Information processing apparatus, information processing method and program

Abstract

To provide an information processing apparatus, an information processing method, and a program that can infer interests (concerns) of a user or determine whether an answer of the user deviates from an expected topic.SOLUTION: An information processing apparatus 1 includes a questioning unit 10 for asking a user question to a user, a receiving unit 20 for receiving a user answer to the user question from the user, and an acquisition unit 30 that generates an input question for inferring interests using the user answer, acquires a first inferred interest from a generation and determination unit 200, or uses an input question for determining topic appropriateness regarding the user answer to acquire determination results of the topic appropriateness from the generation and determination unit 200.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program for receiving a response from a user. [Background technology]

[0002] Conventionally, attempts have been made to grasp user needs by asking a user a question and receiving the answer. For example, Patent Document 1 proposes an information processing device that, when providing a local government service on a chat-style screen, includes: a reception unit that receives from the user a selection of a menu item included in a displayed message or an answer to the question; a determination unit that refers to a storage unit that stores a hierarchical table in which the menu items or the answers to the question are associated one-to-one with information on a transition destination, and determines information on a transition destination corresponding to the received selection of the item or the answer to the question; a generation unit that generates a message displaying information on a new menu or the service based on the determined information on the transition destination; and a communication control unit that transmits the generated message to the user's terminal device. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-051551 Summary of the Invention [Problem to be solved by the invention]

[0004] Even when receiving answers from users, it may not be possible to understand what the user's interests are. Also, the answers from users may deviate from the expected topic.

[0005] The present invention provides an information processing device, an information processing method, and a program that can infer a user's interests or determine whether a response from a user deviates from an expected topic. [Means for solving the problem]

[0006] [Concept 1] The information processing device according to the present invention comprises: a questioning unit that asks a user a question; a receiving unit that receives a user answer to the user question from the user; an acquisition unit that acquires the inferred first interest from the generation determination unit using an input question for inferring an interest using the user's answer; may also be provided.

[0007] [Concept 2] In the information processing device according to concept 1, The receiving unit may receive an answer from the user to concretize the inferred first interest.

[0008] [Concept 3] In the information processing device according to concept 2, When the receiving unit receives an answer for concretizing the inferred first interest from the user, the obtaining unit may obtain the inferred second interest from the generation determining unit using the answer for concretizing the inferred first interest.

[0009] [Concept 4] In the information processing device according to concept 3, The receiving unit may receive information relating to the process of the inferred second interest from the user.

[0010] [Concept 5] In any one of the information processing devices of concepts 1 to 4, the acquisition unit causes the generation determination unit to determine the relevance to the process using the user's answer; When the acquisition unit determines that the relevance to the process is high based on the determination result by the generation determination unit, the question unit asks the user questions about specific details of the process, The receiving unit may receive a response from the user regarding specific details of the process.

[0011] [Concept 6] In the information processing device of concept 4 or 5, The acquisition unit may acquire an inferred third interest regarding the process from the generation determination unit using information regarding the process received from the user or a response to specific content of the process.

[0012] [Concept 7] In any one of the information processing devices of concepts 1 to 6, The acquisition unit may acquire a determination result of the theme suitability from the generation determination unit using an input question for determining the theme suitability of an answer input by the user.

[0013] [Concept 8] The information processing device according to the present invention comprises: a questioning unit that asks a user a question; a receiving unit that receives a user answer to the user question from the user; an acquisition unit that acquires a result of the determination of the theme suitability from the generation determination unit using an input question for determining the theme suitability of the user answer; may also be provided.

[0014] [Concept 9] The information processing method according to the present invention comprises: a step of asking a user question to a user by a questioning unit; receiving, by a receiving unit, a user answer to the user question from the user; acquiring, by an acquisition unit, an inferred first interest from a generation determination unit using an input question for inferring an interest using the user's answer; may also be provided.

[0015] [Concept 10] The information processing method according to the present invention comprises: a step of asking a user question to a user by a questioning unit; receiving, by a receiving unit, a user answer to the user question from the user; acquiring, by an acquisition unit, a result of the determination of the theme suitability from the generation determination unit using an input question for determining the theme suitability of the user answer; may also be provided.

[0016] [Concept 11] The program according to the present invention comprises: A program installed on a computer, On the computer where the program is installed, A function for asking a user question to a user; a function of receiving a user answer to the user question from the user; a function of acquiring a first inferred interest from a generation determination unit using an input question for inferring an interest using the user's answer; may be realized.

