Disaster prevention support device, disaster prevention support method, and disaster prevention support program

Abstract

This invention provides a disaster prevention support device, a disaster prevention support method, and a disaster prevention support program that can support disaster response by using AI to convert large amounts of information into text in the event of an earthquake or other disaster, thereby enabling rapid decision-making. [Solution] The disaster prevention support device comprises: a first acquisition unit that, when a receiver receives information transmitted from a caller during a disaster, converts the transmitted information into text using AI to generate and acquire the text; an analysis unit that analyzes the text by AI; a registration unit that divides the text into predetermined items and registers them; an information extraction unit that extracts missing information from the transmitted information using AI that has learned past call content; and a suggestion unit that suggests additional questions to the caller to the receiver based on the missing information.

Description

Technical Field

[0005]

[0001] The present invention relates to a disaster prevention support device, a disaster prevention support method, and a disaster prevention support program. When a disaster such as an earthquake occurs, a large amount of information is texturized by AI, and by making a decision quickly, it is possible to provide disaster response support. The present invention relates to a disaster prevention support device, a disaster prevention support method, and a disaster prevention support program.

Background Art

[0002] Conventionally, when a disaster such as an earthquake occurs, the report from the informant is described in natural language, so it contains ambiguity, and the disaster situation also changes complexly during a disaster. Therefore, when predicting response actions with a conventional disaster prevention support system, it is necessary to consider the ambiguity of the report and the diversity of the situation. However, conventionally, although a series of processes and concepts for predicting response actions from the report are disclosed, there is no specific disclosure on what actions should actually be taken when an emergency occurs.

[0003] Therefore, a response action prediction system and a response action prediction method for confirming what actions should actually be taken for disaster prevention during an emergency such as a disaster or during training assuming an emergency have been proposed (see Patent Document 1). In Patent Document 1, a natural language analysis unit analyzes a telegram in natural language and divides the character string into morphemes.

[0004] In addition, a dispatch team formation system has been proposed that receives a request for dispatch of a fire department or an emergency medical team via a public communication line, forms the corresponding dispatch team, and issues a dispatch order (see Patent Document 2). In Patent Document 2, a map search device searches for surrounding map data of the incident location based on the received data input for the first time and displays and outputs it to a map search display device.

[0005] However, there are problems such as the resources of the disaster response headquarters being diverted to handling the large volume of calls from callers, and inconsistencies in the skills of callers due to internal transfers within the disaster response headquarters and support from other departments. In addition, there are problems such as difficulty in making quick decisions due to uncertainty about the reliability of the collected information, and difficulty in sharing analysis results and the latest information among staff. Therefore, it was desired that in the event of an earthquake or other disaster, AI could be used to transcribe large amounts of information into text, enabling rapid decision-making and thus supporting disaster response. [Prior art documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Patent Publication No. 2010-271857 [Patent Document 2] Patent No. 4576757 [Overview of the project] [Problems that the invention aims to solve]

[0007] Therefore, the present invention has been made in view of the above points, and provides a disaster prevention support device, a disaster prevention support method, and a disaster prevention support program that can support disaster response by converting a large amount of information into text using AI when a disaster such as an earthquake occurs and making rapid decisions. [Means for solving the problem]

[0008] In other words, the disaster prevention support device according to the first embodiment is characterized by comprising: a first acquisition unit that, when a receiver receives information transmitted from a caller during a disaster, converts the transmitted information into text using AI to generate and acquire the text; an analysis unit that analyzes the text by AI; a registration unit that divides the text by predetermined items and registers them; an information extraction unit that extracts missing information from the transmitted information using AI that has learned past call content; and a proposal unit that proposes additional questions to the caller to the receiver based on the missing information.

[0009] In a second embodiment, the disaster prevention support device according to the first embodiment may further include a second acquisition unit that uses AI to convert the content of the informant's answers to additional questions proposed by the proposal unit into text, thereby generating and acquiring the text information.

