Device and method

Abstract

This device comprises: a reception unit that receives information pertaining to problems of a city; a determination unit that determines at least one among information pertaining to the credibility of the information pertaining to the problems and information pertaining to the severity of the information pertaining to the problems on the basis of the information pertaining to the problems; a decision unit that decides information pertaining to a response policy on the basis of the information pertaining to the problems and the at least one among the information pertaining to the credibility and the information pertaining to the severity; and a generation unit that generates, on the basis of the information pertaining to the response policy, a prompt for instructing generative AI to generate response content.

Description

Apparatus and method

[0001] The present invention relates to an apparatus and method for generating a prompt to instruct the generation of answer content for information about defects in a city.

[0002] Patent document 1 describes a device that, upon receiving an inquiry from a user regarding infrastructure supply services, automatically generates a response based on information about the residence, information about the infrastructure supply services, and information entered into an input form.

[0003] Japanese Patent Application Laid-Open No. 2020-126406

[0004] However, since infrastructure supply services and problems in a town are different from each other, the technology described in Patent Document 1 cannot instruct the generation of answer content for information about problems in a town.

[0005] Therefore, an object of the present disclosure is to provide an apparatus and method capable of instructing the generation of answer content for information relating to defects in a town.

[0006] The device disclosed herein includes a reception unit that receives information regarding defects in a city; a determination unit that determines, based on the information regarding the defect, at least one of information regarding the credibility of the information regarding the defect and information regarding the seriousness of the information regarding the defect; a determination unit that determines information regarding an answer policy based on at least one of the information regarding the defect, the credibility, and the seriousness; and a generation unit that generates, based on the information regarding the answer policy, a prompt to instruct a generation AI to generate answer content.

[0007] According to the present disclosure, it is possible to instruct the generation of answer content for information regarding defects in a town.

[0008] FIG. 1 is a diagram showing the system configuration of an infrastructure management system including a prompt generation device according to an embodiment of the present disclosure. FIG. 2 is a block diagram showing the functional configuration of the prompt generation device. FIG. 3 is an example of a score table for information related to malfunctions. FIG. 4 is an example of a score table for information stored in a knowledge database. FIG. 5 is an example of a score table for information related to malfunctions. FIG. 6 is an example of a score table for information stored in a knowledge database. FIG. 7 is a flowchart showing the operation of the prompt generation device. FIG. 8 is a diagram showing an example of a prompt. FIG. 9 is a diagram showing an example of the hardware configuration of a prompt generation device according to an embodiment of the present disclosure.

[0009] The present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0010] 1 is a diagram showing the system configuration of an infrastructure management system including a prompt generation device 100 according to an embodiment of the present disclosure. As shown in the figure, this system includes the prompt generation device 100, a large-scale language model 200, and a knowledge database 300. These servers and databases are configured to be communicatively connected via a network.

[0011] Residents report defects to the government by inputting information about defects in the town into user terminal 400. Hereinafter, information about defects in the town will be referred to as "information about the defect." A resident who inputs information about a defect into user terminal 400 is a reporter who has reported a defect to the government. When information about a defect is input into user terminal 400, user terminal 400 transmits the input information about the defect to prompt generation device 100.

[0012] When the prompt generation device 100 receives information about a malfunction from the user terminal 400, it determines at least one of information about the credibility of the information about the malfunction and information about the severity of the information about the malfunction, based on the information about the malfunction. Hereinafter, information about the credibility of the information about the malfunction will be referred to as information about credibility. Also, information about the severity of the information about the malfunction will be referred to as information about severity.

[0013] When the prompt generation device 100 determines at least one of information regarding the credibility of the information regarding the malfunction and information regarding the seriousness of the information regarding the malfunction, it determines information regarding the response policy based on the information regarding the malfunction and at least one of information regarding the credibility of the information regarding the malfunction and information regarding the seriousness of the information regarding the malfunction.