[0017] [Concept 12] The program according to the present invention comprises: A program installed on a computer, On the computer where the program is installed, A function for asking a user question to a user; a function of receiving a user answer to the user question from the user; a function of acquiring a result of the determination of the theme suitability from a generation determination unit using an input question for determining the theme suitability of the user answer; may be realized. [Effects of the Invention]

[0018] In one aspect of the present invention, when an input question for inferring an interest is generated using a user's answer and a first inferred interest is obtained from the generation determination unit, it is possible to infer the user's interest. In another aspect of the present invention, when an input question for determining thematic suitability of a user's answer is generated and a determination result of thematic suitability is obtained from the generation determination unit, it is possible to determine whether the user's answer deviates from the expected theme. [Brief explanation of the drawings]

[0019] [Figure 1] FIG. 10 is a diagram for explaining an example of an anchoring process performed in the information processing system according to the embodiment of the present invention. [Figure 2] 1 is a diagram for explaining an example of a process for grasping interests that is executed in an information processing system according to an embodiment of the present invention; [Figure 3] 10A and 10B are diagrams for explaining an example of processing for drilling down into specific content of interests, which is performed in the information processing system according to the embodiment of the present invention; [Figure 4] FIG. 10 is a diagram for explaining an example of processing for grasping process-related interests, which is executed in the information processing system according to the embodiment of the present invention. [Figure 5] 10A and 10B are diagrams for explaining an example of a process for confirming addition or termination performed in the information processing system according to the embodiment of the present invention; [Figure 6] 1 is a block diagram of an information processing system according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0020] Embodiment "composition" Hereinafter, an embodiment of an information processing device according to the present invention will be described. In this embodiment, an information processing method using the information processing device, a program installed to generate the information processing device, and a storage medium such as a USB or DVD storing the program are also provided. Also provided are programs installed on various terminals such as personal computers, smartphones, and tablets. In this application, "or" is a concept that includes "and," and the term "A or B" means either "A," "B," or "A and B."

[0021] The information processing device 1 (see FIG. 6) of this embodiment may be a system made up of multiple devices. Some or all of the elements making up the information processing device 1 may be provided in a server. Typically, an information processing system is made up of the information processing device 1 made up of a server or the like, and other devices such as multiple user terminals 100.

[0022] An information processing device 1, which may be a server or the like, is capable of communicating with multiple user terminals 100 (see FIG. 6). The user terminals 100 are terminals such as personal computers, smartphones, tablet terminals, and wearable computers such as smartwatches and smart glasses, and are typically mobile terminals such as smartphones and tablet terminals. The user terminal 100 may have an operation unit 120 for user input and a display unit 110 for displaying information. In a smartphone or tablet terminal, the operation display unit has the functions of the operation unit 120 and the display unit 110.

[0023] Responses from users and questions to users are made via user terminal 100. Input from an administrator and output to the administrator are made via one or more administrator terminals 150. Examples of administrator terminal 150 include terminals such as personal computers, smartphones, tablet terminals, and wearable computers such as smartwatches and smart glasses. Administrator terminal 150 may have an operation unit 170 through which the administrator makes inputs and a display unit 160 for displaying information. In a smartphone or tablet terminal, the operation display unit has the functions of operation unit 170 and display unit 160. Input from users and administrators may be made by text input or by voice input.

[0024] As shown in FIG. 6, the information processing device 1 may include a questioning unit 10 that asks a user a user question, a receiving unit 20 that receives a user answer to the user question from the user, and an acquiring unit 30 that generates an input question for determining thematic suitability of the user answer and acquires the determination result of thematic suitability from a generating and determining unit 200. The generating and determining unit 200 may use a generating AI such as ChatGPT, or may use an internal generating and determining unit that stores a prepared model in a storage unit 80 and outputs an answer by applying the model. Interactions with the user may be conducted in a conversational format by a chatbot. Interactions with the user may be associated with user identification information such as an identification number and stored in the storage unit 80, and past chat conversations with the user may be read from the storage unit 80 when the user logs in.

[0025] The information processing device 1 may have a communication unit 5 and may be capable of communicating with multiple user terminals 100, an administrator terminal 150, and a generation / determination unit 200 via the communication unit 5. The determination of theme suitability does not need to be performed at the beginning of an exchange with the chatbot, but may be performed during the exchange. Furthermore, the determination of theme suitability may be performed at the beginning and during an exchange with the chatbot, or may be performed multiple times. Adopting such an embodiment is beneficial in that it more reliably prevents user responses from deviating from the theme. The theme is set, for example, by an administrator and stored in the storage unit 80. The administrator may also store in advance in the storage unit 80 definition information for the results to be achieved and for the determination. The theme may include subthemes. For example, themes such as transportation, economy, and environment may be set as themes, with traffic congestion, impact on the local economy, noise and vibration, etc. set as subthemes.