[0010] A third embodiment is a disaster prevention support device relating to the first embodiment, in which the transmitted information is information that the caller has handwritten and entered into the caller's tablet terminal.

[0011] A fourth embodiment may be a disaster prevention support device according to the first embodiment, wherein the transmitted information includes location information, and further comprises a plotting unit that plots the location information of the caller on a map.

[0012] A fifth embodiment may further include a disaster prevention support device according to the first embodiment, which includes a priority determination unit that determines the priority of the information transmitted by the caller according to the severity of the disaster, and a priority display unit that displays the priority determined by the priority determination unit together with a map.

[0013] A sixth embodiment may further include an investigation instruction unit that, based on location information, uses AI to collect device information from at least one of the following: a drone, a live camera, a mobile terminal application, and a vehicle's driving record, and generates candidate instructions for investigating the disaster site of the person making the report.

[0014] The seventh aspect of disaster prevention support method includes an acquisition step in which a computer, upon receiving information from a caller during a disaster, converts the information into text using AI to generate and acquire the text; an analysis step in which the text is analyzed by AI; a registration step in which the text is divided into predetermined items and registered; an information extraction step in which the AI, having learned past call content, extracts missing information from the information; and a suggestion step in which the AI ​​suggests additional questions to the caller to the receiver based on the missing information. It is characterized by causing the execution of [this action].

[0015] The disaster prevention support program according to the eighth aspect includes, on a computer, an acquisition function that generates and acquires text information by converting outgoing information from a disaster caller into text using AI when the receiver receives the outgoing information; an analysis function that analyzes the text information using AI; a registration function that divides the text information into predetermined items and registers them; and an information extraction step that extracts missing information from the outgoing information using AI that has learned past call content from the text information. The system is characterized by providing a suggestion function that proposes additional questions to the recipient based on missing information. [Effects of the Invention]

[0016] The disaster prevention support device according to the present invention is characterized by comprising: a first acquisition unit that generates and acquires text information by converting outgoing information from a caller into text using AI when the receiver receives outgoing information from a caller during a disaster; an analysis unit that analyzes the text information using AI; a registration unit that divides the text information into predetermined items and registers them; an information extraction unit that extracts missing information from the outgoing information using AI that has learned past call content from the text information; and a proposal unit that proposes additional questions to the caller to the receiver based on the missing information. Therefore, in the event of a disaster such as an earthquake, a large amount of information can be converted into text using AI, enabling rapid decision-making and thus supporting disaster response.

[0017] Furthermore, the disaster prevention support method and disaster prevention support program according to the present invention, like the disaster prevention support device according to the present invention, can support disaster response by converting a large amount of information into text using AI when a disaster such as an earthquake occurs, and by making rapid decisions. [Brief explanation of the drawing]

[0018] [Figure 1] This is a schematic diagram showing an example of the configuration of the disaster prevention support system according to the embodiment. [Figure 2] This is a block diagram showing an example of the functional configuration of a disaster prevention support device according to an embodiment. [Figure 3] It is a diagram showing the processing flow of the disaster prevention support device of the embodiment. [Figure 4] It is a diagram showing the AI call recording function of the disaster prevention support device of the embodiment. [Figure 5] It is a diagram showing the AI item analysis and registration function of the disaster prevention support device of the embodiment. [Figure 6] It is a diagram showing the AI question suggestion function of the disaster prevention support device of the embodiment. [Figure 7] It is a diagram showing the text conversion and additional registration function of the shortage information of the disaster prevention support device of the embodiment. [Figure 8] It is a diagram showing the state where the disaster information is plotted on the map by the plot unit of the disaster prevention support device of the embodiment. [Figure 9] It is an example of a flowchart of the disaster prevention support method of the embodiment.

Embodiments for Carrying Out the Invention

[0019] Referring to FIG. 1, an example of the configuration of the disaster prevention support system 1 according to the embodiment will be described. FIG. 1 is a schematic diagram showing an example of the configuration of the disaster prevention support system 1.

