Information processing device and information processing method

Abstract

An information processing device 10 according to an embodiment of the present invention comprises: a question acquisition unit 12 that acquires a question sentence asking for a user's opinion about a prescribed question subject; a generation unit 13 that generates a plurality of answer sentences which include an answer sentence expressing an affirmative opinion regarding the question subject and an answer sentence expressing a negative opinion regarding the question subject and which are adjusted so that the opinion regarding the question subject is neutral overall; and a presentation unit 14 that presents the question sentence together with the plurality of answer sentences to the user.

Description

Information processing device and information processing method

[0001] One aspect of the present disclosure relates to an information processing device and an information processing method.

[0002] Patent Document 1 discloses a system for determining whether the opinions of responses to a questionnaire are positive or negative based on the responses.

[0003] Japanese Patent Application Laid-Open No. 2005-322239

[0004] According to the mechanism described in the above-mentioned Patent Document 1, it is possible to determine a user's opinion (positive or negative) about a predetermined question target from the user's response to a questionnaire (question sentence) about the question target. However, if the user has been exposed to biased opinions about the question target (for example, many negative opinions), psychological biases such as the herding effect (conformity effect) may occur, and it may be impossible to properly elicit the user's true answer (i.e., the user's opinion in a situation where the above-mentioned psychological bias does not exist).

[0005] Therefore, an object of one aspect of the present disclosure is to provide an information processing device and an information processing method that can appropriately elicit a user's opinion on a predetermined question target.

[0006] An information processing device according to one aspect of the present disclosure includes a question acquisition unit that acquires a question sentence requesting a user's opinion on a specified question subject, a generation unit that generates a plurality of answer sentences that include answer sentences that have a positive opinion on the question subject and answer sentences that have a negative opinion on the question subject, and that are adjusted so that the opinions on the question subject are generally neutral, and a presentation unit that presents the question sentence together with the plurality of answer sentences to the user.

[0007] An information processing method according to another aspect of the present disclosure includes the steps of: acquiring a question sentence requesting a user's opinion on a specified question subject; generating a plurality of answer sentences, including answer sentences with a positive opinion on the question subject and answer sentences with a negative opinion on the question subject, and adjusted so that the opinions on the question subject are generally neutral; and presenting the question sentence along with the plurality of answer sentences to the user.

[0008] According to one aspect of the present disclosure, it is possible to appropriately elicit a user's opinion on a predetermined question subject.

[0009] FIG. 1 is a block diagram showing the overall configuration of a system including an information processing device according to an embodiment; FIG. 2 is a diagram showing an example of question information stored in a memory unit; FIG. 3 is a diagram showing an example of answer information stored in a memory unit; FIG. 4 is a diagram showing an example of a prompt output to a generation AI model in a first processing example of the generation unit; FIG. 5 is a diagram showing an example of a prompt output to a generation AI model in a third processing example of the generation unit; FIG. 6 is a diagram showing an example of a screen presented to a user by a presentation unit; FIG. 7 is a diagram schematically showing an example of a second processing example of the answer acquisition unit; FIG. 8 is a flowchart showing an example of the operation of an information processing device; FIG. 9 is a diagram showing an example of the hardware configuration of an information processing device.

[0010] Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same or equivalent elements are designated by the same reference numerals, and redundant description will be omitted.

[0011] Fig. 1 is a diagram showing the overall configuration of a system including an information processing device 10 according to one embodiment. The system shown in Fig. 1 includes the information processing device 10, a user terminal 20, and a generation AI server 30. The information processing device 10 is configured to be able to communicate with each of the user terminal 20 and the generation AI server 30 via any communication network.

[0012] The information processing device 10 presents a question sentence requesting an opinion on a predetermined question target to a user using the user terminal 20. In addition, the information processing device 10 presents, together with the question sentence, a plurality of answer sentences that have been adjusted so that the opinion on the question target is generally neutral, in order to suppress the occurrence of a psychological bias in the user toward the question target.