[0014] When the prompt generation device 100 determines the information on the answer policy, it generates a prompt for instructing the generation AI to generate answer content based on the information on the answer policy and transmits this prompt to the large-scale language model 200. The large-scale language model 200 generates answer content based on this prompt and transmits the generated answer content to the prompt generation device 100. When the prompt generation device 100 acquires the answer content generated by the large-scale language model 200, it transmits the acquired answer content to the user terminal 400. In this way, the resident who is the reporter acquires the answer content for the defect that he or she has reported.

[0015] The large-scale language model 200 is a generative AI model that can generate content in response to a prompt containing input information, according to the instructions, context, question, and output format indicated by the prompt, and return the content as response information. In this disclosure, a prompt refers to information indicating an instruction or question entered by a user in an interactive system such as a dialogue with a generative AI model or a command line interface (CLI). The prompt can also include input information, in which case the generative AI model generates response information targeted at the input information. The generative AI model may be, for example, an interactive AI model that includes a large-scale language model (LLM) and a user interface (UI) for dialogue with a user, enabling text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, PaLM2, etc.

[0016] All or some of the large-scale language model 200 and the knowledge database 300 may be located in the prompt generation device 100 or the user terminal 400. The user terminal 400 may also have the functions of the prompt generation device 100 and function as the prompt generation device 100. There is a type of generation AI model, such as "tsuzumi," in which the generation AI model is located inside the user terminal 400. In this type, a Retrieval-Augmented Generation (RAG) application is also provided in the user terminal 400. However, the information (knowledge database) accessed by the RAG application may be located inside the user terminal 400 or on a network. There is also a type, such as ChatGPT, in which the generation AI model is located on a network. In this type, the RAG application is provided in the user terminal 400. However, the information (knowledge database) accessed by the RAG application is located on a network.

[0017] 2 is a block diagram showing the functional configuration of the prompt generation device 100. The prompt generation device 100 includes a receiving unit 101, a determining unit 102, a determining unit 103, a prompt generation unit 104, and an LLM access unit 105.

[0018] The receiving unit 101 receives information about a defect input to the user terminal 400 from the user terminal 400, and extracts various information included in the received information about the defect. The receiving unit 101 also receives answer content generated by the large-scale language model 200, and transmits the received answer content to the user terminal 400.

[0019] The information about defects received by the receiving unit 101 may be information about defects in infrastructure facilities in the city, such as broken playground equipment, potholes in the road, cracks in the road, unlit lights, and illegal dumping of garbage.

[0020] The information related to the malfunction may include, for example, information indicating the details of the malfunction, information indicating the infrastructure facility where the malfunction occurs, information indicating the date and time, information indicating the location, information encouraging an action such as repair or maintenance, information indicating the person who reported the information about the malfunction, or a photograph or video showing the state of the malfunction. Hereinafter, information encouraging an action such as repair or maintenance will be referred to as information encouraging an action.

[0021] The information prompting a response may include, for example, information indicating the level of response or information regarding the urgency. The information indicating the informant may include, for example, any of the informant's identification information that uniquely identifies the informant, the informant's name, the informant's address, the informant's attribute information, the informant's past application history, etc. This information indicating the informant may be stored in the knowledge database 300. If the information indicating the informant is stored in the knowledge database 300, when the prompt generation device 100 acquires the informant's identification information from the information related to the defect, the prompt generation device 100 may acquire other information about the informant that corresponds to the informant's identification information from the knowledge database 300, and accept this acquired information as information indicating the informant.

[0022] For example, when information about a malfunction is received, such as "Yesterday, when I was walking down XX street, the light was out and it was dangerous. Please do something about it," the information about this malfunction includes information indicating the date and time, information indicating the location, details of the malfunction, information about the malfunctioning infrastructure, and information urging action such as repair or maintenance. Also, when information about a malfunction is received, such as "There is a hole in the road. It is dangerous if a child falls in, so please close it immediately," the information about this malfunction includes information about the malfunctioning infrastructure, and information urging action such as repair or maintenance.

