Information processing method, program, and information processing device

Abstract

To provide a new mechanism capable of using a correction result outputted by an AI for subsequent learning in an online learning service with the AI introduced.SOLUTION: An information processing method for making an information processing device execute: acquiring sentence data about a sentence created by a user; inputting the sentence data and a correction instruction to a generation AI; acquiring a correction result of the sentence data from the generation AI; extracting a prescribed part from the correction result on the basis of a prescribed extraction condition about correction; and associating the prescribed part with identification information of the user to record them.SELECTED DRAWING: Figure 17

Description

[Technical Field]

[0001] The disclosed technology relates to an information processing method, a program, and an information processing device. [Background technology]

[0002] There are known learning services that users can use online. For example, Patent Document 1 discloses that a user (learner) and a teacher communicate with each other regarding writing correction and guidance via a terminal device connected to a network. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-305943 Summary of the Invention [Problem to be solved by the invention]

[0004] In recent years, there has also been an increase in online learning services that incorporate generative AI (artificial intelligence). For example, instead of a human instructor, generative AI can correct input answers, thereby improving learning efficiency.

[0005] However, although the correction results by the generation AI are output to the user, there is no mechanism in place to allow the user to use the correction results for learning purposes when writing sentences later.

[0006] Therefore, the disclosed technology aims to provide a new mechanism in online learning services that incorporate AI, which enables the correction results output by AI to be effectively used when writing subsequent documents. [Means for solving the problem]

[0007] An information processing method that is one aspect of the disclosed technology includes an information processing device that acquires text data related to a text created by a user, inputs the text data and correction instructions into a generation AI, acquires correction results for the text data from the generation AI, extracts a specified portion from the correction results based on specified extraction conditions related to correction, and records the specified portion in association with the user's identification information. [Effects of the Invention]

[0008] According to the disclosed technology, in an online learning service that uses AI, the correction results output by the AI can be effectively used when writing later documents. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of an information processing system according to an embodiment. [Figure 2] FIG. 2 is a block diagram illustrating an example of a server according to an embodiment. [Figure 3] FIG. 10 is a diagram illustrating an example of a screen for setting a level related to a user according to an embodiment. [Figure 4] FIG. 10 is a diagram showing an example of a screen for setting a character of a generated AI according to an embodiment. [Figure 5] FIG. 10 is a diagram illustrating an example of a screen for inputting a sentence according to an embodiment. [Figure 6] FIG. 10 is a diagram illustrating an example of a screen on which a sentence is input according to an embodiment. [Figure 7] FIG. 10 is a diagram showing an example of a correction instruction. [Figure 8] FIG. 10 is a diagram showing an example of a correction result. [Figure 9] FIG. 10 is a diagram illustrating an example of a screen displaying a correction result according to an embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of a screen on which a text theme is presented according to an embodiment. [Figure 11] FIG. 10 is a diagram showing an example of an instruction to generate a theme of a text. [Figure 12] FIG. 10 is a diagram showing an example of a screen on which a message prompting the user to create a sentence is displayed. [Figure 13] FIG. 10 is a diagram showing an example of an instruction to generate a message prompting the user to create a sentence. [Figure 14] FIG. 10 is a diagram showing an example of a question creation instruction. [Figure 15] FIG. 10 is a diagram illustrating an example of user information according to an embodiment. [Figure 16] FIG. 1 illustrates an example of a processing apparatus according to an embodiment. [Figure 17] FIG. 10 is a sequence diagram illustrating an example of processing related to a present service according to an embodiment. [Figure 18] 10 is an example of a home screen according to an embodiment. [Figure 19] 10 is an example of a list screen showing a predetermined portion of the correction results according to an embodiment. [Figure 20] 10 is an example of a quiz screen according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. The following embodiments are merely examples for explaining the present invention, and are not intended to limit the present invention to the embodiments. Furthermore, the present invention can be modified in various ways without departing from the spirit of the present invention. Furthermore, the same components in each drawing will be designated by the same reference numerals whenever possible, and redundant explanations will be omitted whenever possible.

[0011] <System configuration> FIG. 1 is a diagram showing an example of the configuration of an information processing system 1 according to an embodiment of the present invention. In the example shown in FIG. 1, the information processing system 1 illustratively includes a server 10, a server 20, a processing device 30, and a network N. The server 10, the server 20, and the processing device 30 are capable of transmitting and receiving data to and from each other via the network N. The server 10 and the server 20 may be configured with a plurality of processing devices. Furthermore, the number of the servers 10, the servers 20, and the processing devices 30 may be any number. Furthermore, when distinguishing between the plurality of servers and processing devices, they may also be referred to as the nth server or the nth processing device (n is an integer equal to or greater than 1).

[0012] The information processing system 1 provides, for example, a text correction service. The information processing system 1 provides, for example, a service in which text created by a user is corrected using large-scale language models (LLMs), for example, a service in which a generative AI performs corrections (hereinafter also referred to as the Service). Hereinafter, the Service will be described taking as an example a service in which text including English is corrected. However, the Service is not limited to this. The Service may also be, for example, a service in which text in Japanese, a foreign language, or a combination of multiple languages is corrected. Note that the Service is not limited to generative AI, and can also be applied to a machine learning model capable of correcting text.

[0013] The server 10 is a server that provides the present service to users. The server 10 acquires text data created by a user from the processing device 30, for example, via the network N. The server 10 transmits the text data and instructions for correcting the text data to the server 20, for example, via the network N. The server 10 acquires the correction results from the server 20 via the network N and executes predetermined processing on the correction results. The server 10 transmits the correction results, etc. to the processing device 30 via the network N.

[0014] The server 20 is a server that corrects text created by a user. The server 20 acquires text data and instructions for correcting the text data from the server 10, for example, via the network N. The server 20, for example, executes a process related to the correction of the text data. The server 20, for example, executes a process related to the correction of the text using a generation AI. Examples of the generation AI include, but are not limited to, ChatGPT provided by OpenAI OpCo, LLC, Bard (registered trademark) from Google, Bing AI Chat, Cohesive, Perplexity, Writesonic, Forefront Chat, HuggingChat, and Catchy (registered trademark).