[0013] The user terminal 20 is a terminal device used by a user who receives questions from the information processing device 10. The user terminal 20 may take any form, such as a personal computer, a smartphone, a tablet terminal, a feature phone, a server device, or a game console. Although only one user terminal 20 is illustrated in FIG. 1 , the information processing device 10 may execute the above-described questions for each of multiple users. In this case, the system illustrated in FIG. 1 may include multiple user terminals 20 used by each of the multiple users.

[0014] The generation AI server 30 includes a generation AI model 31 and is a device that generates and provides content using the generation AI model 31. The generation AI model 31 is a model that, in response to a prompt input from the information processing device 10 (in this embodiment, the generation unit 13 and the answer acquisition unit 15), generates content (in this embodiment, text indicating an answer to the question from the generation unit 13 and the answer acquisition unit 15) according to any one or a combination of the instructions, context, question, and output format indicated by the prompt, and returns the content as response information. A prompt is information indicating an instruction or question input to the generation AI model 31 in an interactive system such as a dialogue with the generation AI model 31 or a command line interface (CLI). The prompt can include various types of input information. In this case, the generation AI model 31 generates response information targeted at the various types of input information. The type and format of the input information are not particularly limited, and the input information can include, for example, data files with file names including a predetermined extension, such as text data, image data, application-related data, audio data, video data, and still image data. Application-related data is data such as document data, table data, and graph data that can be processed by a predetermined application program.

[0015] The generative AI model 31 may be configured to include, for example, a large-scale language model (LLM) and a user interface (UI) for interacting with the user. The generative AI model 31 may be an interactive AI model capable of text-based or voice-based chat with the user. Examples of such a generative AI model 31 include ChatGPT, GPT-3.5, GPT-4V, and PaLM2. However, the generative AI model 31 may also be an AI model other than the large-scale language models described above. Furthermore, the generative AI model 31 may be located on a generative AI server 30 separate from the information processing device 10, as shown in FIG. 1, or may be located on the information processing device 10. Alternatively, the generative AI model 31 may be located on a user terminal 20, such as Tsuzumi. In this case, each function of the information processing device 10 may also be located on the user terminal 20. In other words, the user terminal 20 itself may function as the information processing device 10 of this embodiment.

[0016] The information processing device 10 includes a storage unit 11 , a question acquisition unit 12 , a generation unit 13 , a presentation unit 14 , and an answer acquisition unit 15 .

[0017] The storage unit 11 stores various data handled by the information processing device 10. In this embodiment, the storage unit 11 is implemented on the information processing device 10, but the storage unit 11 may be implemented outside the information processing device 10 (for example, on a network). In this embodiment, the storage unit 11 stores question information and answer information.

[0018] FIG. 2 is a diagram illustrating an example of question information. The question information is information in which a unique question ID and a question sentence are associated with each other for each question. The question information can be arbitrarily registered, added, or deleted by an operator of the information processing device 10, etc. In the example of FIG. 2, a question ID "Q001" is associated with the question sentence "What do you think about service X?" As in the above example, the question sentence included in the question information seeks the user's (answerer's) opinion on the question subject included in the question sentence (in this example, "service X"). However, the wording of the question sentence is not limited to the above example, and other wordings such as "Do you like XX?" and "What is your impression of XX?" may be used.

[0019] FIG. 3 is a diagram illustrating an example of answer information. The answer information is information that associates a question ID, an answer sentence, and an opinion (the opinion of the answerer of the answer sentence regarding the question target corresponding to the question ID). In the example of FIG. 2, the answer sentence "It's good in that it has XX points!" indicating a positive opinion regarding the question target "Service X" for the question "What do you think about Service X?" with question ID "Q001" and the answer sentence "It's a bit disappointing in that it has △△ points" indicating a negative opinion are stored. In this embodiment, the answer information is registered in the storage unit 11 by processing of the answer acquisition unit 15, which will be described later. Alternatively, the answer information may be registered in advance as an example answer sentence by an operator of the information processing device 10, etc.