[0020] The disaster prevention support system 1 is a system that collects, analyzes, and shares information in cooperation between citizens and the disaster countermeasure headquarters 2, and the collection, analysis, and sharing of these information are performed by the disaster prevention support device 100.

[0021] As the first means of collection, there are the telephone from the general public 3 and the lifeline information 4. When a disaster such as an earthquake occurs, the general public 3 who is the informant calls the disaster countermeasure headquarters 2 and contacts the disaster countermeasure headquarters 2 about the situation of the disaster. The contact from the general public 3 may be not only voice by telephone but also an image or video. In addition, the Ministry of Land, Infrastructure, Transport and Tourism provides the disaster countermeasure headquarters 2 with the lifeline (electricity, water, gas, communication, etc.) information 4 of the area where the disaster has occurred. The disaster countermeasure headquarters 2 can aggregate the lifeline information 4.

[0022] As a second means of collection, registered citizens (registered collaborators) 5 contact the disaster response headquarters 2 with their location information. The disaster response headquarters 2 analyzes this information. The reception staff 6 at the disaster response headquarters 2 receive, register, and organize contact information 4 from the general public 3 and lifeline information. The person in charge 7 at the disaster response headquarters 2 registers instructions to the staff member 8. Staff member 8 confirms the instructions and registers the results. The person in charge 7 confirms the results. The field responders 9 register reliable information from the field obtained from drones 11, dashcams 12, JIII (mobile terminal application) 13, live cameras 14, far-infrared cameras 15, etc. The disaster response headquarters 2 shares this analyzed information. The disaster response headquarters 2 transmits this information to relevant parties 16, who then refer to this information and transmit it to external parties (citizens, media) 17.

[0023] The functional configuration of the disaster prevention support device 100 will be explained with reference to Figure 2. Figure 2 is a block diagram showing an example of the functional configuration of the disaster prevention support device 100. The disaster prevention support device 100 includes functional units such as a first acquisition unit 10, an analysis unit 20, a registration unit 30, an information extraction unit 40, a proposal unit 50, a second acquisition unit 60, a plotting unit 70, a priority determination unit 80, a priority display unit 90, and a survey instruction unit 110. These functional units are realized by the CPU executing an information processing program deployed on RAM. The disaster prevention support device 100 also appropriately implements storage units such as ROM, RAM, SSD, HDD, and various interfaces (not shown).

[0024] The first acquisition unit 10 acquires the transmitted information from a disaster caller by converting the transmitted information into text using AI when the receiver receives the information. Text conversion refers to converting voice, images, videos, etc., from phone calls into text. Text conversion can be performed using natural language processing. Natural language processing is a technology that allows computers to process natural language, which humans use on a daily basis. Outgoing information includes both calls and messages. Since calls are voice-based and difficult to organize, converting them to text using AI makes registration and organization easier. Outgoing information may also include information handwritten by the caller on their tablet device. Location information may also be included in the outgoing information.

[0025] The analysis unit 20 analyzes the text information using AI. Analysis involves classifying the large amount of information sent from the general public 3, such as what kind of disaster occurred in which region and who was seriously injured, into predetermined categories, while removing duplicates. These predetermined categories, as will be detailed later in Figure 5, include, for example, Category 1, Category 2, Name, Location, Content, Assigned Department, and Assignment Details.

[0026] The registration unit 30 registers the text information, dividing it into predetermined items, so that it can be included in the disaster information report 21 shown in Figure 5. The predetermined items are as described above. Registration means inputting the text information so that it can be displayed as a form in the disaster information report 21.

[0027] The information extraction unit 40 uses an AI that has learned from past phone calls to extract missing information from the transcribed text. Missing information refers to information that is insufficient for the predetermined items mentioned above. For example, if the AI ​​has learned from past phone calls that included the street number of an address in a disaster-stricken area, it will extract the detailed street number as missing information regarding the location.