[0023] The receiving unit 101 may set required information for the information regarding the malfunction in advance. The required items may be, for example, information indicating the malfunctioning infrastructure facility, information indicating the location, and information indicating the person who reported the information regarding the malfunction. If the required information is not entered into the user terminal 400, the receiving unit 101 may ask the resident about the required items via the user terminal 400 and prompt the resident to enter the required items into the user terminal 400. This asking may be performed, for example, by a well-known chatbot or the like.

[0024] The determination unit 102 determines at least one of information regarding the credibility of the information regarding the defect and information regarding the seriousness of the information regarding the defect, based on the information regarding the defect. The determination unit 102 may determine at least one of information regarding the credibility of the information regarding the defect and information regarding the seriousness of the information regarding the defect, using the information regarding the defect and the knowledge database 300. In the present embodiment, as an example, the determination unit 102 determines both information regarding the credibility of the information regarding the defect and information regarding the seriousness of the information regarding the defect.

[0025] The information regarding credibility is information indicating the credibility of the declaration. The information regarding credibility may be, for example, information indicating the degree of credibility, such as "reliable," "needs verification," or "unknown," or may be information that scores the credibility. The information that scores the credibility may be, for example, a score output by applying a scoring method using an existing AI model. Note that "reliable" is information that indicates high credibility. "needs verification" is information that indicates low credibility. "Unknown" is information that indicates that the credibility is unknown.

[0026] When determining information about credibility using information about a defect, for example, a score may be assigned to the photograph, location information, etc. included in the information about the defect, and the sum of these scores may be used as information about credibility. Alternatively, a weighted score may be assigned to the photograph, location information, etc., and the sum of these scores may be used as information about credibility.

[0027] FIG. 3 is an example of a score table for information about defects. In the example shown in FIG. 3, the information about defects includes items for photos and information indicating a location. In the photo item, if there are multiple photos, the score is 5 points, if there is one photo, the score is 4 points, and if there is no photo, the score is 2 points. In the location information item, if there is information about a location, the score is 3 points, and if there is no information about a location, the score is 1 point. The determination unit 102 may then determine the information about credibility by adding up the scores registered in this score table from the photos and information about the location included in the information about the defect.

[0028] If the knowledge database 300 stores information about the defect reporting history or the percentage of responses to the reporter's past reports, the determination unit 102 may determine information about credibility based on, for example, the information about the defect reporting history stored in the knowledge database 300 or the percentage of responses to the reporter's past reports. The percentage of responses to the reporter's past reports is the percentage of the number of responses actually taken to the number of the reporter's past reports. The information about the defect reporting history may include, for example, the reporter's past reporting history and the reporter's attribute information. In this case, for example, scores may be assigned to the reporter's past reporting history and the reporter's attribute information, and the sum of these scores may be used as information about credibility. Furthermore, weighted scores may be assigned to the reporter's past reporting history and the reporter's attribute information, and the sum of these scores may be used as information about credibility.

[0029] FIG. 4 is an example of a score table for information stored in the knowledge database. In the example shown in FIG. 4, the information stored in the knowledge database includes an item for the percentage of response to the informant's past reports. In the item for the percentage of response to the informant's past reports, 80-100% is given 5 points, 60-80% is given 4 points, 40-60% is given 3 points, 20-40% is given 2 points, 0-20% is given 1 point, and if this is the informant's first report, 3 points. The determination unit 102 may then determine information regarding credibility based on the scores registered in this score table for the percentage of response to the informant's past reports stored in the knowledge database 300.

[0030] The determination unit 102 may determine the information regarding the credibility by adding up the scores described above. Alternatively, the determination unit 102 may determine the information regarding the credibility using a score with 100 points as the maximum score.