[0026] By determining thematic suitability, the anchoring process (see Figure 1) can be performed. Figure 2 shows a mode in which the anchoring process is performed following the opening, but as mentioned above, the mode is not limited to this. Note that performing the anchoring process at the initial stage (following the opening) is beneficial in that it can prevent the user's answer from deviating from the theme at an early stage.

[0027] The anchoring process prevents any answer from deviating from the theme of the question. If an answer does not fit the theme of the question, the process repeatedly asks about its relevance to the theme, thereby bringing the dialogue back to the original theme of the question. For example, the generation and discrimination unit 200 determines the theme compatibility of a user's answer. If the compatibility is low, the questioning unit re-asks the user about the relevance to the theme of the question, thereby eliciting an answer related to the theme. For example, if the theme is "Purpose of Road Construction," a user response such as "I hope it will help stimulate the economy" would increase the theme compatibility, whereas a user response such as "I'm worried about the noise it will cause to the school" would decrease the theme compatibility. This is because the response "I hope it will help stimulate the economy" refers to economic revitalization, which is the "purpose of road construction," while the response "I'm worried about the noise it will cause to the school" refers to the user's own concerns and is not related to the "purpose of road construction." When the generation and determination unit 200 determines the theme suitability, the acquisition unit 30 may use a prompt stored in advance in the storage unit 80, for example, to input to the generation and determination unit 200 the following content: "The theme is set as 'Purpose of road construction.' If the answer is 'I want it to help stimulate the economy,' then the answer is judged to be 5 out of 5, and if the answer is 'I'm worried about the noise it will cause to schools,' then how many points out of 5 would the answer 'XX' get?" The generation and determination unit 200 may then evaluate the theme suitability of the user answer "XX." A typical example is that the user answer "XX" varies from user to user. A threshold may be set in advance, such that a score of 3 or more indicates high theme suitability, and a score below that score indicates low theme suitability, and the acquisition unit 30 may determine the level of theme suitability.

[0028] The prompt to be input to the generation determination unit 200 is input in advance by, for example, an administrator from the administrator terminal 150 and stored in the storage unit 80. An answer input by a user ("XXX" in the above example) may be embedded in the prompt by the acquisition unit 30 to generate input information, which is then input to the generation determination unit 200. This applies not only to the above example but also to examples described later. In other words, information such as an answer input by a user may be incorporated into a prompt stored in the storage unit 80 in correspondence with information to be generated or determined by the generation determination unit 200, to generate information to be input to the generation determination unit 200.

[0029] Even if the acquisition unit 30 determines that the user's answer has low thematic relevance, if the level of interest related to the process described below is high, the process may proceed to a process for confirming the interest related to the process described below. In this application, "interest" means "concern," and for example, "guessing interests" means guessing the user's interest, such as deep needs.

[0030] If the relevance is low multiple times in succession, the process may proceed to the addition / termination confirmation process shown in FIG. 5 without further investigation.

[0031] An example will be described using Figure 1. When a user inputs an answer to a question ("Answer" in Figure 1), the generation and determination unit 200, which receives the input from the acquisition unit 30, evaluates the suitability for the theme, and the acquisition unit 30 determines the level of suitability for the theme (S11 in Figure 1). If the suitability for the theme is high, the process proceeds to the interest level determination process (see Figure 2), which will be described later. On the other hand, if the suitability for the theme is low, the level of interest related to the process is determined (S12 in Figure 1), and if the level of interest related to the process is high, the process of grasping the interest related to the process shown in Figure 4 is performed.

[0032] When the generation and determination unit 200 determines the relevance to a process, for example, the acquisition unit 30 may input the following content to the generation and determination unit 200: "Please determine the level of relevance to the process. The theme is 'Purpose of road construction,' and if the answer is 'Aren't they hiding the negative impacts?', it is a comment about how to proceed, such as transparency, so it will be judged as 5 out of 5 points. If the answer is 'Vibration is more of a problem than noise,' it will be judged as 1 out of 5 points because there is no mention of how to proceed. How many points out of 5 would the answer 'XX' get?" and determine the level of relevance to the process for the user answer 'XX'. In this case, the acquisition unit 30 may determine that a given score (e.g., 3 points) or more indicates a high relevance to the process, but if the score is less than that score, it indicates a low relevance to the process.