[0015] The processing device 30 is an information processing device used by a user who creates a sentence. The processing device 30 (hereinafter also referred to as a user terminal 30) is, for example, a personal computer, a smartphone, a tablet terminal, etc. The user terminal 30 transmits sentence data of a sentence input and created by the user to the server 10, for example, via the network N. The processing device 30 obtains correction results of the sentence data from the server 10, for example, via the network N.

[0016] The processing device 30 may also use a web browser to access a website (or web page) provided by the server 10 and receive the service provided on this website.

[0017] An application program (app) for using the present service provided by the server 10 may be installed on the processing device 30. This app causes the processing device 30 to execute at least a part of the processing related to the present service disclosed in the following embodiments. When this app is executed, the processing device 30 accesses the server 10 and transmits and receives information used to execute the app.

[0018] The network N may be realized by, for example, a network such as the Internet or a mobile phone network, a LAN (Local Area Network), or a combination of these. The following describes in detail each component of the information processing system 1 that enables the present service to be executed.

[0019] <Server configuration> 2 is a block diagram illustrating an example of a server 10 according to one embodiment. The server 10 includes one or more processors (e.g., CPUs) 110, one or more network communication interfaces 120, a storage device (storage unit) 130, and one or more communication buses 150 for interconnecting these components.

[0020] The server 10 may optionally include a user interface 140, which may include a display and input devices (such as a keyboard and / or mouse or some other pointing device).

[0021] The storage device 130 may be, for example, a high-speed random access memory (main storage device) such as a DRAM, an SRAM, or other random access solid-state storage device. Alternatively, the storage device 130 may be a non-volatile memory (auxiliary storage device) such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. Alternatively, the storage device 130 may be a non-transitory computer-readable recording medium that stores programs and the like. Alternatively, the storage device 130 may be either a main storage device (memory) or an auxiliary storage device (storage), or may include both.

[0022] The storage device 130 stores data, programs, etc. used by the information processing system 1. For example, the storage device 130 stores user information (for example, FIG. 15) of users who use the system.

[0023] Another example of storage device 130 may be one or more storage devices located remotely from processor 110. In some embodiments, storage device 130 stores programs, modules, and data structures, or a subset thereof, that are executed by processor 110.

[0024] The processor 110 executes a program stored in the storage device 130 to control, for example, the processing executed by the information processing unit 111.

[0025] The information processing unit 111 includes, for example, an acquisition unit 112, an input unit 113, an extraction unit 114, a recording unit 115, a creation unit 116, a generation unit 117, a presentation unit 118, and a selection unit 119.

[0026] (Outline of the process up to the creation of text in this service) Before describing each unit of the information processing unit 111, an overview of the process up to writing in the present service provided by the information processing unit 111 will be described using an example screen of the user terminal 30. The process up to writing in the present service includes setting the user's level, setting the corrector, and writing the text.

[0027] FIG. 3 is a diagram showing an example of a screen for setting a user's writing level according to one embodiment. In the example shown in FIG. 3, the user selects his / her writing level based on a level selection screen D10 displayed on the user terminal 30. For example, the user selects his / her writing level by operating (using tapping as an example below) buttons B10 to B60 displayed on the screen D10. As an example, the user can set his / her writing level to "Basic: Able to make basic greetings" by tapping button B10. This allows the user to receive the service according to his / her level.

[0028] Furthermore, by setting the user's writing level based on screen D10, information relating to the level (hereinafter also referred to as user level information) is transmitted from user terminal 30 to server 10. The timing for setting the user's own level is arbitrary, and may be, without limitation, for example, when the user accesses the service, downloads the app for the service, and creates the necessary account, or when the app for the service is launched, or when the website (or web page) is accessed. Furthermore, when any of buttons B10 to B60 shown in FIG. 3 is tapped, the screen may transition to the screen shown in FIG. 4.

[0029] FIG. 4 is a diagram showing an example of a screen for setting an instructor character according to one embodiment. In the example shown in FIG. 4, the user sets the character based on an instructor character selection screen D20 displayed on the user terminal 30. The instructor corrects writing, chats, and gives quizzes related to improving writing ability. In the present service, for example, the set character provides the user with writing correction guidance, etc., in a tone of voice appropriate to the character. For example, the user can set the character displayed in area C10 on screen D20 as the character that will provide correction guidance, etc. The user can change the character displayed in area C10 by tapping buttons displayed in areas C20 and C30 on screen D20.

[0030] In the example shown in FIG. 4, character CH1 is initially displayed in area C10. By tapping the button in area C20 once, the user can change the character displayed in area C10 to character CH2. By tapping the button in area C20 twice, the user can change the character displayed in area C10 to character CH3. By tapping the button in area C30 once, the user can change the character displayed in area C10 to character CH4. By tapping the button in area C30 twice, the user can change the character displayed in area C10 to character CH5. By directly tapping any of CH3 to CH5, the user can change the character displayed in area C10 to any character of their choice. This allows the user to receive corrections and guidance on their writing from a character that suits their preferences. Although not shown, the user may be able to create their own dedicated character, set character information (e.g., the character's gender, name, occupation, type, etc.), and select the character.

[0031] 4 is tapped, the screen may transition to the screen shown in Fig. 5. Furthermore, by setting a character based on screen D20, information about the character (hereinafter also referred to as character information) is transmitted from user terminal 30 to server 10. The character information may include, for example, the character's attributes (for example, the character's gender, name, occupation, type, etc.).

[0032] FIG. 5 is a diagram showing an example of a screen for inputting text according to an embodiment. In the example shown in FIG. 5, a user inputs a diary entry based on a text entry screen D30 displayed on the user terminal 30. For example, the user can input the title of the diary entry in area A10 by tapping "Enter title" in area A10. For example, the user can input the main text of the diary entry in area A20 by tapping "Enter main text" in area A20. Note that the text data may be input using a voice recognition function provided in the user terminal 30, and the text data may include voice data.

[0033] In addition, in the example shown in Fig. 5, the user can get hints for creating a diary by tapping button B100. In the example shown in Fig. 5, when button B100 is tapped, "Questions for starting writing," "Help for expressing emotions," "Useful expressions to use in a diary," and "Suggest a topic" are displayed.