[0028] The suggestion unit 50 suggests to the recipient additional questions for the informant based on the missing information. These additional questions may include, for example, as shown in Figure 6, "Let's confirm the address details," "Let's confirm that your safety is ensured," and "Let's confirm if there are any other people who need help, such as those with mobility difficulties or the elderly."

[0029] The second acquisition unit 60 uses AI to convert the informant's answers to additional questions proposed by the proposal unit 50 into text, generating and acquiring the text information.

[0030] Based on the information acquired by the first acquisition unit 10 and the second acquisition unit 60, the plotting unit 70 plots the location information of the caller on a map, as shown in Figure 8.

[0031] The priority determination unit 80 determines the priority of the information transmitted by the caller according to the severity of the disaster. The priority according to the severity of the disaster is expressed as follows, for example, as shown in Figure 8, with a priority of "high" if there are seriously injured patients, and a priority of "medium" if vehicles are submerged.

[0032] The priority display unit 90 displays the priority determined by the priority determination unit 80 along with the map. Furthermore, as detailed in Figure 8, the priority display unit 90 can display the priority of each damage situation as the damage situation improves moment by moment, next to the map.

[0033] The investigation instruction unit 110 uses AI to collect device information from at least one of the following devices based on location information: a drone 11, a vehicle's driving record (dashcam) 12, a mobile terminal application 13, a live camera 14, and a far-infrared camera 15, and generates candidate instructions for investigating the disaster site of the person making the report. Collecting device information means collecting information such as the current location, range of motion, and charging status of each device. Based on this information, the investigation instruction unit 110 generates candidate instructions for the drone 11 to investigate the disaster site by taking images of the disaster area. The investigation instruction unit 110 also generates candidate instructions for the dashcam 12 to investigate the status of vehicles driving in the disaster area. The investigation instruction unit 110 also generates candidate instructions for the mobile terminal application 13 to input the damage situation in the disaster area. The investigation instruction unit 110 also generates candidate instructions for the live camera 14 to investigate the disaster site by taking images of rivers, etc., in the disaster area. The investigation instruction unit 110 also generates candidate instructions for the far-infrared camera 15 to investigate the disaster site by taking nighttime images of the disaster area.

[0034] Refer to Figure 3 to explain the processing flow of the disaster prevention support system 1. Figure 3 is a diagram illustrating the processing flow of the disaster prevention support system 1. First, when a call comes in from a member of the public 3, the receptionist 6 who receives the call listens to the content and instructs the disaster prevention support device 100. In the disaster prevention support device 100, the first acquisition unit 10 uses AI to transcribe the outgoing information into text and generates and acquires the transcribed information. Next, the analysis unit 20 uses AI to analyze the transcribed information, and the registration unit 30 registers the transcribed information, dividing it into predetermined items. Next, the information extraction unit 40 uses AI, which has learned from past call content, to extract any missing information from the outgoing information. Next, the proposal unit 50 proposes additional questions for member of the public 3 to the receptionist 6 based on the missing information. The receptionist 6 calls member of the public 3 and asks the additional questions based on the missing information. When member of the public 3 answers the additional questions, the second acquisition unit 60 uses AI to transcribe the caller's answers to the additional questions proposed by the proposal unit 50 and generates and acquires the transcribed information. Next, the analysis unit 20 analyzes the text information using AI, and the registration unit 30 registers the text information, dividing it into predetermined categories. This process is repeated.

[0035] Referring to Figure 4, the processing content of the first acquisition unit 10 will be explained. Figure 4 is a diagram showing the function of the AI ​​call recording of the disaster prevention support device of the embodiment. Assume that a call comes in from a regular citizen 3, and as shown in Figure 4, the call content (voice) is either, "This is XX from XX town. My house is flooded above floor level. I want to evacuate, but where should I go?" or, as written, "XX town XX flooded." In the former case, the parts "XX town," "XX," "My house is flooded above floor level," and "I want to evacuate, but where should I go?" are selected and converted to text. In the latter case, it is converted to text as "XX town XX flooded."