[0033] On the other hand, if the level of interest related to the process is low, the user's answer is summarized and a question is asked to the user regarding the relationship of the answer to the theme, prompting the user to enter a different answer. For example, a question such as, "Your answer was ____, but how does it relate to theme ____? Please answer in a way that makes the relationship with theme ____ clear" may be asked to prompt the user to enter a different answer. The same process is performed for the second user answer, but if the user's answer is again deemed low in theme relevance, the process proceeds to the addition / termination confirmation process shown in Figure 5. The "process" refers to a process related to the theme, such as the transparency, fairness, participation opportunities, and method (way of doing things) of the process.

[0034] 1 to 5 show the user terminal 100, and hatched content basically indicates input or output from the user terminal 100. 1 to 5 also show the generation determination unit 200, but non-hatched content indicates that it may be performed using the generation determination unit 200, and typically processing is performed by the generation determination unit 200 and other units other than the generation determination unit 200, such as the acquisition unit 30 and the interrogation unit 10.

[0035] Before the anchoring process shown in FIG. 1, an opening interaction may take place with the user via a chatbot (see FIG. 2). In the opening, a start declaration is made, and then information such as consent to recording, use, consent to the reflection process, consent to the use of the generation AI, access to and understanding of prior information, subject attribute information, and intention to continue may be acquired from the user, and may be stored in the storage unit 80 in association with user identification information such as a user ID. "Anchoring" shown in FIG. 2 refers to the anchoring process shown in FIG. 1. As described above, the anchoring process may be performed at an appropriate timing. For example, each time the user inputs an answer, the suitability of the answer to the theme may be determined, and if the user's answer deviates from the theme, the user may be prompted to input an answer that is in line with the theme.

[0036] When the anchoring process for theme compatibility is completed, the acquisition unit 30 may input the user's answer together with a prompt to the generation determination unit 200, causing the generation determination unit 200 to determine the interest level of the user's answer (S21 in FIG. 2). If the acquisition unit 30 determines that the interest level of the user's answer is low, the question unit 10 may ask the user a question inquiring about the user's interest, and this may be repeated until the interest level reaches a certain level. However, an upper limit may be set for the number of repetitions, and if the interest level does not seem to be increasing, the process may proceed to the addition / end confirmation process shown in FIG. 5 without further investigation.

[0037] As an example of a high level of interest, for example, when the theme is set as "Purpose of road construction" as mentioned above, if the user inputs a response such as "I'm worried about traffic accidents," the level of interest will be high, while if the user inputs a response such as "I want a crosswalk installed," the level of interest will be low. The content "I'm worried about traffic accidents" expresses an interest for the user, so the level will be high, but the external request "I want a crosswalk installed" is not an interest for the user, so the level will be low. When the generation and determination unit 200 determines the level of interest, for example, the acquisition unit 30 may input the following content to the generation and determination unit 200: "Please determine the level of interest of the questioner. The theme is set as 'Purpose of road construction.' If the answer "I'm worried about traffic accidents" is judged to be 5 out of 5, and if the answer "I want a crosswalk installed," it is judged to be 1 out of 5. How many points out of 5 would the answer "XX" be?", and the acquisition unit 30 may then determine the level of interest for the user answer "XX." In this case, the acquisition unit 30 may determine that the level of interest is high if the answer is equal to or higher than a predetermined score (for example, 3 points), and low if the answer is below that score.

[0038] The acquisition unit 30 may generate an input question for inferring an interest using the user's answer, and acquire an inferred first interest from the generation determination unit 200. In this case, an anchoring process may be performed to acquire an inferred first interest for a user opinion linked to the theme of the dialogue, and the interest may be identified by confirming the inferred first interest with the user, and the inferred first interest may be stored together with the original opinion in the storage unit 80. Acquiring an inferred first interest from the generation determination unit 200 in this manner is beneficial in that it allows the user's interests to be grasped.

[0039] As mentioned above, interests are concerns and represent situations desired by the user. Conversely, complaints and dissatisfaction that are not relevant to the user will have a low interest level because they are not of interest to the user. For example, the acquisition unit 30 may input a question such as "The theme is set as 'Purpose of road construction,' and the user is asked the question, 'What do you expect from road construction?' and the answer is 'XX.' Can you guess this user's interest in 'Purpose of road construction,'" to the generation and determination unit 200, and the first inferred interest inferred from the user's answer, 'XX.'