[0034] For example, by tapping "Question for writing", the user can receive hints about writing a diary entry from a character on a chat screen, which will be described later. To realize this process, when "Question for writing" is tapped, a request about writing a diary entry is transmitted from the user terminal 30 to the server 10 via the network N. The server 10 inputs, for example, via the network N, instructions (prompts) about writing a diary entry to the generation AI provided by the server 20. The server 20 outputs, via the network N, hints about writing a diary entry created by the generation AI in response to the instructions to the server 10. The server 10 transmits the output results of the generation AI to the user terminal 30 via the network N.

[0035] For example, by tapping "Help on emotional expression," the user can receive hints about emotional expression in the diary from a character on a chat screen, which will be described later. To realize this process, when "Help on emotional expression" is tapped, a request about emotional expression in the diary is transmitted from the user terminal 30 to the server 10 via the network N. The server 10 inputs, for example, instructions about emotional expression in the diary to the generation AI provided by the server 20 via the network N. The server 20 outputs hints about emotional expression created by the generation AI in response to the instructions to the server 10 via the network N. The server 10 transmits the output results of the generation AI to the user terminal 30 via the network N.

[0036] For example, by tapping "useful expressions for diary entries," the user can receive hints from a character on useful expressions for diary entries on a chat screen, which will be described later. To achieve this process, when "useful expressions for diary entries" is tapped, a request for useful expressions for diary entries is transmitted from the user terminal 30 to the server 10 via the network N. The server 10 inputs, for example, instructions on useful expressions for diary entries to the generation AI provided by the server 20 via the network N. The server 20 outputs, via the network N, hints on useful expressions for diary entries that are created by the generation AI in response to the instructions to the server 10. The server 10 transmits the output results of the generation AI to the user terminal 30 via the network N.

[0037] For example, by tapping "Suggest a topic," the user can receive a hint about the topic of the diary from the character on a chat screen described later. To realize this process, when "Suggest a topic" is tapped, a request about the topic of the diary is transmitted from the user terminal 30 to the server 10 via the network N. The server 10, for example, inputs an instruction about the topic of the diary to the generation AI provided by the server 20 via the network N. The server 20 outputs a hint about the topic created by the generation AI in response to the instruction to the server 10 via the network N. The server 10 transmits the output result of the generation AI to the user terminal 30 via the network N. Note that after button B100 is selected and transition is made to a chat screen described later, questions for creating a diary may be answered in succession according to the user's answers. In other words, the character may transmit questions necessary for the user to create a diary in an interview format. For example, suppose that the character asks "What did you do today?" and the user answers "I went out to eat," and the character asks "Where did you go with whom?" and the user answers "I went to yakiniku with a friend." Based on this series of exchanges, the server 10 or server 20 can send the user an outline for creating a diary (e.g., "Today I went to eat yakiniku with a friend") or a rough draft of the diary. This allows the user to use the acquired outline to proceed with creating the diary. Note that the user's message may be sent from the user terminal 30 to the server 10 or server 20 via the server 10, and a message from the system side (character side) following the user's message may be sent from the server 10 or server 20 via the server 10. However, if the server 20 can directly access the user terminal 30, it is not necessary to go through the server 10.

[0038] 5, the user can save the title and text they have entered by tapping button B80. When button B80 is tapped, the entered title and text are associated with information identifying the user (for example, an ID, which will be described later) and stored in storage device 130 of server 10. This allows the user to edit the saved title and text, for example, when logging out of the service and then logging back in.

[0039] 5, the user can receive guidance on correcting the diary entry by tapping button B90. When button B90 is tapped, information about the diary entry (an example of text data) and a request for correction of the diary entry are sent from the user terminal 30 to the server 10, and the information may be stored in the storage device 130 of the server 10. At this time, the server 10 that has received the correction request may send information about the diary entry to the server 20.

[0040] Fig. 6 is a diagram showing an example of a screen on which a sentence is input according to an embodiment. In the example shown in Fig. 6, the diary title "Balancing Work and Play" is input in area A10 shown in Fig. 5. Furthermore, the main text of the diary, "Today was specially... week." is input in area A20 shown in Fig. 5. As shown in Fig. 6, the diary may contain both Japanese and foreign languages.

[0041] (Example of processing performed by the server) Next, a specific example of the process executed by the server 10 will be described.

[0042] 2, the acquisition unit 112 acquires text data relating to text created by the user. The acquisition unit 112 acquires, for example, information relating to the diary including at least one of the title and the text (hereinafter also referred to as diary information) from the user terminal 30. The acquisition unit 112 may also acquire diary information that has been temporarily saved in the storage device 130.

[0043] The input unit 113 inputs the text data and the correction instructions to the generation AI. The input unit 113 transmits the diary information and the correction instructions for the diary content to the generation AI provided by the server 20, for example, using the API (Application Programming Interface) of the generation AI.

[0044] FIG. 7 is a diagram showing an example of a correction instruction. In the example shown in FIG. 7, the input unit 113 transmits the correction instruction "Please correct the following sentence..." and diary information (e.g., the main text of the diary) acquired from the user terminal 30 to the server 20. To realize the processing of the input unit 113, the storage device 130 stores, for example, a fixed phrase related to the correction instruction in advance. As a specific example, the storage device 130 may store the fixed phrase "Please correct the following sentence...". The input unit 113 may create the correction instruction by, for example, reading out the fixed phrase that will be at least a part of the prompt from the storage device 130 and adding the diary information to the fixed phrase.

[0045] The acquisition unit 112 acquires the correction results of the sentence data from the generation AI provided by the server 20. The acquisition unit 112 acquires, for example, the correction results of the diary from the server 20. Upon acquiring the correction results, the server 10 transmits the correction results to the user terminal 30.

[0046] The correction results may include, for example, a correct expression for an incorrect expression, and a more appropriate expression for a predetermined expression. FIG. 8 is a diagram showing an example of the correction results. In the example shown in FIG. 8, "especially" is corrected to "exceptionally." In the example shown in FIG. 8, "big update" is corrected to "major update." In the example shown in FIG. 8, "sometimes wish" is corrected to "I often find myself wishing."

[0047] The extraction unit 114 extracts a predetermined portion from the correction result based on predetermined extraction conditions related to the correction. Specific processing by the extraction unit 114 will be described later.

[0048] The recording unit 115 records a predetermined portion included in the correction result in association with the user's identification information. For example, the recording unit 115 associates the user's identification information (for example, an ID, which will be described later) with the predetermined portion extracted by the extraction unit 114 and stores them in the storage device 130. The recording unit 115 may also record the predetermined portion by dividing it into categories. For example, the recording unit 115 may record words, grammar, idioms that the user used incorrectly, and expressions that are not incorrect but are more appropriate, for each category.