[0020] The question acquisition unit 12 acquires a question requesting a user's opinion on a predetermined question subject. In this embodiment, the question acquisition unit 12 acquires one question (e.g., a question "What do you think about service X?" corresponding to question ID "Q001") by referring to question information ( FIG. 2 ) stored in the storage unit 11. However, the method by which the question acquisition unit 12 acquires a question is not limited to the above. For example, an operator operating the information processing device 10 may input a question he or she wants to pose to the user on the spot. The question acquisition unit 12 may then acquire the question thus input. In this embodiment, the following description will be given using an example in which the question acquisition unit 12 acquires the question "What do you think about service X?"

[0021] The generation unit 13 generates a plurality of answer sentences, including answer sentences with positive views on the question target and answer sentences with negative views on the question target, and adjusted so that the views on the question target are generally neutral. In this embodiment, the question target is "Service X." Here, "adjusting so that the views on the question target are generally neutral" typically means adjusting so that the number of answer sentences with positive views (hereinafter "number of positive answers") matches the number of answer sentences with negative views (hereinafter "number of negative answers"), but is not limited to the above. For example, "adjusting so that the views on the question target are generally neutral" includes adjusting so that the number of positive answers and the number of negative answers are approximately equal (including cases where one is slightly greater than the other (e.g., when the difference is equal to or less than a predetermined threshold)), and does not necessarily require adjusting so that the number of positive answers and the number of negative answers are strictly equal. Below, several examples of the processing by the generation unit 13 are described.

[0022] (First processing example of the generation unit 13) The generation unit 13 may cause the generation AI model 31 to generate multiple answer sentences by inputting instruction information (prompt) to the generation AI model 31 that includes at least the content of the question sentence and content that instructs the generation of multiple answer sentences that are adjusted so that the views on the question subject are generally neutral.

[0023] 4, the generation unit 13 may generate a prompt P1 corresponding to instruction information including at least a question sentence, "What do you think about service X?" and an instruction content such as, "Generate N answer sentences with positive views and N answer sentences with negative views in succession, as if they were having a discussion in a text chat," and output the generated prompt P1 to the generation AI server 30. Here, "N" included in the prompt P1 is actually an arbitrarily determined integer.

[0024] For example, a template including a parameter Pq corresponding to a question may be prepared in advance, such as "Generate N positive and N negative response sentences in succession for the question 'Pq', as if they were being discussed in a text chat." In this case, the generation unit 13 can easily generate a prompt P1 corresponding to the question by inputting the question "What do you think about service X?" into the parameter Pq of the template.

[0025] The generation AI server 30, which has acquired the prompt P1 from the information processing device 10 (generation unit 13), inputs the prompt P1 into the generation AI model 31 and returns the output result from the generation AI model 31 to the information processing device 10 (generation unit 13). In this way, the generation unit 13 can acquire multiple answer sentences generated by the generation AI model 31.

[0026] According to the first processing example of the generation unit 13 described above, by using the generation AI model 31, it is possible to appropriately and easily generate multiple answer sentences that include a balanced mix of both positive and negative views and are adjusted so that the views on the question subject are generally neutral.

[0027] (Second Processing Example of Generation Unit 13) The generation unit 13 may generate a plurality of answer sentences for a second user different from the first user, based on the answer sentence of the first user stored in the storage unit 11. Typically, the first user is a user who has answered the same question sentence in the past, and the second user is a user to whom the question is to be asked this time.

[0028] For example, when generating a plurality of answer sentences for a second user, if an answer sentence previously given by a first user (e.g., "It's good that it has XX points!" in FIG. 3 ) is stored in the storage unit 11 in association with the first user's opinion ("positive" in the above example), the generation unit 13 may generate a plurality of answer sentences for the second user using the answer sentence "It's good that it has XX points!" For example, the generation unit 13 may include the answer sentence "It's good that it has XX points!" as an example of a positive answer sentence in the plurality of answer sentences to be presented to the second user.