[0036] Referring to Figure 5, the processing contents of the analysis unit 20 and the registration unit 30 will be explained. Figure 5 is a diagram showing the AI ​​item analysis and registration function of the disaster prevention support device according to the embodiment.

[0037] The analysis unit 20 analyzes information transmitted from ordinary citizens 3. Specifically, it analyzes whether the information falls under category 1 (22) and whether it concerns safety, damage, fire, rescue, supplies, lifelines, or other. It also analyzes whether the information falls under category 2 (23) and whether it is a report, request, confirmation, information provision, inquiry, or other. The analysis unit 20 also analyzes the assigned departments 27 and assignment details 28 of the disaster response headquarters 2 from the text information converted into text by the first acquisition unit 10.

[0038] The registration unit 30 registers information in the disaster information report 21 based on the analysis results performed by the analysis unit 20. If the analysis unit 20 analyzes the case as "damage" under category 1 (22), the "damage" box is checked. If the analysis unit 20 analyzes the case as "confirmation" under category 2 (23), the "confirmation" box is checked. In addition, based on the text information converted into text by the first acquisition unit 10, the text information is entered into the name (24), location (25), content (26), assigned department (27), and assigned content (28) fields. As shown in Figure 5, the location (25) field only contains "XX town" and does not contain an address, so there is insufficient information. Also, nothing is entered in the assigned content (28) field, so there is insufficient information.

[0039] Referring to Figure 6, the processing contents of the information extraction unit 40 and the suggestion unit 50 will be explained. Figure 6 is a diagram showing the AI ​​question suggestion function of the disaster prevention support device of the embodiment. As mentioned above, only "XX town" is entered in the location (25) field, so the street address is not entered and there is insufficient information. Also, nothing is entered in the assignment content (28) field, so there is insufficient information. Therefore, the information extraction unit 40 extracts the details of the address (25) and the assignment content (28) as missing information. The suggestion unit 50 generates instructions for creating additional questions based on the missing information extracted by the information extraction unit 40. For example, as shown in Figure 6, it generates an instruction 1 (51) that says, "Let's check the details of the address." It also generates an instruction 2 (52) that says, "Let's check if your safety is ensured." It also generates an instruction 3 (53) that says, "Let's check if there are any other people who need help, such as people with mobility difficulties or the elderly."

[0040] Referring to Figure 7, the processing content of the second acquisition unit 60 will be explained. Figure 7 is a diagram showing the AI ​​call function and AI item analysis / registration function of the disaster prevention support device of the embodiment. The second acquisition unit 60 receives responses from the general public 3 to the additional questions generated by the proposal unit 50, i.e., instructions 1(51) through 3(53) above, and reflects them in the disaster information report 21. First, in response to instruction 1(51), "Let's confirm the details of the address," when it receives the response "XX town 1-1-1" from the general public 3, it obtains the response as text. Next, in response to instruction 2(52), "Let's confirm whether your safety is ensured," when it receives the response "My safety is ensured," it obtains the response as text. Finally, in response to instruction 3(53), "Let's confirm if there are any other people who need help, such as those with mobility difficulties or the elderly," when it receives the response "There is a wheelchair user" from the general public 3, it obtains the response as text. The analysis unit 20 analyzes the text information, and the registration unit 30 registers the text information. As a result, as shown in Figure 7, additional information is added, with "1-1-1 XX Town" in the Location (25) column and "Safety is ensured. There is a wheelchair user." in the Assignment Details (28) column.

[0041] Referring to Figure 8, the processing contents of the plotting unit 70, the priority determination unit 80, and the priority display unit 90 will be explained. Figure 8 is a diagram showing the state in which disaster information is plotted on a map by the plotting unit 70 of the disaster prevention support device of the embodiment.