[0049] By performing the above process, a specific portion of the text data corrected by the generation AI is associated with the user's identification information and stored in the storage device 130, making it possible for the correction results output by the AI to be effectively used for subsequent learning related to writing in online learning services that use AI. For example, the following is an example of an effective use method.

[0050] Here, an example in which the correction results can be used for subsequent learning related to writing will be described. FIG. 9 is a diagram showing an example of a screen displaying the correction results according to one embodiment. By way of example and not limitation, the user can receive correction guidance in a chat format from the character set as shown in FIG. 4. In the example shown in FIG. 9, when button B90 shown in FIG. 6 is tapped, chat screen D50 opens on user terminal 30, and a message from the user, "Please correct my writing," is displayed in speech bubble SB1. At this time, message information is sent from server 10 to user terminal 30 as a reply from the character, and "You did a great job" is displayed in speech bubble SB2.

[0051] Furthermore, when the correction results are sent from the server 10 to the user terminal 30, the correction results shown in Fig. 8 are displayed in a speech bubble SB3, for example. This allows the user to check the correction results for the diary. Furthermore, a message in a tone of voice corresponding to the set character is displayed in a speech bubble SB2, which can increase the motivation of the user who selected the character to study.

[0052] 9, message information is transmitted from the server 10 to the user terminal 30, and a message bubble SB4 displays the message "'Specially' is expressed as 'exceptionally' and 'big update' as 'major update'." This allows the user to easily check appropriate expressions for inappropriate expressions in the diary entry (e.g., an expression entered in Japanese in an English diary, an expression with a spelling error, etc.). To achieve this process, the server 10 may transmit message information including, for example, "exceptionally" and "major update" extracted by the extraction unit 114 to the user terminal 30.

[0053] 9, message information is transmitted from the server 10 to the user terminal 30, and the message "You can emphasize the emotion more by changing "sometimes wish" to "I often find myself wishing"" is displayed in the speech bubble SB5. This allows the user to easily check a more appropriate expression for the expression in the diary entry they entered. To achieve this process, the server 10 may transmit the message information to the user terminal 30, including, for example, "I often find myself wishing" extracted by the extraction unit 114.

[0054] The extraction unit 114 may specify, as an extraction condition, a corrected portion of the text data, and extract inappropriate portions of the text data from the correction results based on the specified portion. When the extraction condition is a corrected portion, the extraction unit 114 may, for example, compare the original text data with the corrected text data and extract the different portions. As a specific example, the extraction unit 114 may extract inappropriate expressions in the diary from the correction results based on the corrected portions of the diary acquired from the server 10. More specifically, the extraction unit 114 extracts "exceptionally" and "major update," which are English translations of the portions expressed by the user in Japanese and included in the correction results shown in FIG. 8.

[0055] This allows a specific portion of the text data corrected by the generation AI to be associated with the user's identification information and stored in the storage device 130, making it possible for the corrected portion to be used for subsequent learning related to writing in online learning services that use AI.

[0056] The acquisition unit 112 may acquire information regarding how the generation AI corrects sentences. As a non-limiting example, the acquisition unit 112 acquires character information from the user terminal 30 as information regarding how the generation AI of the server 20 corrects the diary.

[0057] The creation unit 116 creates correction instructions based on the correction method. For example, when character information is acquired by the acquisition unit 112, the creation unit 116 creates correction instructions for the diary based on the character information. More specifically, the creation unit 116 creates a correction instruction (prompt) including "You are a female idol who is also an English teacher," as shown in FIG. 7, based on the character information acquired by the acquisition unit 112. As an example, the creation unit 116 acquires a fixed phrase "You are {C} who is also an English teacher," and inputs attribute information of the character information acquired by the acquisition unit 112 into the parameter portion of {C} to create the correction instruction.

[0058] As a result, the correction instruction is input to the generation AI of the server 20, and a correction result corresponding to the character set by the user is output. As a result, the user can receive correction guidance according to the character set by the user, which can increase the user's motivation to learn.

[0059] The acquisition unit 112 may acquire information about the user's level related to writing. For example, the acquisition unit 112 acquires the user level information from the user terminal 30 by selecting the user's level related to writing based on the screen D10 shown in FIG.

[0060] When the user level information is acquired by the acquisition unit 112, the creation unit 116 may create a correction instruction for the diary based on the user level information. For example, the creation unit 116 may acquire a template indicating an answer content including the user level information as a parameter, and create a correction instruction by inputting content based on the user level information acquired by the acquisition unit 112 into the parameter portion. As an example, in the case of a template saying "However, please speak in {L}," the creation unit 116 inputs the user's native language (e.g., Japanese) into {L} if the level indicated by the user level information is basic, beginner, or intermediate, and inputs the language being studied (e.g., English) into {L} if the level indicated by the user level information is upper-intermediate, advanced, or proficient.

[0061] That is, when the level included in the user level information acquired by the acquisition unit 112 is relatively low (for example, basic, beginner, intermediate), the creation unit 116 creates a correction instruction including "However, please speak in Japanese when speaking." as shown in Fig. 7. Also, when the level included in the user level information acquired by the acquisition unit 112 is relatively high (for example, upper-intermediate, advanced, proficient), the creation unit 116 creates a correction instruction including "However, please speak in English when speaking." (not shown).

[0062] As a result, correction instructions based on the user level information are input to the generation AI of the server 20, and correction results according to the user's writing level are output. As a result, the user can receive correction guidance according to their own level.

[0063] The extraction unit 114 may extract one or more common words from multiple sentence data. For example, the extraction unit 114 may extract one or more common words from multiple diaries stored in the storage device 130. As an example, if the words "home run," "strike," and "bat" are included in multiple diaries stored in the storage device 130, the extraction unit 114 extracts "home run," "strike," and "bat" included in the multiple diaries.

[0064] The storage device 130 may store in advance a plurality of predetermined words that are frequently used by users. For example, the storage device 130 may store a plurality of predetermined words including "home run," "strike," and "bat." The extraction unit 114 may, for example, read a plurality of diaries acquired by the acquisition unit 112, identify words such as "home run," "strike," and "bat" contained in the plurality of diaries, and store the identified words in the storage device 130.