[0029] Note that there may be multiple first users (i.e., users who have been asked the same question and answered it before the second user) for a second user who is scheduled to be asked a question. When answer information of multiple first users (e.g., answer information including multiple answer sentences that express a positive view and multiple answer sentences that express a negative view regarding question ID "001" (see FIG. 3)) is stored, the generation unit 13 may randomly extract multiple answer sentences from the multiple answer sentences that express a positive view and the multiple answer sentences that express a negative view so that the number of positive answers matches the number of negative answers. Then, the generation unit 13 may randomly arrange the multiple answer sentences extracted in this way to generate multiple answer sentences that are adjusted so that the views regarding the question target are generally neutral.

[0030] According to the second processing example of the generation unit 13 described above, when asking a question to a second user, the quality of the multiple answer sentences presented to the second user can be improved by using the answer sentences of the first user who answered the same question in the past.

[0031] (Third Processing Example of Generation Unit 13) The generation unit 13 may use the generative AI model 31 as in the first processing example, and may generate multiple answer sentences based on answer sentences stored in the storage unit 11 as in the second processing example. That is, when the generation unit 13 causes the generative AI model 31 to generate multiple answer sentences for the second user, the generation unit 13 may include information on answer sentences corresponding to at least one opinion of the first user stored in the storage unit 11 in the instruction information (in this embodiment, a prompt for the generative AI model 31).

[0032] 5, the generation unit 13 may generate a prompt P2 corresponding to instruction information including at least a question, "What do you think about service X?", and an instruction, "Generate N answer sentences with positive views and N answer sentences with negative views in succession, as if they were being discussed in a text chat," and output the generated prompt P2 to the generation AI server 30. The prompt P2 differs from the prompt P1 in that it includes an additional supplementary item, "Please refer to the answer sentence examples," and includes "#positive answer sentence example," which is information on answer sentences corresponding to positive views (sentences previously answered by the first user), and "#negative answer sentence example," which is information on answer sentences corresponding to negative views (sentences previously answered by the first user).

[0033] According to the third processing example of the generation unit 13 described above, it is possible to obtain the effects of both the first and second processing examples described above.

[0034] In the example of FIG. 5 , the instruction information (prompt P2) includes information on answer sentences corresponding to each of the multiple opinions (positive and negative) of the multiple first users. However, the instruction information may include information on answer sentences corresponding to any one of the opinions. In the example of FIG. 5 , answer sentences previously given by the first user are included in the instruction information as an example. However, the positioning of the answer sentences previously given by the first user in the instruction information is not limited to the above. For example, the generation unit 13 may include in the prompt P2 an instruction such as, "Generate N answer sentences with positive opinions and N answer sentences with negative opinions in succession, as if discussing in a text chat, so as to include an affirmative answer sentence such as, 'That's a good thing!' and a negative answer sentence such as, 'That's a bit disappointing.'"

[0035] In this embodiment, by the processing of the generation unit 13 as described above, multiple answer sentences are generated for the question "What do you think about service X?", including answer sentences that express positive views on the question target "service X" of the question sentence (for example, "It's good that it has that ○○ point!", "I like it too!", etc.) and answer sentences that express negative views (for example, "It's a bit disappointing that it has that △△ point.", "I understand! I hope they improve that!", etc.), and are adjusted to be neutral overall (in this embodiment, so that "the number of positive answers = the number of negative answers").

[0036] The presenting unit 14 presents the question sentence to the user together with the plurality of answer sentences generated by the generating unit 13. Fig. 6 is a diagram showing an example of a screen presented to the user by the presenting unit 14 (a screen displayed on a display provided in the user terminal 20).

[0037] 6 , a screen 50 presented to a user by the presentation unit 14 includes a virtual space image 51 showing a virtual space including the user's avatar A, a question image 52 including a question, and a display area 53 in which a plurality of answers generated by the generation unit 13 are displayed. In the example of FIG. 6 , the question image 52 is represented as a speech bubble indicating a question from a character N (an NPC object simulating an animal in the example of FIG. 6 ) that is an NPC located in front of the avatar A in the virtual space. In addition, a plurality of answers are displayed in the display area 53 like a text chat between multiple people.