[0031] The information on severity is information that indicates the severity of the impact that a defect in a town has on the lives of residents, etc. The information on severity may be, for example, information that indicates the degree of severity, such as "serious," "normal," "minor," or "no problem," or may be information that scores the severity. The information that scores the severity may be, for example, a score output by applying a scoring method using an existing AI model. Note that "serious" is information that indicates a high degree of seriousness. "normal" is information that indicates a medium degree of seriousness. "minor" is information that indicates a low degree of seriousness. "No problem" is information that indicates a zero degree of seriousness.

[0032] When using information about a defect to determine information about the severity, for example, scores may be assigned to the defect content, location, photograph, etc. included in the information about the defect, and the sum of these scores may be used as information about the credibility. Also, weighted scores may be assigned to the defect content, location, photograph, etc., and the sum of these scores may be used as information about the credibility.

[0033] FIG. 5 is an example of a score table for information about defects. In the example shown in FIG. 5, the information about the defect includes items such as defect content, location, and photo judgment. The photo judgment is performed by performing image processing on the photos included in the information about the defect using a generative AI model or the like to determine the degree of the defect. In the item about the defect content, broken playground equipment is given a score of 5, potholes in the road a score of 4, unlit lights a score of 2, and illegal dumping of garbage a score of 1. In the item about the location, a location with many users is given a score of 5, a location with an average number of users a score of 3, and a location with few users a score of 1. In the item about the photo judgment, a major defect is given a score of 5, a minor defect a score of 2, and a minor defect a score of 1. The judgment unit 102 may then judge the information about the severity by adding up the scores registered in this score table based on the defect content, location, and photo included in the information about the defect.

[0034] If the knowledge database 300 stores the number of reported defects, the determination unit 102 may determine information related to the severity based on the number of reported defects stored in the knowledge database 300. The number of reported defects is the number of reports of the same defect from other residents (other user terminals 400). In this case, for example, a score may be assigned to the number of reported defects, and the sum of these scores may be used as information related to the severity. Alternatively, a weighted score may be assigned to the number of reported defects, and the sum of these scores may be used as information related to the severity.

[0035] 6 is an example of a score table for information stored in the knowledge database. In the example shown in FIG. 6, the information stored in the knowledge database includes an item for the number of reported cases. In the item for the number of reported cases, 5 points are assigned if there are 5 or more reports, 4 points if there are 3 to 4 reports, 3 points if there are 1 to 2 reports, and 2 points if there are 0 reports. The determination unit 102 may then determine information regarding the severity from the scores registered in this score table for the number of reported cases stored in the knowledge database 300.

[0036] The determination unit 102 may determine the information regarding the severity by adding up the scores described above. Alternatively, the determination unit 102 may determine the information regarding the severity using a score with 100 points as the maximum score.

[0037] The determination unit 103 determines information regarding a response policy based on information regarding the defect and at least one of information regarding credibility and information regarding severity. In the present embodiment, the determination unit 103 determines information regarding a response policy based on both information regarding credibility and information regarding severity, for example. If the knowledge database 300 stores information such as the current status of the defect or a future response plan, the determination unit 103 may also determine information regarding a response based on the current status of the defect or a future response plan stored in the knowledge database 300.

[0038] The information regarding the response policy is information indicating a policy for creating response content for residents who have input information regarding the defect into the user terminal 400. The information regarding the response policy may be information regarding how to respond to the defect, such as "immediate response," "on-site check," or "wait and see." "Immediate response" is information indicating that the defect will be responded to immediately. "On-site check" is information indicating that the site will be checked first and then a response will be considered. "Wait and see" is information indicating that a response will be postponed for the time being.

[0039] When the information on the credibility indicates higher reliability than the set condition, the determination unit 103 may determine "immediate response" or "on-site check" as the information on the response policy. When the information on the credibility does not indicate higher reliability than the set condition, the determination unit 103 may determine "on-site check" or "wait and see" as the information on the response policy.