[0042] The plotting unit 70 plots the location information of the informants on the map. Informants refer to both ordinary citizens 3 and citizens (registered collaborators) 5. When the first acquisition unit 10 acquires location information from these informants, the plotting unit 70 plots the location of the informants on the map as shown in Figure 8. Furthermore, on the right side of the map, information such as "20xx / xx / xx 8:45 Power outage in 20 houses in Sangenjaya 1-chome", "20xx / xx / xx 8:42 Man in his 50s seriously injured by a sign that fell due to strong winds", "20xx / xx / xx 8:40 Vehicle submerged in underbus", "20xx / xx / xx 8:38 Street tree falls onto the road, obstructing traffic", and "20xx / xx / xx 8:35 Fallen sign blocks road, obstructing traffic" is displayed.

[0043] The priority determination unit 80 determines the priority of this information, and the priority display unit 90 displays the priority. For example, clicking on the section that reads "20xx / xx / xx 8:42 A man in his 50s is seriously injured when a sign falls due to strong winds" will display the priority as "Medium," and clicking on the section that reads "20xx / xx / xx 8:40 A vehicle is submerged in an underbus" will display the priority as "High."

[0044] Referring to Figure 9, the disaster prevention support program according to the embodiment will be described along with the disaster prevention support method. Figure 9 is an example of a flowchart of the disaster prevention support method according to the embodiment. The disaster relief method is executed by the CPU of the disaster relief device 100 based on the disaster relief program. The disaster prevention support program includes steps such as acquisition step S10, analysis step S20, registration step S30, information extraction step S40, and proposal step S50. The disaster prevention support program enables the CPU of the disaster prevention support device 100 to perform acquisition, analysis, registration, information extraction, and proposal functions. These functions are executed in the order shown in the flowchart of Figure 9, but the order can be changed as needed. Since each function overlaps with the explanation of the various functions of the disaster prevention support system 1 described above, a detailed explanation is omitted.

[0045] The acquisition function generates and acquires text information by converting the transmitted information from a disaster caller into text using AI when the recipient receives the transmitted information (Step S10: Acquisition Step).

[0046] The analysis function uses AI to analyze the text information (Step S20: Analysis Step).

[0047] The registration function registers the text information by dividing it into predetermined fields (Step S30: Registration Step).

[0048] The information extraction function uses an AI that has learned from past call content to extract missing information from the transcribed text (Step S40: Information Extraction Step).

[0049] The suggestion function suggests to the recipient additional questions to the informant based on the missing information (Step S50: Suggestion Step).

[0050] According to each aspect of this disclosure described above, when a disaster such as an earthquake occurs, it is possible to support disaster response by converting a large amount of information into text using AI and making rapid decisions.

[0051] The positioning program of this embodiment can be implemented on the CPU of the disaster prevention support device 100 using a computer language such as the Arduino® IDE.

[0052] [Regarding functions and circuitry] Next, the functions and circuitry of the CPU of the disaster prevention support device 100 described above will be explained. Each functional part of the CPU of the disaster prevention support device 100 may be implemented as a function of a computer's processing unit or the like. That is, the CPU of the disaster prevention support device 100 may be implemented as an acquisition function, analysis function, registration function, information extraction function, and proposal function, respectively, by a computer's processing unit or the like. The disaster relief program can implement the above-described functions on a computer. The disaster relief program may be recorded on a computer-readable, non-temporary storage medium such as memory, a solid-state drive, a hard disk drive, or an optical disc. The storage medium can be rephrased as a non-temporary, computer-readable medium for storing the disaster relief program. Furthermore, the disaster relief program may be transmitted online. Furthermore, the computer's arithmetic processing unit and the like described above may be composed of, for example, integrated circuits. That is, the CPU of the disaster prevention support device 100 may be implemented as an acquisition circuit, analysis circuit, registration circuit, information extraction circuit, and proposal circuit that constitute the computer's arithmetic processing unit and the like.