[0038] Note that the manner in which the presenting unit 14 presents the question sentence and the multiple answer sentences is not limited to the manner in which text is displayed on the screen 50 as in the example of FIG. 6 . For example, the presenting unit 14 may output the question sentence and the multiple answer sentences as audio. For example, the presenting unit 14 may output the audio of the question sentence and then output the audio of the multiple answer sentences. In this case, the question sentence image 52 and the display area 53 may not be displayed on the screen 50. Alternatively, the presenting unit 14 may display the question sentence image 52 and output the audio of the question sentence. Similarly, the presenting unit 14 may display the display area 53 and output the audio of the multiple answer sentences.

[0039] The answer acquisition unit 15 acquires the answer of the user to the question presented to the user as in the example of FIG. 6, and the user's opinion on the question subject.

[0040] The answer acquiring unit 15 may acquire the user's answer sentence by, for example, acquiring the text of the answer sentence input by the user via an input device (e.g., a keyboard, a touch panel display, etc.) included in the user terminal 20. Alternatively, the answer acquiring unit 15 may acquire the user's answer sentence by acquiring voice information input by the user via a voice input device (e.g., a microphone, etc.) included in the user terminal 20 and performing voice recognition processing on the voice information.

[0041] Furthermore, the answer acquisition unit 15 acquires the user's opinion on the question subject, for example, in the following manner.

[0042] (First Processing Example of the Answer Acquisition Unit 15) The answer acquisition unit 15 may acquire the user's opinion on the question target "service X" of the question sentence "What do you think about service X?" by inquiring of the user. For example, when the answer acquisition unit 15 acquires an answer sentence such as "I think it's XX" from the user, the answer acquisition unit 15 may subsequently inquire of the user, such as "Is your answer positive, negative, or neutral?" The answer acquisition unit 15 may then acquire the user's opinion by acquiring the user's answer to the inquiry (positive, negative, or neutral). In this case, the answer acquisition unit 15 may display answer buttons of "positive," "negative," and "neutral" on the screen 50 and have the user perform an operation (e.g., a click operation, a touch operation, etc.) to select one of the answer buttons. In this case, the answer acquisition unit 15 may acquire the opinion corresponding to the answer button selected by the user as the user's opinion.

[0043] According to the first processing example of the answer acquisition unit 15 described above, by directly confirming the user's opinion on the question subject with the user, it is possible to more reliably obtain the user's opinion corresponding to the user's answer sentence.

[0044] (Second Processing Example of Answer Acquisition Unit 15) The answer acquisition unit 15 may acquire a user's opinion by inputting the user's answer sentence into an estimation model configured to be able to input any answer sentence to a question sentence and output an estimation result of an opinion on the question target. As an example, the estimation model may be the generative AI model 31 used by the generation unit 13 when generating multiple answer sentences. Alternatively, the estimation model may be a trained model generated by predetermined machine learning (e.g., a neural network, a multilayer neural network (deep learning), etc.) separately from the generative AI model 31. As an example in the present embodiment, the generative AI model 31 is used as the estimation model.

[0045] 7 is a diagram schematically illustrating an example of the second processing example of the answer acquisition unit 15. As shown in Fig. 7, the answer acquisition unit 15 may generate a prompt P3 including at least the question "What do you think about service X?" and content instructing the user to answer the question with their opinion (opinion on the question target) of "I think it's XX.", and output the generated prompt P3 to the generation AI server 30.

[0046] For example, a template "Regarding the question 'Pq,' do you think the opinion 'Pa' is positive, negative, or neutral?" may be prepared in advance, including a parameter Pq corresponding to the question sentence and a parameter Pa corresponding to the user's answer sentence. In this case, the answer acquisition unit 15 can easily generate the prompt P3 by inputting the question sentence "What do you think about service X?" into the parameter Pq of the template and inputting "I think it's XX" into the parameter Pa.

[0047] The generation AI server 30, which has acquired the prompt P3 from the information processing device 10 (answer acquisition unit 15), inputs the prompt P3 into the generation AI model 31 and returns the output result from the generation AI model 31 to the information processing device 10 (answer acquisition unit 15). In this way, the answer acquisition unit 15 can acquire the inference result (either positive, negative, or neutral) output by the generation AI model 31.