[0040] For example, if the score of the information regarding credibility is 60 points or more out of 100 points, the determination unit 103 may determine "immediate response" as the information regarding the response policy. Also, if the score of the information regarding credibility is 30 points or more but less than 60 points out of 100 points, the determination unit 103 may determine "on-site inspection" as the information regarding the response policy. Also, if the score of the information regarding credibility is less than 30 points out of 100 points, the determination unit 103 may determine "wait and see" as the information regarding the response policy.

[0041] For example, if the information on the severity indicates a severity higher than the set condition, the determination unit 103 may determine "immediate response" or "on-site check" as the information on the response policy. Also, if the information on the severity does not indicate a severity higher than the set condition, the determination unit 103 may determine "on-site check" or "wait and see" as the information on the response policy.

[0042] For example, if the score of the information regarding the severity is 60 or more out of 100, the determination unit 103 may determine "immediate response" as the information regarding the response policy. Alternatively, if the score of the information regarding the severity is 30 or more but less than 60 out of 100, the determination unit 103 may determine "on-site inspection" as the information regarding the response policy. Alternatively, if the score of the information regarding the severity is less than 30 out of 100, the determination unit 103 may determine "wait and see" as the information regarding the response policy.

[0043] If the information regarding the response policy determined based on the information regarding credibility and the information regarding the response policy determined based on the information regarding the severity are different from each other, the decision unit 103 may prioritize either one when determining the information regarding the response policy.

[0044] If the knowledge database 300 stores the current status of the defect or future response plans, the determination unit 103 may determine, as information regarding the response policy, information such as "immediate response," "on-site check," or "wait and see," plus information such as the current status of the defect or future response plans.

[0045] The prompt generation unit 104 generates a prompt for instructing the generation AI to generate answer content based on information related to the answer policy. The prompt may include, for example, at least one of a role, a task, a condition, etc. The role is a person who will be the entity that generates the answer content in the large-scale language model 200. The task is a task that indicates the generation of the answer content. The condition is a condition for generating the answer content. The condition may include, for example, at least one of information related to the defect, information related to the reporter, information related to the credibility, information related to the severity, information related to the answer policy, example sentences, etc.

[0046] The example sentences may be sentences stored in the knowledge database 300. For example, a plurality of example sentences associated with information about the defect and a response policy may be stored in the knowledge database 300. The prompt generation unit 104 may then acquire the plurality of example sentences corresponding to the information about the defect and the response policy from the knowledge database 300, and use the acquired plurality of example sentences as example sentences for the prompt conditions.

[0047] The LLM access unit 105 sends the generated prompt to the large-scale language model 200 and obtains the answer content generated by the large-scale language model 200 .

[0048] Next, a description will be given of the operation of the prompt generation device 100 configured as above. FIG.

[0049] As shown in FIG. 7, the receiving unit 101 receives information about defects in a town from the user terminal 400 (S101).

[0050] Next, the determination unit 102 determines at least one of information regarding credibility and information regarding severity based on the information regarding the defect received by the reception unit 101 (S102). In the present embodiment, as an example, the determination unit 102 determines both information regarding credibility and information regarding severity.

[0051] Next, the determination unit 103 determines information about a response policy based on the information about the defect and at least one of the information about the credibility and the information about the severity determined by the determination unit 102 (step S103). In this embodiment, the determination unit 103 determines information about a response policy based on both the information about the credibility and the information about the severity, as an example.

[0052] Next, the prompt generating unit 104 generates a prompt for instructing the generation AI to generate answer content based on the information about the answer policy determined by the determining unit 103 (step S104).

[0053] Next, the LLM access unit 105 transmits the prompt generated by the prompt generation unit 104 to the large-scale language model 200 and obtains the answer content generated by the large-scale language model 200 (step S105).

[0054] Next, the receiving unit 101 transmits the answer content acquired by the LLM access unit 105 to the user terminal 400 (step S106).