[0053] Furthermore, the present invention is not limited to the disaster prevention support device 100, disaster prevention support method, and disaster prevention support program according to the above-described embodiment, and can be implemented by various other modifications or applications without departing from the gist of the present invention as described in the claims. Also, although the word "information" is used in the above-described embodiment, the word "information" can be replaced with "data," and the word "data" can be replaced with "information."

[0054] [Aspects and Effects of This Embodiment] Next, an embodiment of this model and the effects of each embodiment will be described. Note that the embodiments described below are examples as of the time of filing, and this embodiment is not limited to the embodiments described below. In other words, this embodiment is not limited to the embodiments described below, and may be realized by appropriately combining the parts described above. Furthermore, lower-level embodiments may be referenced in any of the higher-level embodiments. Furthermore, the effects of this embodiment described below are merely examples, and the effects achieved by each embodiment are not limited to those described below. Also, each embodiment may achieve, for example, at least one of the effects described below.

[0055] (Aspect 1) One embodiment of a disaster prevention support device includes: a first acquisition unit that, when a receiver receives information transmitted from a caller during a disaster, converts the transmitted information into text using AI to generate and acquire the text; an analysis unit that analyzes the text by AI; a registration unit that divides the text into predetermined items and registers them; an information extraction unit that extracts missing information from the transmitted information using AI that has learned past call content; and a suggestion unit that suggests additional questions to the caller to the receiver based on the missing information. This allows the disaster response support device to use AI to convert large amounts of information into text when an earthquake or other disaster occurs, enabling rapid decision-making and thus supporting disaster response.

[0056] (Aspect 2) In one embodiment of the disaster prevention support device, a second acquisition unit may be further provided, which uses AI to convert the content of the caller's responses to additional questions proposed by the proposal unit into text, thereby generating and acquiring the text information. This allows for the use of AI to transcribe large amounts of information into text in the event of an earthquake or other disaster, enabling rapid decision-making and thus supporting disaster response.

[0057] (Aspect 3) In one embodiment of the disaster prevention support device, the transmitted information may be information that the caller has handwritten and entered into their tablet device. This allows for the use of AI to transcribe large amounts of information into text in the event of an earthquake or other disaster, enabling rapid decision-making and thus supporting disaster response.

[0058] (Aspect 4) In one embodiment of the disaster prevention support device, the transmitted information may include location information, and the device may further include a plotting unit that plots the location information of the caller on a map. This allows for the use of AI to transcribe large amounts of information into text in the event of an earthquake or other disaster, enabling rapid decision-making and thus supporting disaster response.

[0059] (Appendix 5) One embodiment of the disaster prevention support device may further include a priority determination unit that determines the priority of the information transmitted by the caller according to the severity of the disaster, and a priority display unit that displays the priority determined by the priority determination unit along with a map. This allows for the use of AI to transcribe large amounts of information into text in the event of an earthquake or other disaster, enabling rapid decision-making and thus supporting disaster response.

[0060] (Aspect 6) In one embodiment of the disaster prevention support device, an investigation instruction unit may further include an AI that, based on location information, collects device information from at least one of the following: a drone, a live camera, a mobile terminal application, and a vehicle's driving record, and generates candidate instructions for investigating the disaster site of the person making the report. This allows for the use of AI to transcribe large amounts of information into text in the event of an earthquake or other disaster, enabling rapid decision-making and thus supporting disaster response.

[0061] (Aspect 7) The disaster relief support method involves having a computer perform the following steps: an acquisition step in which, upon receiving information from a caller during a disaster, the caller's information is converted into text using AI to generate and acquire the text; an analysis step in which the text is analyzed using AI; a registration step in which the text is divided into predetermined categories and registered; an information extraction step in which the AI, having learned from past call content, extracts missing information from the caller's information; and a suggestion step in which the computer suggests additional questions to the caller based on the missing information. This allows the disaster response support device to use AI to convert large amounts of information into text when an earthquake or other disaster occurs, enabling rapid decision-making and thus supporting disaster response.