[0048] 7, the answer acquiring unit 15 acquires an estimated result of the opinion "negative." Furthermore, the record information R that associates the user's answer sentence "I think it's XX" acquired by the answer acquiring unit 15 with the user's opinion "negative" is newly stored (registered) in the storage unit 11 as the above-mentioned answer information (see FIG. 3).

[0049] According to the second processing example of the answer acquisition unit 15 described above, the user's opinion on the question target is estimated by the estimation model (generative AI model 31), thereby making it unnecessary to query the user as in the first processing example. As a result, it is possible to acquire the user's opinion (estimated result) corresponding to the user's answer sentence while reducing the user's effort (annoyance).

[0050] The first and second processing examples of the answer acquisition unit 15 may be used in combination. For example, the answer acquisition unit 15 may execute the second processing example to acquire an estimation result (e.g., "negative") from the generative AI model 31, and then query (confirm) the user by notifying the user of a message such as, "Is it correct to understand that your answer is 'negative' regarding the question subject?" According to the combined method, after acquiring an estimation result of the user's opinion from the generative AI model 31, the user can be queried as to whether the result is correct (YES or NO). This reduces the burden on the user in responding, while reliably obtaining the correct user opinion.

[0051] Next, an example of the operation (information processing method) of the information processing device 10 will be described with reference to Fig. 8. Fig. 8 is a flowchart showing an example of the operation of the information processing device 10.

[0052] In step S1, the question acquisition unit 12 acquires a question sentence requesting a user's opinion on a specific question subject (for example, a question sentence requesting an opinion on the question subject "service X": "What do you think about service X?").

[0053] In step S2, the generation unit 13 generates a plurality of answer sentences, including answer sentences with positive views on the question target and answer sentences with negative views on the question target, and adjusted so that the views on the question target are generally neutral. The generation unit 13 generates the plurality of answer sentences, for example, by executing any one of the first to third processing examples of the generation unit 13 described above.

[0054] In step S3, the presenting unit 14 presents to the user the question sentence acquired in step S1 together with the plurality of answer sentences generated in step S2. As an example, the presenting unit 14 presents to the user a screen 50 including the question sentence (question sentence image 52) and the plurality of answer sentences (display area 53), as shown in FIG.

[0055] In addition, in step S2, when the first processing example of the generation unit 13 described above is executed, the following steps S4 and S5 may be omitted. On the other hand, in step S2, when the second or third processing example of the generation unit 13 described above is executed, the following steps S4 and S5 are executed for the current question recipient user (first user) in order to generate multiple answer sentences for future question recipient users (second users).

[0056] In step S4, the answer acquisition unit 15 acquires the user's answer to the question and the user's opinion on the question target. The answer acquisition unit 15 acquires, for example, the user's answer input to the user terminal 20 after step S3. The answer acquisition unit 15 also acquires the user's opinion on the question target by executing the first processing example or the second processing example of the answer acquisition unit 15 described above.

[0057] In step S5, the storage unit 11 stores the information (e.g., record information R in FIG. 7) that associates the user's answer sentence and opinion acquired in step S4 as new answer information (see FIG. 3). As a result, the answer information (record information R) stored this time can be used when generating multiple answer sentences for the user to whom a question is to be asked next time (step S2).

[0058] In the information processing device 10 described above, when a question is posed to a user requesting the user's opinion on a specific question target, the question is not presented to the user alone, but is presented to the user along with a plurality of answer sentences, including answer sentences with positive views on the question target and answer sentences with negative views on the question target, and adjusted so that the user's opinion on the question target is generally neutral. According to the above configuration, by placing the user in a situation where the user's opinion on the question target is generally neutral, it is possible to effectively suppress the occurrence of psychological biases (i.e., psychological biases biased toward either positive or negative opinions) in the user, such as the herding effect (conformity effect). Therefore, according to the information processing device 10 and its information processing method, by presenting the question to the user in a state where such psychological bias is suppressed, it is possible to appropriately elicit the user's opinion on the specific question target.