[0065] The generation unit 117 may generate a theme of a sentence based on one or more words extracted by the extraction unit 114. For example, the generation unit 117 generates a diary theme based on one or more words extracted by the extraction unit 114. As an example, the generation unit 117 generates "baseball" as the diary theme based on "home run," "strike," and "bat" extracted by the extraction unit 114.

[0066] In this case, the storage device 130 may store in advance a plurality of predetermined words and terms related to the predetermined words in association with one another. For example, the storage device 130 may store "baseball" in association with each of "home run," "strike," and "bat." As an example, the generation unit 117 may generate "baseball," which is associated with each of "home run," "strike," and "bat," as a diary theme.

[0067] The presentation unit 118 presents the theme of the text to the user. For example, the presentation unit 118 outputs the diary theme generated by the generation unit 117 to the user terminal 30 used by the user. The user terminal 30 displays the acquired diary theme on the screen, thereby presenting and suggesting a diary theme suitable for the user to the user.

[0068] 10 is a diagram showing an example of a screen on which a theme of a sentence is presented according to one embodiment. In the example shown in FIG. 10, when the generation unit 117 generates "baseball" as the theme of a diary, the presentation unit 118 transmits a message related to the theme ("Let's write a diary entry on the theme of baseball!") to the user terminal 30. When the user terminal 30 receives the message, it displays a speech bubble SB6 including the message on the chat screen D60.

[0069] This makes it possible to present the user with a topic that will interest them, thereby stimulating their motivation to write and increasing the likelihood that they will engage in learning.

[0070] The generation unit 117 may input to the generation AI one or more words extracted by the extraction unit 114 and a prompt to generate a theme of a sentence related to the one or more words. For example, the generation unit 117 may input to the generation AI of the server 20 the words "home run", "strike", and "bat" extracted by the extraction unit 114 and a prompt to generate a diary theme related to "home run", "strike", and "bat".

[0071] 11 is a diagram showing an example of an instruction to generate a theme for a sentence. As shown in FIG. 11, the generation unit 117 may generate an instruction to generate a diary theme such as, "Please generate a diary theme related to 'home run,' 'strike,' and 'bat.'"

[0072] In this case, the storage device 130 may store in advance a standard phrase related to an instruction to generate a theme. For example, the storage device 130 may store a standard phrase such as "Please generate a diary theme related to {T}." The generation unit 117 may read the standard phrase from the storage device 130 and create a generation instruction by inputting the words extracted by the extraction unit 114 into the parameter portion of {T}.

[0073] This allows the process of generating a theme for a text to be executed in the server 20. As a result, the processing load on the server 10 can be reduced.

[0074] The acquisition unit 112 may acquire predetermined information from an external service. For example, the acquisition unit 112 acquires a predetermined character string included in an external service, such as an SNS, blog, or news site, from the external service. For example, the acquisition unit 112 acquires a character string included in an article on a news site published on the Internet. For example, software for providing the present service in the server 10 and software for providing the external service may be linked via an API. As an example, the acquisition unit 112 may link with a search engine via an API, input a search keyword, and acquire a predetermined character string.

[0075] The generation unit 117 may identify whether the character string acquired by the acquisition unit 112 includes "baseball," which is the theme of the generated diary. As a non-limiting example, the identification process may be performed by natural language processing. If the character string includes "baseball" as a result of the identification process, the generation unit 117 may extract one or more predetermined characters from the character string. For example, the generation unit 117 extracts "Today, player X of team ABC hit a walk-off home run" from a character string included in an article on a news site. As a non-limiting example, the extraction process may be performed by natural language processing.

[0076] The presentation unit 118 transmits a message including one or more characters extracted by the generation unit 117 to the user terminal 30. For example, the presentation unit 118 transmits a message including "Player X of team ABC hit a walk-off home run today! It would be nice to write your thoughts about it" to the user terminal 30. As shown in Fig. 10, when the user terminal 30 receives the message, it displays a speech bubble SB7 including the message.

[0077] This allows the user to be presented with a topic that will interest the user based on a predetermined character string included in the external service, thereby increasing the user's motivation to study.

[0078] The extraction unit 114 may extract appropriate portions generated by the generation AI from the correction results based on the corrections made to the corrected sentence data. For example, the extraction unit 114 extracts "exceptionally" and "major update" from the correction results shown in FIG. 8. As shown in FIG. 8, "exceptionally" is a correction of "extraordinarily" in the diary. As shown in FIG. 8, "major update" is a correction of "major update" in the diary. Furthermore, the extraction unit 114 extracts, for example, "I often find myself wishing" from the correction results shown in FIG. 8. As shown in FIG. 8, "I often find myself wishing" is a correction of "sometimes wish" in the diary. Note that although the expression "sometimes wish" is not incorrect in the diary, it has been corrected by the generation AI of the server 20 as an appropriate expression.

[0079] The generation unit 117 may generate a message that prompts the user to create a sentence based on an appropriate portion of the corrected sentence. The generation unit 117 may generate a message that prompts the user to create a diary based on, for example, "exceptionally" extracted by the extraction unit 114. Alternatively, the generation unit 117 may generate a message that prompts the user to create a diary based on "I often find myself wishing" extracted by the extraction unit 114. As an example, the generation unit 117 may obtain a fixed phrase "Let's write a diary using {M}!" and generate a message by inputting at least one word extracted by the extraction unit 114 into {M}.

[0080] 12 is a diagram showing an example of a screen on which a message prompting the user to create a sentence is displayed. The generation unit 117 generates a message prompting the user to create a diary ("Let's write a diary using exceptionally!") based on, for example, "exceptionally" extracted by the extraction unit 114. The message is transmitted to the user terminal 30 via the network N by, for example, the network communication interface 120 of the server 10. In the example shown in FIG. 12, on the user terminal 30, "Let's write a diary using exceptionally!" is displayed in a speech bubble SB8 as a message from a character on the chat screen D70.

[0081] Furthermore, the generation unit 117 generates a message ("Let's write a diary using 'I often find myself wishing'!") that prompts the user to create a diary, for example, based on "I often find myself wishing" extracted by the extraction unit 114. The message is transmitted to the user terminal 30 via the network N, for example, by the network communication interface 120 of the server 10. In the example shown in FIG. 12, on the user terminal 30, "Let's write a diary using 'I often find myself wishing'!" is displayed in a speech bubble SB9 on the chat screen D70 as a message from a character.