[0062] (Pattern 8) The disaster relief support program implements the following functions on a computer: an acquisition function that generates and acquires text information by converting the outgoing message from the caller into text using AI when the receiver receives the message during a disaster; an analysis function that analyzes the text information using AI; a registration function that divides the text information into predetermined categories and registers it; an information extraction step that extracts missing information from the outgoing message using AI that has learned from past call content; and a suggestion function that suggests additional questions to the caller to the receiver based on the missing information. This allows the disaster response support device to use AI to convert large amounts of information into text when an earthquake or other disaster occurs, enabling rapid decision-making and thus supporting disaster response. [Explanation of Symbols]

[0063] 1. Disaster Prevention Support System 2. Disaster Countermeasures Headquarters 3 General public 4. Lifeline Information 5. Citizens (registered collaborators) 6. Receptionist 7. Person in Charge 8 Person in Charge 9. Environmental Response Officer 10 Acquisition Department 11 Drones 12 Dashcams 13 Mobile device applications 14 Live Cameras 15. Far-infrared camera 16 Related parties 17. External (Citizens, Media) 20 Analysis Department 21 Disaster Information Report 22 Category 1 23 Category 2 24 Name 25 locations 26 Contents 27 Assigned Departments 28. Assignment Details 30 Registration Department 40 Information extraction part 50 Proposal Department 60 Second acquisition part 70 Plot section 80 Priority judgment section 90 Priority display area 100 Disaster prevention support equipment 110 Investigation and Instruction Department

Claims

1. A first acquisition unit that, when a recipient receives information transmitted from a disaster caller, converts the transmitted information into text using AI to generate and acquire the text information, An analysis unit that analyzes the aforementioned text information using AI, A registration unit that registers the aforementioned text information by dividing it into predetermined items, An information extraction unit extracts missing information from the outgoing call information using an AI that has learned from past call content, based on the aforementioned text information. A proposal unit that proposes additional questions to the recipient based on the missing information, A disaster prevention support device equipped with the following features.

2. The second acquisition unit uses AI to convert the content of the informant's responses to the additional questions proposed by the proposal unit into text, and generates and acquires the text information. The disaster prevention support device according to claim 1, further comprising:

3. The disaster prevention support device according to claim 1, wherein the transmitted information is information that the caller has handwritten and entered into the caller's tablet device.

4. The disaster prevention support device according to claim 1, wherein the transmitted information includes location information, and further comprises a plotting unit for plotting the location information of the caller on a map.

5. A priority determination unit that determines the priority of the information transmitted by the aforementioned informant according to the severity of the disaster, The disaster prevention support device according to claim 1, further comprising a priority display unit that displays the priority determined by the priority determination unit together with the map.

6. The disaster prevention support device according to claim 4, further comprising an investigation instruction unit that, based on the location information, uses AI to collect device information from at least one of a drone, a live camera, a mobile terminal application, and a vehicle's driving record, and generates candidate instructions for investigating the disaster site of the person making the report.

7. On the computer, The acquisition step involves, when a recipient receives information transmitted from a disaster informant, using AI to convert the transmitted information into text and generate and acquire the text information; An analysis step in which the aforementioned text information is analyzed by AI, A registration step in which the aforementioned text information is divided into predetermined items and registered, The information extraction step involves using an AI that has learned past call content to extract missing information from the outgoing call information from the aforementioned text information. A suggestion step of proposing to the recipient additional questions to the informant based on the missing information, A disaster relief support method that enables implementation.

8. On the computer, A function to acquire information generated by converting information transmitted from a disaster caller into text using AI when the recipient receives the transmitted information, An analysis function that analyzes the aforementioned text information using AI, A registration function that divides the aforementioned text information into predetermined items and registers them, The information extraction step involves using an AI that has learned past call content to extract missing information from the outgoing call information from the aforementioned text information. A suggestion function that proposes additional questions to the recipient based on the missing information, A disaster relief program that makes this a reality.

Images