[0055] Here, an example of a prompt generated by the prompt generation device 100 will be described. Fig. 8 is a diagram showing an example of a prompt. The prompt shown in Fig. 8 includes a role, a task, and a condition.

[0056] The role of this prompt is written as "You are a government telephone operator." The role field contains predetermined, standardized information. This prompt requires the large-scale language model 200 to determine that the person is a government telephone operator.

[0057] The task field of this prompt reads, "Please create a response to the message from the resident." The task field contains predetermined, standardized information. This prompt requires the user to create a response to a resident who has reported information about a problem in the town from the user terminal 400.

[0058] The condition column of this prompt lists five conditions for the large-scale language model 200 to generate an answer sentence.

[0059] The first condition written in this condition column is "The reply should fully address the contents of the report from the resident" and "Report from a resident: Today, I was walking down XX street and noticed a hole in the road. It's dangerous if a child falls in, so could you do something about it?" This condition indicates that the message contained in the received information about a problem in the town and the reply to be generated should fully address the contents of that message.

[0060] The second condition listed in the condition column is "Please respond to the following resident," "Resident: AA, declaration credibility 24 points out of 100," and "Past response history: None." This condition indicates the information, credibility, and past response history of the declarant that will be taken into consideration when generating the response text.

[0061] The third condition listed in the conditions column is "Please write your answer taking into account the following severity level" and "Severity level: 75 points out of 100 points." This condition indicates the severity level of the bug that will be taken into consideration when generating the answer.

[0062] The fourth condition written in the condition column is "The policy for the message to be created should be to convey that the situation is being checked, that the customer should wait for a response, and that a response will be made promptly." This condition indicates a response policy. This response policy may be the same as the response policy determined by the determination unit 103, or may be an answer policy generated based on the response policy determined by the determination unit 103.

[0063] The fourth condition listed in the condition column is "You may output the following example sentence as a reference.", "Example sentence: Thank you for contacting us. We apologize for any inconvenience. We have received a report of a pothole in the town of △△. We will check the local situation and take action.", "Example sentence: Thank you for contacting us. We apologize for any inconvenience. We have received a report of a pothole in the town of △△. We will check the local situation immediately, so please wait a moment.", "Example sentence: Thank you for contacting us. We apologize for any inconvenience. "We have received a report of a pothole on ◇◇ Street. We will be checking the site on XX / XX, so please wait for a while." Another example sentence is "Thank you for contacting us. We apologize for any inconvenience. We have received a report of a pothole in front of XX Elementary School. We have received similar reports from other residents, and we plan to dispatch a representative to the site in an hour, so please wait. We will also be setting up barriers to prevent children from getting injured, but we would like the school to inform the school of this." This condition indicates an example sentence to be taken into consideration when generating an answer. There may be one or more example sentences. This prompt includes four example sentences.

[0064] The fifth condition written in the condition column is "Future response plan, scheduled for on-site check tomorrow evening. (Person in charge: BB)." This condition indicates a response plan for the defect to be included in the generated response document. This response plan may be the same as the response plan stored in the knowledge database 300, or may be a response plan generated based on the information stored in the knowledge database 300.

[0065] Next, the effects of the embodiments of the present disclosure will be described.

[0066] The prompt generation device 100 according to an embodiment of the present disclosure generates a prompt for instructing the large-scale language model 200 to generate answer content based on received information about a defect. This allows the large-scale language model 200 to be instructed to generate answer content for information about defects in a city. Furthermore, based on the received information about the defect, the prompt generation device 100 determines at least one of information about the credibility of the information about the defect and information about the severity of the information about the defect, determines information about an answer policy based on the information about the defect and at least one of the information about the credibility and the information about the severity, and generates a prompt for instructing the large-scale language model 200 to generate answer content based on the information about the answer policy. This allows the large-scale language model 200 to generate appropriate answer content.

[0067] Furthermore, since the information regarding the malfunction is information regarding a malfunction of infrastructure facilities in the town, it is possible to instruct the generation of answer content for the information regarding the malfunction of infrastructure facilities in the town.