[0082] This encourages the user to write sentences using appropriate expressions included in the correction results.

[0083] The input unit 113 may input to the generation AI an appropriate portion contained in the corrected document data and an instruction to generate a message that prompts the user to create a sentence. For example, the input unit 113 inputs "exceptionally" extracted by the extraction unit 114 and an instruction to generate a message that prompts the user to create a diary to the generation AI of the server 20 via the network N.

[0084] Fig. 13 is a diagram showing an example of an instruction to generate a message prompting the user to create a sentence. In the example shown in Fig. 13, the input unit 113 may generate an instruction to generate a diary theme such as "Please generate a message to prompt the user to create a diary using exceptionally." Note that the generation unit 117 may execute the process of generating an instruction (prompt) to generate a message prompting the user to create a sentence.

[0085] In this case, the storage device 130 may store in advance a fixed phrase related to an instruction to generate a message that prompts the user to create a sentence. The storage device 130 may store, for example, a fixed phrase such as "Please generate a message to prompt the user to create a diary using {M}." As an example, the generation unit 117 may read the fixed phrase from the storage device 130 and input "exceptionally" extracted by the extraction unit 114 into {M} to generate the generation instruction.

[0086] This allows the server 20 to execute a process of generating a message that prompts the user to create sentences using appropriate expressions included in the correction results, thereby reducing the processing load on the server 10.

[0087] The selection unit 119 selects at least one predetermined portion from a plurality of predetermined portions in the corrected text data. The selection unit 119 selects at least one predetermined portion from a plurality of predetermined portions (e.g., inappropriate expressions, appropriate expressions, etc.) extracted by the extraction unit 114 and stored in the storage device 130. As an example, the selection unit 119 extracts "exceptionally" stored in the storage device 130.

[0088] The input unit 113 may input to the generation AI an instruction to create a question related to at least one predetermined part selected by the selection unit 119. FIG. 14 is a diagram showing an example of an instruction to create a question. In the example shown in FIG. 14, the input unit 113 generates an instruction to create a question related to "exceptionally" selected by the selection unit 119.

[0089] In this case, the storage device 130 may store in advance a fixed phrase related to an instruction to create a question. For example, the storage device 130 may store a fixed phrase such as "Please generate a quiz about the English word {Q}." The input unit 113 may read the fixed phrase from the storage device 130 and generate an instruction to create a question by inputting the extracted predetermined part, "exceptionally," into the parameter part of {Q}. Note that the process of generating an instruction to create a question may be executed by the generation unit 117. The input unit 113 inputs the generated instruction to the generation AI of the server 10.

[0090] This makes it possible to generate questions that can be used for the user's future learning regarding writing, based on expressions corrected in the diary written by the user.

[0091] <Examples of each data> Next, a description will be given of an example of each piece of data stored in the storage device 130 of the server 10. Fig. 15 is a diagram showing an example of user information according to an embodiment.

[0092] The user information shown in FIG. 15 includes, by way of example and not limitation, various data / information such as "ID," "name," "contact information," "password," "character," "text," "correction results," and "specified portion."

[0093] The "ID" includes, for example, identification information for identifying the user. The "Name" includes, for example, the user's nickname (which may be the user's real name) set when creating an account for the Service. The "Contact Information" includes, for example, at least one of the user's email address, SNS ID, and phone number. The "Password" includes, for example, a password set by the user or a password set by the system. The "Character" includes, for example, character information set on screen D20 shown in FIG. 4. The "Text" includes, for example, at least one of the title, text, and date of the diary entry entered by the user. Note that the "Text" may include multiple texts entered by a single user, and in this case, each text may be associated with an "ID" or the like and stored in the storage device 130. The "Correction Result" includes, for example, the correction result acquired from the generation AI of the server 20. The "Predetermined Portion" includes, for example, a predetermined portion of the correction result extracted by the extraction unit 114.

[0094] For ease of explanation, the text, correction results, and specified parts are associated with user information, but this information may also be associated with the user's ID and managed as data separate from the user information, for example, as text-related information.

[0095] <Configuration of the processing device> 16 is a diagram illustrating an example of a processing device 30 (user terminal 30) according to an embodiment of the present disclosure. For example, the processing device 30 is a device used by a user. The processing device 30 includes one or more processors (e.g., CPUs) 310, one or more network communication interfaces 320, a storage device (storage unit) 330, a user interface 340, and one or more communication buses 350 for interconnecting these components.

[0096] The user interface 340 includes a display 351 and an input device 352 (such as a keyboard and / or a mouse or some other pointing device).

[0097] The storage device 330 may be, for example, a high-speed random access memory (primary storage device) such as a DRAM, an SRAM, or other random access solid-state storage device. The storage device 330 may also be a non-volatile memory (secondary storage device) such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. The storage device 330 may also be a non-transitory computer-readable recording medium. The storage device 330 may be either a primary storage device (memory) or a secondary storage device (storage), or may include both.

[0098] The storage device 330 stores data and programs used by the information processing system 1. For example, the storage device 330 stores application programs for the processing device 30 in the information processing system 1.

[0099] The processor 310 executes a program stored in the storage device 330 to configure an information processing unit 311. The information processing unit 311 includes, for example, a web browser and an email application. The web browser enables browsing of websites provided by the server 10. Furthermore, by executing an installed application for the user terminal, browsing of websites providing the above-mentioned services becomes possible. The information processing unit 311 has a communication control unit 312, a display control unit 313, and an operation control unit 314 to receive the provided services.

[0100] The communication control unit 312 acquires screen information for each of the above-mentioned screens. In addition, the communication control unit 212 acquires, for example, message information displayed in a speech bubble on the screen where the chat is performed.

[0101] Furthermore, the communication control unit 312 outputs to the server 10 information selected, set, or input using each screen, and / or a predetermined request (for example, a registration request, a creation request, or the like).

[0102] The display control unit 313 controls the display 351 to display the above-mentioned screens related to the present service based on the screen information acquired by the communication control unit 312.

[0103] The operation control unit 314 receives user operations on UI widgets displayed on each screen, transmits the operation information to an application of the processing device 30, and outputs it to the server 10. For example, the operation control unit 314 may output user information of the present service and input text data to the server 10.

