Disaster Safety Knowledge Integrated Management System by AI

Abstract

This invention relates to an AI-based disaster safety knowledge integrated management system that utilizes an intelligent analysis service of disaster safety data to enable an AI-based question-answering service of specialized knowledge in the field of disaster safety and an automatic reporting service that supports the creation of policy plans and report materials for specific subjects. It is composed of a disaster safety knowledge base part connected to a data network and an artificial intelligence part for realizing high-dimensional information processing using artificial intelligence. The disaster safety knowledge base unit is composed of a data collection unit for collecting and aggregating various information from external organizations; a data transmission unit for transmitting the aggregated information to a server via LTE, 5G, WIFI, etc.; and a big data unit for analyzing and storing the transmitted data. The artificial intelligence unit utilizes the data accumulated and analyzed in the big data unit to make judgments and inferences based on human cognitive abilities (language, voice, vision, sensitivity, etc.) and learning and inference functions, thereby featuring machine intelligence through rapid learning using data.

Description

Technical Field

[0001] The present invention relates to an AI-based integrated disaster safety knowledge management system that enables a question-and-answer service for specialized knowledge in the field of disaster safety by AI through the use of an intelligent analysis service for disaster safety data, and also enables an automatic reporting service to assist in the creation of policy planning and report materials for specific topics.

Background Art

[0002] The Fourth Industrial Revolution has brought about a dramatic improvement in productivity through the intelligentization of machines and a fundamental transformation of the industrial structure, which has been made possible by changes in intelligent information technology.

[0003] Intelligent information technology is a combination of "intelligence" that realizes high-dimensional information processing of humans by artificial intelligence and "information" based on data network technologies (IoT, cloud, big data, mobile). Artificial intelligence is a technology that realizes human cognitive abilities (language, voice, vision, sensibility, etc.) and intelligence such as learning and inference, including the software / hardware and basic technologies of artificial intelligence. Data network technologies have become essential ICT technologies for generating, collecting, transmitting, storing, and analyzing data, serving as an important foundation to support the rapid performance improvement, popularization, and diffusion of artificial intelligence technologies.

[0004] Intelligent information technology has developed to the level where machines can perform unmanned decision-making (machines exercise human high-dimensional judgment capabilities), real-time responses, autonomous evolution, and have learning capabilities that enable the dataization of all things, and has been applied to specialized fields.

[0005] Recently, in order to promote the joint use of data among administrative and public institutions, improve the integrated management system, and activate objective and scientific policy-making and decision-making, the state has emphasized the use of intelligent information technology and is in the process of formulating relevant laws and policies.

[0006] On the one hand, in order to efficiently respond to disasters such as new types of disasters and compound disasters and formulate scientific and systematic safety policies, database-based disaster safety management is urgently required. To establish a legal and institutional foundation for systematically managing data for database disaster management and aiming for private-sector sharing and opening, it is necessary to construct integrated disaster safety data.

[0007] "Integrated disaster safety data" is used to promote disaster safety policies based on databases, such as damage prediction and selection of safety-vulnerable areas. It consists of a process of collecting and linking data, a process of accumulating and storing it, and a process of opening and utilizing it for the target persons who need it.

[0008] In this situation where the promotion of data-centered government policies and related policies has been in full swing, it is urgently required to replace the decision-making method based on conventional experience and intuition with an objective and scientific administrative system method based on data, in order to enhance reliability and improve the quality of the people's lives (see Figure 9).

[0009] However, the necessary data is fragmented and difficult to utilize as information. Not only is experiential knowledge not digitized, but also an integrated management system has not been established, so database-based disaster safety management has not been properly utilized.

[0010] In addition, in the decision-making method based on conventional experience and intuition, intellectual activities such as policy planning and decision-making are carried out in a labor-intensive manner, resulting in high time and personnel costs. Furthermore, there are problems such as dependence on skilled experts and blind spots in analysis, and high-level intellectual labor is required from the data investigation and search necessary for policy reports to the creation of reports.

[0011] That is, there is a limit for beginners and inexperienced people to directly analyze, and there is a very high dependence on highly trained professional analysts, resulting in blind spots in analysis, so there is a problem that integrated analysis of all related fields is impossible.

Summary of the Invention

Problems to be Solved by the Invention

[0012] The first object of the present invention is to enable a question-and-answer service for specialized knowledge in the field of disaster safety by AI and an automatic reporting service for assisting in the creation of policy plans and reports on specific topics by utilizing disaster safety data sharing and intelligent analysis services for disaster safety data.

[0013] In addition, by dramatically reducing the time, personnel, costs, and workload involved in data exploration, analysis, and reporting work, enabling non-experts and those with relatively little experience to generate analysis and policy materials with the assistance of machines, and supporting comprehensive judgment considering information other than disasters, new technologies, and new issues, another object is to significantly improve the level of disaster safety policy support through insights based on an integrated knowledge base in the field of disaster safety.

Means for Solving the Problems

[0014] To achieve the above object, in the AI-based disaster safety knowledge integration management system of the present invention, by utilizing disaster safety data sharing and intelligent analysis services for disaster safety data, it is related to an AI-based disaster safety knowledge integration management system that enables a question-and-answer service for specialized knowledge in the field of disaster safety by AI and an automatic reporting service for assisting in the creation of policy plans and reports on specific topics.

[0015] The disaster safety knowledge integration management system is composed of a disaster safety knowledge base unit combined with a data network and an artificial intelligence unit for realizing human high-dimensional information processing by artificial intelligence.