[0040] An example will be described using FIG. 2. When the anchoring process is completed, the generation determination unit 200 determines the interest level (S21 in FIG. 2). If the interest level is high, the generation determination unit 200 proceeds to a process of acquiring an inferred first interest. On the other hand, if the interest level is low, the user's interests are summarized, and a question about the interest is asked to the user, prompting the user to input an answer regarding the interest. The acquisition unit 30 determines the interest level of the user's answer to the interest using the generation determination unit 200 (S22 in FIG. 2). If the interest level of the user's answer is low, the user's interests are summarized again, and the question unit 10 asks the user a question about the interest, prompting the user to input an answer regarding the interest. On the other hand, if the interest level is high, the generation determination unit 200 proceeds to a process of acquiring an inferred first interest. If the acquisition unit 30 using the generation determination unit 200 determines that the interest level of the user's answer is low even after receiving the second answer regarding the interest, the process proceeds to the addition / end confirmation process of FIG.

[0041] When the generation determination unit 200 acquires the inferred first interest, the system checks with the user whether or not the user agrees with the inferred first interest (S23 in FIG. 2). If the user agrees, the system stores the inferred first interest in the storage unit 80, and proceeds to the process of digging deeper into the specific content of the interest shown in FIG. 3. At this time, the storage unit 80 may store the inferred first interest together with the user's identification information, such as a user ID, the date and time, etc. On the other hand, if the user does not agree, the system returns to a1 in FIG. 2, summarizes the user's interests, and then asks the user questions about the interests, prompting the user to input a response regarding the interests.

[0042] The receiving unit 20 may receive an answer from the user to specify the inferred first interest. This is advantageous in that information specifying the inferred first interest can be obtained from the user.

[0043] For example, the questioning unit 10 may ask the user a question inquiring about specific content, and if the specificity of the answer reaches a certain level, the questioning unit 10 may rephrase the answer as an interest including the specific content and confirm it with the user. The storage unit 80 may record the interest including the specific content for which the user's consent has been confirmed, or the user's answer for which no interest could be inferred. If the specific content cannot be obtained after asking the question a certain number of times, the process may proceed to a step of grasping process-related interests, which will be described later, without pursuing further.

[0044] An example of a high level of specificity would be a user response such as "Air pollution and noise get worse on slopes," while a user response such as "It's bad for the environment" would result in a low level of specificity. To set the level of specificity, prompts incorporating location, time, time period, target, etc. may be prepared and input to the generation and determination unit 200.

[0045] For example, an answer such as "Air pollution and noise get worse on slopes" has a high level of specificity because it provides detailed information such as the type and degree of environmental pollution, location, and reason, whereas an answer such as "It's bad for the environment" has a low level of specificity because the details are unclear. When the generation / discrimination unit 200 determines the level of specificity, the acquisition unit 30 may input, for example, the following content to the generation / discrimination unit 200: "Please determine the level of specificity of the answer. The theme is "Purpose of road development." The answer "Air pollution and noise get worse on slopes" is given a score of 5 out of 5, and the answer "It's bad for the environment" is given a score of 1 out of 5. How many points out of 5 would the answer "XX" get?" The acquisition unit 30 may then evaluate the level of specificity of the user's answer "XX." In this case, the acquisition unit 30 may determine that a given score (e.g., 3 points) or higher is a high level of specificity, whereas a score below that score is a low level of specificity.

[0046] When the receiving unit 20 receives an answer from the user for specifying the inferred first interest, the obtaining unit 30 may use the answer to specify the inferred first interest to generate an input question for inferring a specific interest, and obtain an inferred second interest that is more specific than the inferred first interest as an output answer. Employing such an embodiment is beneficial in that it allows for a more specific understanding of the user's interests.

[0047] For example, a question such as "The theme is set as 'Purpose of road construction,' and the question, 'What are your interests regarding road construction?' is asked, and the user answers, 'XX.' Can you guess this user's more specific interests regarding the 'purpose of road construction?'?" may be input to the generation and determination unit 200, and an inferred second interest inferred from the user's answer, 'XX.' may be acquired. Alternatively, an input question may be asked to summarize specific interests, and an inferred second interest that is more specific than the inferred first interest may be acquired as an output answer. For example, a question such as "The theme is set as 'Purpose of road construction,' and the question, 'What are your interests regarding road construction?' is asked, and the user answers, 'XX.' Can you summarize this user's more specific interests regarding the 'purpose of road construction?'?" may be input to the generation and determination unit 200, and an inferred second interest summarized based on the user's answer, 'XX.'