[0104] <Operation description> Next, a description will be given of each operation of the information processing system 1. Fig. 17 is a sequence diagram showing an example of processing related to the present service according to an embodiment.

[0105] In step S11, the communication control unit 312 of the processing device 30 (user terminal 30) transmits text data relating to a text created by the user to the server 10. The acquisition unit 112 of the server 10 acquires the text data. The acquisition unit 112 acquires, for example, diary information including at least one of a title and / or a body of text from the user terminal 30. The acquisition unit 112 may also acquire diary information temporarily stored in the storage device 130.

[0106] In step S12, the input unit 113 of the server 10 inputs the text data and the correction instructions to the generation AI of the server 20. The method for generating the correction instructions is as described above.

[0107] In step S13, the acquisition unit 112 of the server 10 acquires the correction results of the text data from the server 20.

[0108] In step S14, the extraction unit 114 of the server 10 extracts a predetermined portion from the correction result based on predetermined extraction conditions related to correction. If the extraction condition is a correction portion, for example, the extraction unit 114 compares the original text data with the text data of the correction result and extracts the different portions.

[0109] In step S15, the recording unit 115 of the server 10 associates the predetermined portion with the user's identification information and records them in the storage device 130. For example, the recording unit 115 associates the user's identification information (e.g., ID) with the predetermined portion extracted by the extraction unit 114 and the user's identification information (e.g., ID) and stores them in the storage device 130. The recording unit 115 may also record the predetermined portion by dividing it into categories. For example, the recording unit 115 may record incorrect words, grammar, idioms used by the user, and expressions that are not incorrect but are more appropriate, for each category.

[0110] <Screen example> Next, an example of another screen displayed on the user terminal 30 will be described. FIG. 18 is an example of a home screen according to an embodiment. On the home screen D80 shown in FIG. 18, for example, information related to the diary of a predetermined day is displayed. In the example shown in FIG. 18, information related to the diary entered on November 24, 2023 is displayed on the home screen D80. In the example shown in FIG. 18, the number of words (number of words) included in the diary entered on November 24, 2023, newly entered words (number of new vocabulary words) in the diary of the previous day (November 23, 2023), an evaluation of today's diary (today's rank), the number of days of continuation, etc. are displayed on the home screen D80 as information related to the entered diary. This allows the user to check information related to the diary of a predetermined day.

[0111] Furthermore, information relating to the diary for a specified week may be displayed on the home screen. In the example shown in Fig. 18, information relating to the diary entered for the week of November 19, 2023 to November 25, 2023 is displayed on the home screen D90. In the example shown in Fig. 18, the number of words in the diary for each day of the week is displayed as a bar graph as information relating to the entered diary. This allows the user to check information relating to the diary for a specified week.

[0112] Furthermore, when button B110 is tapped on home screen D90, the screen transitions to home screen D90. Similarly, when button B120 is tapped on home screen D80, the screen transitions to home screen D90. This allows the user to easily switch between home screen D80 and home screen D90.

[0113] Furthermore, when button B130 is tapped on home screen D80 or home screen D90, the screen transitions to diary entry screen D40 shown in Fig. 6. This allows the user to easily switch from home screen D80 or home screen D90 to the diary entry screen.

[0114] Furthermore, when button B140 is tapped on home screen D80 or home screen D90, the screen transitions to screen D100 shown in Fig. 19. Screen D100 will be described later. Furthermore, when button B150 is tapped on home screen D80 or home screen D90, the screen transitions to screen D110 shown in Fig. 20. Screen D110 will be described later.

[0115] FIG. 19 is an example of a screen of a list on which predetermined portions of the correction results are posted according to one embodiment. In the example shown in FIG. 19, a list (hereinafter also referred to as a lesson note) on which predetermined portions of the correction results extracted by the extraction unit 114 are posted is displayed on screen D100. On screen D100, for example, "exceptionally" extracted by the extraction unit 114 is displayed in association with its Japanese translation. On screen D100, for example, "major" included in "major update" extracted by the extraction unit 114 is displayed in association with its Japanese translation. This allows the user to check a list of all the corrected portions of the diary they entered, and enables them to study based on the list.

[0116] In addition, the extracted words are displayed with a "check mark" indicating "not mastered" or "mastered." "Not mastered" indicates, for example, that the word has not been correctly answered in a quiz (described later) or has not been used in writing, and a "check mark" indicates that the word has been correctly answered a predetermined number of times in a quiz or has been used more than a predetermined number of times in writing.

[0117] Furthermore, when button B150 is tapped on screen D100, the screen transitions to screen D110 shown in Fig. 20. Screen D110 will be described later. Furthermore, when button B160 is tapped on screen D100, the screen transitions to home screen D80 (or home screen D90) shown in Fig. 18.

[0118] Fig. 20 is an example of a quiz screen according to one embodiment. In the example shown in Fig. 20, on screen D110, the user can set questions (hereinafter also referred to as quizzes) regarding predetermined portions (e.g., inappropriate expressions, appropriate expressions, etc. included in the correction results) generated by the generation AI of the server 20.

[0119] In the example shown in FIG. 20, the user can set the content of the quiz questions on screen D110. For example, the user can select one or more of "unlearned expressions contained in the lesson notes," "learned expressions contained in the lesson notes," and "other than the lesson notes" as the content of the questions. "Unlearned expressions contained in the lesson notes" includes, for example, expressions that have not been answered correctly in the quiz. "Learned expressions contained in the lesson notes" includes, for example, expressions that have been answered correctly in the quiz. "Other than the lesson notes" includes, for example, expressions that are not included in the user's lesson notes, and includes, by way of example and not limitation, expressions included in other users' lesson notes.

[0120] In the example shown in FIG. 20, the user can set the quiz question format on screen D110. For example, the user can select one or more of a "matching quiz" and a "fill-in-the-blank quiz" as the question format. A "matching quiz" is a quiz that asks, for example, a combination of an expression and its correct meaning. A "fill-in-the-blank quiz" is a quiz that asks, for example, a sentence in which a portion of the expression is missing, to complete the correct sentence by entering the missing expression.

[0121] In the example shown in FIG. 20, when a button B170 is tapped on the screen D110, the screen transitions to a screen D120 for answering a quiz.