[0068] Furthermore, the determination unit 102 determines the information regarding the credibility using at least either the information regarding the defect or the information regarding the person who reported the information regarding the defect, thereby improving the accuracy of determining the information regarding the credibility.

[0069] Furthermore, since the knowledge database 300 storing information about the informant is provided, by referring to the knowledge database 300, it is possible to easily determine information about the credibility based on the information about the informant.

[0070] Furthermore, the information about the defect includes at least either a photograph or information indicating a location, and the determination unit 102 determines the information about the credibility based on at least either the photograph or the information indicating the location, thereby improving the accuracy of determining the information about the credibility.

[0071] Furthermore, the determining unit 102 determines the information about the severity using at least information about the defect or information about the defect report history, thereby improving the accuracy of determining the information about the severity.

[0072] In addition, since the system is provided with a knowledge database 300 that stores information regarding the defect reporting history, by referring to the knowledge database 300, it is possible to easily determine information regarding the severity based on the information regarding the defect reporting history.

[0073] Furthermore, the information about the defect includes at least one of the details of the defect, information about the location, and a photograph, and the determination unit 102 determines the information about the severity based on at least one of the details of the defect, information about the location, and a photograph, thereby improving the accuracy of determining the information about the severity.

[0074] Furthermore, since the information regarding credibility is a score indicating the credibility and the information regarding seriousness is a score indicating the seriousness, it is possible to easily determine the information regarding credibility and the information regarding seriousness.

[0075] The device and method of the present disclosure include the following configuration.

[0076] [1] An apparatus comprising: a reception unit that receives information about defects in a city; a determination unit that determines at least one of information about the credibility of the defect and information about the seriousness of the defect based on the information about the defect; a determination unit that determines information about an answer policy based on the information about the defect and at least one of information about the credibility and information about the seriousness; and a generation unit that generates a prompt to instruct a generation AI to generate answer content based on the information about the answer policy.

[0077] [2] The device according to [1], wherein the information about the malfunction is information about a malfunction of infrastructure facilities in a city.

[0078] [3] The device according to [1] or [2], wherein the determination unit determines the information regarding the credibility using at least either information regarding the defect or information regarding a person who reports the information regarding the defect.

[0079] [4] The device according to [3], further comprising a knowledge database storing information about the informant.

[0080] [5] The device according to any one of [1] to [4], wherein the information about the defect includes at least either a photograph or information indicating a location, and the determination unit determines the information about the credibility based on at least either the photograph or the information indicating the location.

[0081] [6] The device according to any one of [1] to [5], wherein the determination unit determines the information regarding the severity using at least information regarding the malfunction or information regarding a reporting history of the malfunction.

[0082] [7] The device according to [6], further comprising a knowledge database that stores information about the defect reporting history.

[0083] [8] The device according to any one of [1] to [7], wherein the information about the defect includes at least one of the details of the defect, information about the location, or a photograph, and the determination unit determines the information about the severity based on at least one of the details of the defect, information about the location, or the photograph.

[0084] [9] The device according to any one of [1] to [8], wherein the information regarding the credibility is a score indicating the credibility, and the information regarding the seriousness is a score indicating the seriousness.

[0085]

[10] A method comprising: a step of accepting information about defects in a city; a step of determining, based on the information about the defect, at least one of information about the credibility of the information about the defect and information about the seriousness of the information about the defect; a step of determining information about an answer policy based on the information about the defect and at least one of information about the credibility and information about the seriousness; and a step of generating, based on the information about the answer policy, a prompt to instruct a generation AI to generate answer content.

[0086] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for 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 connected directly or indirectly (e.g., via wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0087] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0088] For example, the prompt generation device 100 according to an embodiment of the present disclosure may function as a computer that performs processing of the prompt generation method of the present disclosure. Fig. 9 is a diagram illustrating an example of the hardware configuration of the prompt generation device 100 according to an embodiment of the present disclosure. The prompt generation device 100 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage device 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0089] In the following description, the term "device" may be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the prompt generation device 100 may be configured to include one or more of the devices shown in the figures, or may be configured to exclude some of the devices.