[0048] An example will be described using FIG. 3. The questioning unit 10 asks the user a question prompting them to provide an answer to concretize the inferred first interest. When the answer to concretize the inferred first interest is received from the user via the user terminal 100 by the receiving unit 20, the acquiring unit 30 uses the generation determining unit 200 to determine the level of concreteness of the answer (S31 in FIG. 3). If the acquiring unit 30 determines that the level of concreteness of the user's answer is high, the process proceeds to acquiring an inferred second interest. On the other hand, if the acquiring unit 30 determines that the level of concreteness of the user's answer is low, the process proceeds to receiving another answer from the user to concretize the inferred first interest by providing a specific example. The acquiring unit 30 then uses the generation determining unit 200 to determine the level of concreteness of the answer again (S31 in FIG. 3). If the level of concreteness is still low after the second input, the process proceeds to grasping the process-related interest shown in FIG. 4.

[0049] If the acquisition unit 30 determines, based on the output from the generation determination unit 200, that the level of specificity of the user's answer is high, the acquisition unit 30 acquires the inferred second interest from the generation determination unit 200. The acquisition unit 30 then asks the user whether they agree with the inferred second interest (S32 in FIG. 3). If the user agrees, the inferred second interest is stored in the storage unit 80, and the process proceeds to the process for grasping the process-related interests shown in FIG. 4. At this time, the storage unit 80 may store the inferred second interest together with the user's identification information, such as a user ID, the date and time, etc. On the other hand, if the user does not agree, the process returns to b1 in FIG. 3 and the system again accepts an answer from the user to concretize the inferred first interest. At this time, the system may display past answers from the user that have been concretized in a series of interactions and prompt the user to enter a different answer. From then on, the process may be performed as described above.

[0050] The receiving unit 20 may ask the user a question about the process of the estimated second interest and receive a response from the user. If the level of interest in the process is determined to be high in S12 of FIG. 1, the questioning unit 10 may generate a question about the specific content of the process and ask the user. Asking a question to grasp the user's interest in the process is beneficial in that it allows the user to grasp the user's interest in the process. A question about the process may be, for example, "I'm asking about the theme 'Purpose of Road Construction.' Please tell me what interests you about the 'Road Construction' process." This is beneficial in that it determines the level of interest in the process, such as the process approach that influences the content of the theme, and distinguishes between the two aspects of "content" and "process" to grasp the interest. The generation and determination unit 200 may determine the process relevance of the answer. If the process relevance is low, the question may be asked again. If the process relevance is low multiple times, the system may proceed to the addition / termination confirmation process shown in FIG. 5. When information about a process is received from a user, the generation determination unit 200 may generate a question for the user.

[0051] An example of high process relevance is when a user inputs a response such as "They're hiding the negative impacts," which would result in high process relevance, whereas a user input such as "Vibration is more of a problem than noise" would result in low process relevance. The response "They're hiding the negative impacts," which is a statement regarding how to proceed (e.g., transparency), would have a high level of process relevance, whereas the response "Vibration is more of a problem than noise" would have a low level of process relevance because it does not mention how to proceed. When the generation and discrimination unit 200 determines the level of process relevance, for example, the acquisition unit 30 may input the following content to the generation and discrimination unit 200: "Please determine the level of process relevance. The theme is "Purpose of road construction." The response "They're hiding the negative impacts," which is a statement regarding how to proceed (e.g., transparency), would be given a score of 5 out of 5, while the response "Vibration is more of a problem than noise" would be given a score of 1 out of 5 because it does not mention how to proceed. How many points out of 5 would the response "XX" receive?" The acquisition unit 30 may then input the following content to the generation and discrimination unit 200, which would then evaluate the level of process relevance for the user response "XX." In this case, if the score is equal to or greater than a predetermined score (for example, 3 points), the level of process relevance is high, but if the score is less than the predetermined score, the level of specificity may be determined to be low by the acquisition unit 30. This question is asked, for example, in S42 of FIG. 4, which will be described later.

[0052] The storage unit 80 may record process-related interests for which the user's consent has been confirmed and user responses for which no process-related interest has been inferred. Instead of or while receiving information about the process of the inferred second interest from the user, the receiving unit 20 may also receive information about the process of the inferred first interest from the user.

[0053] The acquisition unit 30 may use the process-related information received from the user to generate an input question for inferring an interest in the process, and may acquire an inferred third interest in the process as an output answer. For example, a question such as "The theme is 'Purpose of road construction,' and the user has responded with 'XX' as an interest in the process. Can you infer the user's specific interest in the process related to 'road construction'?" may be input to the generation and determination unit 200, and an inferred third interest inferred from the user's answer, 'XX', may be acquired.