[0059] In the above embodiment, the generation unit 13 generates a plurality of answer sentences that are neutral overall by generating a balance of answer sentences corresponding to negative views and answer sentences corresponding to positive views. However, the generation unit 13 may also include answer sentences that have a neutral view on the question target among the plurality of answer sentences. For example, in response to the question, "What do you think about service X?", the generation unit 13 may include an answer sentence that shows a neutral view, such as, "At this point, I can't decide whether it's good or bad." among the plurality of answer sentences. In this way, by mixing neutral answer sentences that are neither positive nor negative among the plurality of answer sentences presented to the user (i.e., by increasing the variety of views on the question target), the occurrence of the above-mentioned psychological bias can be more effectively suppressed.

[0060] [Additional Notes] The present disclosure includes the following aspects.

[0061] [1] An information processing device comprising: a question acquisition unit that acquires a question sentence requesting a user's opinion on a specified question subject; a generation unit that generates a plurality of answer sentences including answer sentences with a positive opinion on the question subject and answer sentences with a negative opinion on the question subject, and that are adjusted so that the opinions on the question subject are generally neutral; and a presentation unit that presents the question sentence together with the plurality of answer sentences to the user.

[0062] [2] The information processing device of [1], wherein the generation unit inputs instruction information to the generative AI model that includes at least the content of the question sentence and content instructing the generation of multiple answer sentences that are adjusted so that the views on the question subject are generally neutral, thereby causing the generative AI model to generate the multiple answer sentences.

[0063] [3] An information processing device according to [1] or [2], further comprising: an answer acquisition unit that acquires the user's answer to the question and the user's opinion on the question subject; and a memory unit that stores the user's answer in association with the user's opinion, wherein the generation unit generates the plurality of answer sentences for a second user different from the first user based on the answer sentence of a first user stored in the memory unit.

[0064] [4] The information processing device of [3], wherein the generation unit inputs instruction information to a generative AI model, the instruction information including at least the content of the question sentence and content instructing the generative AI model to generate a plurality of answer sentences that are adjusted so that the views on the question target are generally neutral, and when the generation unit causes the generative AI model to generate the plurality of answer sentences for the second user, the generation unit includes information on answer sentences that correspond to at least one view of the first user stored in the memory unit in the instruction information.

[0065] [5] The information processing device of [3] or [4], wherein the answer acquisition unit acquires the user's opinion on the question subject by making an inquiry to the user.

[0066] [6] The information processing device according to any one of [3] to [5], wherein the answer acquisition unit acquires the user's opinion by inputting the user's answer sentence into an estimation model configured to be able to input any answer sentence to the question sentence and output an estimated result of an opinion on the question subject.

[0067] [7] The information processing device according to any one of [1] to [6], wherein the generation unit includes, among the plurality of answer sentences, answer sentences that have a neutral view on the question subject.

[0068] [8] An information processing method including the steps of: acquiring a question sentence requesting a user's opinion on a specified question subject; generating a plurality of answer sentences including answer sentences with a positive opinion on the question subject and answer sentences with a negative opinion on the question subject, and adjusted so that the opinions on the question subject are generally neutral; and presenting the question sentence together with the plurality of answer sentences to the user.

[0069] [Definition of Terms, etc.] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are directly or indirectly connected (e.g., using wires, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining software with the single device or the multiple devices.

[0070] Functions include, but are not limited to, judging, determining, calculating, computing, processing, deriving, investigating, searching, verifying, receiving, transmitting, outputting, accessing, resolving, selecting, choosing, establishing, comparing, expecting, anticipating, regarding, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assigning.

[0071] For example, the information processing device 10 according to an embodiment of the present disclosure may function as a computer that performs the information processing method of the present disclosure. Fig. 9 is a diagram illustrating an example of a hardware configuration of the information processing device 10 according to an embodiment of the present disclosure. The information processing device 10 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0072] In the following description, the term "apparatus" can be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the information processing device 10 may be configured to include one or more of the apparatuses shown in Fig. 9, or may be configured to exclude some of the apparatuses.

[0073] Each function of the information processing device 10 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0074] The processor 1001 controls the entire computer by running, for example, an operating system, and may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc.