[0122] <Modification> In the above embodiment, the service has been described as an example of diary correction, but the service is not limited to this. For example, the service may also correct essays, impressions, book manuscripts, and other writings. In the above embodiment, the service has been described using diary correction as an example, but the diary may also include image data (still image data or video data). For example, a user attaches image data related to the diary content to the diary. The user terminal 30 transmits the diary including the image data to the server 10. The server 10 inputs the diary including the image data to the server 20. The server 20 analyzes the image data, for example, by image analysis processing. The server 20 generates correction results according to the content of the image, for example, based on the results of the image analysis processing. As a result, the server 20 can recognize the content of the sentence based on the diary with higher accuracy by analyzing the content of the image, making it possible to improve correction accuracy compared to correction processing based on a diary that does not include image data.

[0123] In the above embodiment, the service has been described with diary correction as an example, but the diary may be set so that other users can view it. For example, the information processing unit 111 may enable users who publish their favorite diaries to be followed, as in a general SNS (Social Networking Service). The information processing unit 111 may also enable the original diary and the results of corrections to be viewed for the published diary. The information processing unit 111 may also attach advertisements to the published diary. This allows popular users to earn advertising revenue, as in a general SNS. This can motivate users to create diaries.

[0124] The information processing unit 111 may also have a billing unit. The billing unit may require a fee to provide the service. The billing unit may provide some services, such as diary corrections, free of charge, and other services, such as quizzes, subject to a fee. A configuration may be adopted in which diary corrections are free up to a certain number of entries, but require a fee for entries greater than the certain number. The billing unit may issue a certain number of tokens in response to a user's fee and reduce the number of tokens as the service is used, or may provide the service through a fixed-fee subscription. The number of tokens issued or the number of functions available through a subscription may be variable depending on the amount charged.

[0125] The acquisition unit 112 may also acquire the correction results from the generation AI as image data or a predetermined file format (Word, PDF, etc.). For example, an image of the created diary with corrections made in a predetermined color may be acquired. This makes it possible to improve the visibility of the corrections.

[0126] Furthermore, the information processing unit 111 may make public the correction settings, for example, the ranking of characters. For example, the correction settings set by the user may be displayed in a ranking format. In this case, the ranking may vary depending on the number of users and the number of times used. The above-mentioned charge amount may be changed depending on the character used. Furthermore, the information processing unit 111 may return a portion of the charge amount to the user who created a character ranked high (for example, 1st to 3rd place). The return method may be a method of reducing the monthly fee of the user who created a character ranked high, or a method of returning points.

[0127] Furthermore, the information processing unit 111 may send a summary instruction to the server 20 to create a summary of a diary created by the user or a diary that has been corrected, and may then obtain the result (summary) and send it to the user terminal 30. This allows the user to obtain a summary of the diary that they have created, which can be used when posting to an SNS, for example. [Explanation of symbols]

[0128] 1...information processing system, 10...server, 20...server, 30...processing device, 110...processor, 111...information processing unit, 112...acquisition unit, 113...input unit, 114...extraction unit, 115...recording unit, 116...creation unit, 117...generation unit, 118...presentation unit, 119...selection unit, 120...network communication interface, 130...storage device, 140...user interface, 150...communication bus, 310...processor, 311...information processing unit , 312...Communication control unit, 313...Display control unit, 314...Operation control unit, 320...Network communication interface, 330...Storage device, 340...User interface, 350...Communication bus, 351...Display, 352...Input device, A10...Area, A20...Area, B10 to B160...Buttons, C10 to C30...Area, CH1 to CH5...Characters, D10 to D120...Screen, N...Network, SB1 to SB9...Speech bubbles

Claims

1. The information processing device Obtaining text data relating to text created by a user; inputting the sentence data and correction instructions into a generating AI; Obtaining the correction results of the sentence data from the generation AI; extracting a predetermined portion from the correction result based on predetermined extraction conditions related to the correction; recording the predetermined portion in association with the user's identification information; An information processing method that performs the above.

2. 2. The information processing method according to claim 1, wherein said extracting step includes extracting an inappropriate portion of said text data from said correction result based on a portion of said text data that has been corrected.

3. Obtaining information regarding how the generation AI corrects the sentence; creating the correction instructions based on the correction method; The information processing method according to claim 1 , further comprising:

4. The information processing method according to claim 3 , wherein creating the correction instructions includes creating the correction instructions corresponding to information about a corrector included in the correction method.

5. obtaining information about the user's writing level; creating the correction instructions based on the level information; The information processing method according to claim 1 , further comprising:

6. extracting one or more common words from the plurality of sentence data; generating a theme for the text based on the one or more words; presenting a theme of the text to the user; The information processing method according to claim 1 , further comprising:

7. Generating a theme for the text includes: The information processing method according to claim 6, further comprising inputting the one or more words and an instruction to generate a theme of a sentence related to the one or more words to the generation AI.

8. The extracting step includes extracting an appropriate portion generated by the generation AI from the correction result based on the corrected portion of the sentence data; generating a message prompting the user to create a sentence based on the relevant portion; The information processing method according to claim 1 , further comprising:

9. The extracting step includes extracting an appropriate portion generated by the generation AI from the correction result based on the corrected portion of the sentence data; inputting the appropriate portion and a message generation instruction to the generation AI to prompt the user to create a sentence; The information processing method according to claim 1 , further comprising:

10. selecting at least one predetermined portion from a plurality of the predetermined portions; inputting a question creation instruction relating to the at least one selected predetermined portion into the generating AI; The information processing method according to claim 1 , further comprising:

11. the text data includes a first language and a second language; The information processing method according to claim 1 , wherein the correction result is expressed in the first language or the second language.

12. In the information processing device, Obtaining text data relating to text created by a user; inputting the sentence data and correction instructions into a generating AI; Obtaining the correction results of the sentence data from the generation AI; extracting a predetermined portion from the correction result based on predetermined extraction conditions related to the correction; recording the predetermined portion in association with the user's identification information; A program that executes the following.

13. Obtaining text data relating to text created by a user; inputting the sentence data and correction instructions into a generating AI; Obtaining the correction results of the sentence data from the generation AI; extracting a predetermined portion from the correction result based on predetermined extraction conditions related to the correction; recording the predetermined portion in association with the user's identification information; An information processing device that executes the above.

Images