

6. The output unit is, When displaying the words contained in the aforementioned English sentence in correspondence with their positions in a diagram showing the structural rules of the aforementioned English sentence, the display format of the text representing the English sentence is sequentially changed from the beginning to the end of the sentence. The English learning support device according to feature 4.

7. The output unit is, The aforementioned correspondence information is overlaid with an explanation of the systematic diagram showing the structural rules of the English text. The English learning support device according to feature 4.

8. The output unit is, The aforementioned correspondence information is superimposed with the classification of the word contained in the English text, the Japanese-English explanation of the word, or the English-English explanation of the word. The English learning support device according to feature 4.

9. The output unit is, Depending on the user's learning progress, the method of highlighting in a predetermined diagram that highlights the position of the structural rules of the English text is changed. The English learning support device according to feature 4.

10. The system further includes a reception unit that receives the aforementioned English text entered based on text input, voice input, or image recognition input. An English learning support device according to any one of features 1 to 3.

11. The system further comprises a generation unit that performs syntactic analysis of English sentences input based on text input, voice input, or image recognition input, and generates tasks for English learning according to a systematic diagram showing the structural rules of the English sentences. An English learning support device according to any one of features 1 to 3.

12. The specified part is, If the Verb contained in the aforementioned English sentence is separated by an adverb, the part corresponding to the auxiliary verb or auxiliary verb is designated as a predetermined attribute, and tense information is assigned to the main verb part to identify the attributes of the words contained in the aforementioned English sentence. An English learning support device according to any one of features 1 to 3.

13. A method of supporting English language learning that is executed by a computer, A procedure for identifying the attributes of words contained in an English sentence by inputting an English sentence into a model trained on a system diagram that defines the rules of sentence structure based on verbs, which are parts of speech that define word order, tense, and the action or state of things in English. Output procedure for outputting correspondence information to the user, which associates the words contained in the aforementioned English text with the position of the said words in a system diagram showing the structural rules of the aforementioned English text, using the attributes of the said words. An English learning support method characterized by including [this].

14. The process involves inputting English text into a model trained based on a system diagram that defines the rules of English sentence structure based on word order, tense, and verbs (parts of speech that define the action or state of things in English), thereby identifying the attributes of the words contained in the English text. Output process: Outputs to the user correspondence information that associates the words contained in the English text with the position of the words in a system diagram showing the structural rules of the English text, using the attributes of the words. An English learning support program characterized by having a computer execute it.

Images