[0090] Each function in the prompt generating device 100 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.

[0091] The processor 1001, for example, runs an operating system to control the entire computer. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control unit, an arithmetic unit, a register, etc. For example, the above-mentioned determination unit 102, decision unit 103, and prompt generation unit 104 may be realized by the processor 1001.

[0092] The processor 1001 also loads 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 these 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, the determination unit 102, the decision unit 103, and the prompt generation unit 104 may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be used 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.

[0093] 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 referred to as a register, a cache, a main memory (primary storage device), etc. The memory 1002 may store executable programs (program codes), software modules, etc. for implementing a prompt generation method according to one embodiment of the present disclosure.

[0094] 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.

[0095] 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 referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned reception unit 101 may be realized by the communication device 1004. The communication device 1004 may be implemented with a transmission unit and a reception unit that are physically or logically separated.

[0096] 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. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0097] 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.

[0098] Furthermore, prompt generation device 100 may 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 such hardware. For example, processor 1001 may be implemented using at least one of these pieces of hardware.

[0099] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0100] 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.

[0101] 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 sent to another device.

[0102] 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).

[0103] 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).

[0104] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0105] 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.

[0106] 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.

[0107] 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.

[0108] Note that terms described in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0109] Furthermore, the information, parameters, etc. described in the present 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. For example, a radio resource may be indicated by an index.

[0110] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0111] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," and the like may be used interchangeably.

[0112] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0113] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0114] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0115] 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."

[0116] 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.

[0117] 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.

[0118] 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.

[0119] 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."

[0120] 100...prompt generation device, 200...large-scale language model, 300...knowledge database, 400...user terminal, 101...reception unit, 102...determination unit, 103...decision unit, 104...prompt generation unit, 105...LLM access unit.

Claims

1. An apparatus comprising: a reception unit that receives information regarding defects in a city; a determination unit that determines at least one of information regarding the credibility of the defect and information regarding the seriousness of the defect based on the information regarding the defect; a determination unit that determines information regarding an answer policy based on the information regarding the defect and at least one of information regarding the credibility and information regarding the seriousness; and a generation unit that generates a prompt to instruct a generation AI to generate answer content based on the information regarding the answer policy.

2. The device according to claim 1, wherein the information relating to the malfunction is information relating to malfunctions in infrastructure facilities of a city.

3. The device according to claim 1, wherein the determination unit determines the information regarding the credibility using at least either the information regarding the defect or information regarding the person who reported the information regarding the defect.

4. The device according to claim 3, further comprising a knowledge database storing information about the filer.

5. The device according to claim 1, wherein the information about the defect includes at least either a photograph or information indicating a location, and the determination unit determines the information about the credibility based on at least either the photograph or the information indicating the location.

6. The device according to claim 1, wherein the determination unit determines the information regarding the severity using at least either information regarding the defect or information regarding a reporting history of the defect.

7. The device according to claim 6, further comprising a knowledge database storing information relating to the defect reporting history.

8. The device described in claim 1, wherein the information about the defect includes at least one of the content of the defect, information about the location, or a photograph, and the determination unit determines the information about the severity based on at least one of the content of the defect, information about the location, or the photograph.

9. The device according to claim 1, wherein the information regarding credibility is a score indicating credibility, and the information regarding seriousness is a score indicating seriousness.

10. A method comprising the steps of: accepting information about defects in a city; determining at least one of information regarding the credibility of the information about the defect and information regarding the seriousness of the information about the defect based on the information about the defect; determining information regarding an answer policy based on the information about the defect and at least one of information regarding the credibility and information regarding the seriousness; and generating a prompt to instruct a generation AI to generate answer content based on the information about the answer policy.

Images