[0054] An example will be described using FIG. 4. The questioning unit 10 asks the user a question prompting an answer regarding a process-related interest. When the answer regarding the process-related interest input by the user terminal 100 is received by the receiving unit 20 from the user, the generation and determination unit 200 receives input from the acquiring unit 30 and determines the level of interest in the answer (S41 in FIG. 4). As described above, the determination of the level of interest may be performed, for example, by inputting the following content to the generation and determination unit 200: "Please determine the level of interest of the questioner. The theme is 'Purpose of road development.' If the answer "I'm worried about traffic accidents" is determined to be 5 points out of 5, and if the answer "I want a pedestrian crossing installed," it is determined to be 1 point out of 5. How many points out of 5 would the answer "XX" be?" The acquiring unit 30 may input this content to the generation and determination unit 200, and the level of interest in the answer "XX" from the user (an answer regarding a process that includes specific details) may be determined.

[0055] If the acquisition unit 30 determines that the answer from the user has a high level of interest, the generation determination unit 200 determines the level of process relevance in the answer (S42 in FIG. 4).

[0056] 4, if the acquisition unit 30 determines that the interest level of the answer from the user is low, or if the acquisition unit 30 determines that the process relevance level of the answer from the user is low in S42 of Fig. 4, the acquisition unit 30 again accepts an answer to the interest related to the process from the user. At this time, a question may be generated using the previous answer from the user, such as "Your previous answer was 'XX', but I'm interested in your interest related to the process. Can you tell me your opinion on something related to the process and your interest?"

[0057] If the level of process relevance of the user's answer is high in S42 of FIG. 4, the acquisition unit 30 causes the generation determination unit 200 to generate and acquire an inferred third interest inferred from the user's answer. Then, the user is asked whether they agree with the inferred third interest (S43 of FIG. 3). If the user agrees, the inferred third interest is stored in the storage unit 80, and the process proceeds to the addition or termination confirmation process shown in FIG. 5. At this time, the storage unit 80 may store the inferred third interest together with user identification information such as a user ID, date and time, etc. On the other hand, if the user does not agree, the process returns to c2 of FIG. 4 and accepts a response from the user to the interest related to the process again. At this time, the user's previous response may be displayed and the user may be prompted to enter a different response.

[0058] If the interest level of the user's response is determined to be low twice in S41 of FIG. 4, or if the process relevance level of the user's response is determined to be low twice in S42 of FIG. 4, the process proceeds to the addition or termination confirmation process shown in FIG. 5.

[0059] In the final step shown in Figure 5, the participant is asked whether they wish to express any additional opinions (S51 in Figure 5), and if they wish to express additional opinions, they may return to the anchoring process shown in Figure 1 and perform a new reframing process (the series of processes shown in Figures 2 to 5) (or return to "Additional Opinion (D)" in Figure 2). On the other hand, if they do not wish to express any additional opinions, they may perform a closing process. Reframing in this embodiment is a series of processes shown in Figures 2 to 5 that are carried out with the aim of converting positions such as demands and criticisms toward the other party, such as the government, into interests such as deep needs, concerns, and expectations toward oneself.

[0060] In the closing process, the user's opinions may be summarized, their identified interests may be presented, relevant information sources may be introduced, contact information may be provided, and then a thank you and closing statement may be presented to the user via the chatbot.

[0061] By carrying out the above process, it is possible to understand the interests behind the user's position and reframe them by digging deeper into the details. It is also possible to distinguish and understand interests related to the decision-making process, such as transparency and fairness, communication with users such as citizens, and the procedures for deciding on organizers. By repeating the reframing process to express additional opinions at the user's request, it is also possible to understand other opinions that the user was unable to fully express.

[0062] The analysis unit 90 may analyze the results based on the results of the user dialogue, such as interests. The analysis unit 90 may automatically generate analysis results, such as event information such as the implementation date, time, and number of accesses, accessor attribute distribution, required time distribution, response count distribution, prior information access and awareness status, interests (deep needs, concerns, expectations, etc.) / positions (requests, criticism, etc.), detailed interests (deep needs, concerns, expectations, etc.) / positions (requests, criticism, etc.), process-related interests (deep needs, concerns, expectations, etc.) / positions (requests, criticism, etc.), position classification category recommendations, reframing results such as interest classification category recommendations, category-specific interests, category classification positions, and position-specific interests, and output the results to one or more customer terminals 190 of customers, such as clients, via the communication unit 5. The customer terminals 190 may be, for example, terminals such as personal computers, smartphones, tablet terminals, and wearable computers such as smart watches and smart glasses. The customer terminal 190 may have an operation unit 191 for the administrator to input data and a display unit 192 for displaying information. In a smartphone or a tablet terminal, the operation display unit has the functions of the operation unit 191 and the display unit 192.