[0075] The processor 1001 also reads programs (program code), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with the programs. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, each functional unit of the information processing device 10 (e.g., the generation unit 13, etc.) may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be made for other functional blocks. While the above-described various processes have been described as being executed by a single processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0076] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be called a register, a cache, a main memory (primary storage device), etc. The memory 1002 can store executable programs (program codes), software modules, etc. for implementing an information processing method according to an embodiment of the present disclosure.

[0077] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0078] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also called, for example, a network device, a network controller, a network card, or a communication module.

[0079] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. The input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0080] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0081] The information processing device 10 may also be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by the hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0082] Although the present embodiment has been described in detail above, it is clear to those skilled in the art that the present embodiment is not limited to the embodiment described in this specification. The present embodiment can be implemented in modified and altered forms without departing from the spirit and scope of the present invention as defined by the claims. Therefore, the description in this specification is intended to be illustrative and does not have any limiting meaning on the present embodiment.

[0083] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0084] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be transmitted to another device.

[0085] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0086] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0087] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0088] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0089] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0090] Furthermore, the information, parameters, etc. described in this disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information.

[0091] The names used for the parameters described above are not intended to be limiting in any way. Furthermore, the mathematical formulas and the like that use these parameters may differ from those explicitly disclosed in this disclosure. The various information elements may be identified by any suitable names, and the various names assigned to these various information elements are not intended to be limiting in any way.

[0092] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0093] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0094] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0095] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0096] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0097] 10...information processing device, 11...memory unit, 12...question acquisition unit, 13...generation unit, 14...presentation unit, 15...answer acquisition unit, 20...user terminal, 30...generation AI server, 31...generation AI model

Claims

1. An information processing device comprising: a question acquisition unit that acquires a question sentence that requests a user's opinion on a specified question subject; a generation unit that generates a plurality of answer sentences that include answer sentences that have a positive opinion on the question subject and answer sentences that have a negative opinion on the question subject, and that are adjusted so that the opinions on the question subject are generally neutral; and a presentation unit that presents the question sentence together with the plurality of answer sentences to the user.

2. The information processing device described in claim 1, wherein the generation unit causes the generative AI model to generate the plurality of answer sentences by inputting instruction information to the generative AI model that includes at least the content of the question sentence and content that instructs the model to generate a plurality of answer sentences that are adjusted so that the views on the question subject are generally neutral.

3. An information processing device as described in claim 1, further comprising: an answer acquisition unit that acquires the user's answer to the question and the user's opinion on the question subject; and a memory unit that stores the user's answer in association with the user's opinion, wherein the generation unit generates the plurality of answer sentences for a second user different from the first user based on the answer sentence of a first user stored in the memory unit.

4. The information processing device described in claim 3, wherein the generation unit causes the generative AI model to generate the plurality of answer sentences by inputting instruction information to the generative AI model, the instruction information including at least the content of the question sentence and content instructing the generative AI model to generate a plurality of answer sentences that are adjusted so that the views on the question target are generally neutral, and when the generation unit causes the generative AI model to generate the plurality of answer sentences for the second user, the generation unit includes information on answer sentences that correspond to at least one view of the first user stored in the memory unit in the instruction information.

5. The information processing device according to claim 3, wherein the answer acquisition unit acquires the user's opinion on the question by making an inquiry to the user.

6. The information processing device described in claim 3, wherein the answer acquisition unit acquires the user's opinion by inputting the user's answer sentence into an estimation model configured to be able to input any answer sentence to the question sentence and output an estimation result of an opinion on the question subject.

7. The information processing device according to claim 1, wherein the generation unit includes, among the plurality of answer sentences, answer sentences that have a neutral view on the question subject.

8. An information processing method comprising the steps of: acquiring a question sentence requesting a user's opinion on a specified question subject; generating a plurality of answer sentences including answer sentences with a positive opinion on the question subject and answer sentences with a negative opinion on the question subject, and adjusted so that the opinions on the question subject are generally neutral; and presenting the question sentence together with the plurality of answer sentences to the user.

Images