[0063] The position classification category recommendation proposal may be made by inputting multiple positions input by the user together with a prompt to the generation and determination unit 200, and the acquisition unit 30 may acquire it as output information. Similarly, the interest classification category recommendation proposal may be made by inputting multiple interests input by the user together with a prompt to the generation and determination unit 200, and the acquisition unit 30 may acquire it as output information. The multiple positions and multiple interests do not need to use all of the information input by the user, and only a portion may be used. The category classification interests are the results of interests by category generated using the recommendation proposal, the category classification positions are the results of positions by category generated using the recommendation proposal, and the position category interests are the results of interests divided by category in the position, for example, showing the interests of a person who takes a position of "opposition" to a certain policy (for example, "I'm worried about noise").

[0064] Each element of the present embodiment, such as the interrogation unit 10, the reception unit 20, the acquisition unit 30, and the analysis unit 90, may be realized by one or more IC chips or electronic modules, or may be realized by a circuit configuration. The information processing device 1 has a processor, and the processor may execute a program to realize various functions of the interrogation unit 10, the reception unit 20, the acquisition unit 30, the analysis unit 90, and the like of the information processing device 1 of the present embodiment.

[0065] The above-mentioned description of the embodiment and the disclosure of the drawings are merely examples for explaining the invention described in the claims, and the invention described in the claims is not limited by the above-mentioned description of the embodiment or the disclosure of the drawings. Furthermore, the description of the claims as originally filed is merely an example, and the description of the claims may be changed as appropriate based on the description in the specification, drawings, etc. [Explanation of symbols]

[0066] 1. Information processing equipment 5. Communications Department 10 Question Section 20 Reception 30 Acquisition Department 80 Memory Department 90 Analysis Department 200 Generate the discrimination part

Claims

1. a questioning unit that asks a user a question; a receiving unit that receives a user answer to the user question from the user; an acquisition unit that acquires the inferred first interest from the generation determination unit using an input question for inferring an interest using the user's answer; An information processing device comprising:

2. The information processing device according to claim 1 , wherein the accepting unit accepts an answer from the user for realizing the inferred first interest.

3. The information processing device according to claim 2 , wherein when the receiving unit receives an answer from the user for concretizing the inferred first interest, the obtaining unit obtains the inferred second interest from the generation determining unit using the answer for concretizing the inferred first interest.

4. The information processing device according to claim 3 , wherein the receiving unit receives information about the process of the inferred second interest from a user.

5. the acquisition unit causes the generation determination unit to determine the relevance to the process using the user's answer; When the acquisition unit determines that the relevance to the process is high based on the determination result by the generation determination unit, the question unit asks the user questions about specific details of the process, The information processing device according to claim 1 , wherein a reception unit receives a response to specific content of the process from the user.

6. The information processing device according to claim 4 , wherein the acquisition unit acquires the inferred third interest about the process from the generation determination unit using information about the process received from the user or a response to specific content of the process.

7. The information processing device according to claim 1 , wherein the acquisition unit acquires a result of the determination of the theme suitability from the generation determination unit using an input question for determining the theme suitability of an answer input by the user.

8. a questioning unit that asks a user a question; a receiving unit that receives a user answer to the user question from the user; an acquisition unit that acquires a result of the determination of the theme suitability from the generation determination unit using an input question for determining the theme suitability of the user answer; An information processing device comprising:

9. a step of asking a user question to a user by a questioning unit; receiving, by a receiving unit, a user answer to the user question from the user; acquiring, by an acquisition unit, an inferred first interest from a generation determination unit using an input question for inferring an interest using the user's answer; An information processing method comprising:

10. a step of asking a user question to a user by a questioning unit; receiving, by a receiving unit, a user answer to the user question from the user; acquiring, by an acquisition unit, a result of the determination of the theme suitability from the generation determination unit using an input question for determining the theme suitability of the user answer; An information processing method comprising:

11. A program installed on a computer, On the computer where the program is installed, A function for asking a user question to a user; a function of receiving a user answer to the user question from the user; a function of acquiring a first inferred interest from a generation determination unit using an input question for inferring an interest using the user's answer; A program that makes this happen.

12. A program installed on a computer, On the computer where the program is installed, A function for asking a user question to a user; a function of receiving a user answer to the user question from the user; a function of acquiring a result of the determination of the theme suitability from a generation determination unit using an input question for determining the theme suitability of the user answer; A program that makes this happen.

Images