[0016] The disaster safety knowledge base department includes a data collection department for collecting and aggregating various information from external institutions, a data transmission department for transmitting the collected information to a server via LTE, 5G, WIFI, etc., and a big data department for analyzing and storing the transmitted data.

[0017] In the data collection department, general materials such as disaster situations, disaster statistics, disaster policies, situation reports, research reports, unstructured data such as news, SNS, laws, situation reports, research reports, and structured data such as damage restoration, meteorology, safety indexes, disaster sirens, 119 reports are provided. Also, in the big data department, the data provided by the data transmission department is distinguished, analyzed, and stored, and classified into real-time data such as steep slope sensors, water level information, photos of the disaster site, structured data such as past damage history information and safety information of various facilities, unstructured data such as the text of disaster situation reports, images of disaster situation reports, and briefing materials. The data classified by these is classified into the damage situation of disaster accidents, the response history of disaster accidents, and disaster safety policy information, stored as archive data, and can be opened and shared with necessary target persons (ordinary citizens, relevant institutions, etc.). Also, in the artificial intelligence department, by utilizing the data stored and analyzed in the big data department and performing judgment and inference by human cognitive abilities (language, voice, vision, sensibility, etc.) and learning and inference functions, the intelligentization of machines through rapid learning using data is characteristic.

[0018] In the above AI department, from the state of data which is objective facts, relevant and correlative relationships are understood and informatized through meaning extraction and pattern recognition. Then, the result information internalized as unique knowledge is structured and knowledge-ized, and finally, wisdom which is a creative result is generated by structuring knowledge.

[0019] In the disaster safety knowledge-based data collection department, overseas information, unstructured data, and structured data are collected, and after going through knowledge resource collection / management, human-mimicking knowledge learning, and natural language understanding knowledge learning, they are accumulated in the big data department of the disaster safety knowledge base. In addition to natural language understanding knowledge learning, knowledge curation and knowledge enhancement through composite inference may also be provided to the big data department.

[0020] In the AI department, the data accumulated in the big data department is utilized to perform judgment and inference through human cognitive abilities (such as language, voice, vision, and sentiment) and learning and inference functions. It conducts problem analysis through query interpretation and situation interpretation from queries and keywords provided externally, a process of understanding the user's intention, a process of searching / inferring candidate answers, making judgments / predictions through self-learning and growth, a process of selecting / generating answers, and finally a process of automatically generating in-depth query responses or analysis reports.

[0021] The disaster safety knowledge base is composed of a composite inference knowledge enhancement department, a natural language understanding knowledge learning department, a human-mimicking knowledge learning department, and a knowledge resource collection / management department.

[0022] The composite inference knowledge enhancement department is configured to extract rules and knowledge relationships from stereotyped and non-stereotyped documents, and based on this, explore / infer new facts to generate knowledge.

[0023] The natural language understanding knowledge learning department is configured to analyze the structure, situation, context, and intention of queries input by users to improve natural language understanding capabilities.

[0024] The human-mimicking knowledge learning department is configured to perform self-learning based on the data accumulated in the knowledge base by mimicking human cognitive and judgment functions to improve performance. In addition, the knowledge resource collection / management department is configured to collect and manage non-stereotyped / stereotyped data and various overseas data.

[0025] The above-mentioned AI unit is composed of a user interface, a query / keyword input unit, a problem analysis unit, a user intention understanding unit, an answer candidate search / inference unit, an answer selection / generation unit, and a response generation unit, serving as a disaster safety knowledge bot capable of in-depth query response via AI.

[0026] The query / keyword input unit is composed of a query / keyword collection unit for collecting queries / keywords from the user, and a query identification unit for identifying contextual errors or typos in the query / keyword itself, requesting re-entry if necessary, or transmitting the query / keyword to the problem analysis unit.

[0027] The problem analysis unit is composed of a query interpretation unit for interpreting the sentence structure and words of the input query sentence, and a situation interpretation unit for interpreting the situation and context included in the query.

[0028] The user intention understanding unit is composed of an intention extraction unit for extracting the user's intention included in the query sentence and transmitting it to the intention interpretation unit, and an intention interpretation unit for interpreting the user's intention included in the data received from the intention extraction unit.

[0029] The answer candidate search / inference unit is composed of an answer candidate search unit for searching for answer candidates based on the analyzed query, and an answer candidate inference unit for inferring and ranking the optimal answer based on the user's intention and context from the answer candidate list.

[0030] The answer selection / generation unit is composed of an answer selection unit for selecting the optimal answer based on the ranked answer candidates, and an answer generation unit for generating an answer based on the selected answer.

[0031] The response generation unit is composed of an answer implementation unit for generating an answer in an easy-to-understand colloquial sentence for the user, and an answer display unit for displaying the answer on the user interface.

Advantages of the Invention

[0032] In the AI-based disaster safety knowledge integration management system according to the present invention, by utilizing the sharing of disaster safety data and intelligent analysis services, it becomes possible to provide a question-and-answer service for specialized knowledge in the field of disaster safety by AI, and an automatic reporting service that supports the creation of policy planning and reports for specific topics is also enabled. In addition, since the time, personnel, costs, and workload required for information search, analysis, and report creation can be dramatically reduced, even non-experts or those with little experience can perform analysis and generate policy materials with the support of machines. Furthermore, comprehensive judgments can be made considering data other than disasters, new technologies, and new problems, and the level of support for disaster safety policies is significantly improved through the integrated knowledge base in the field of disaster safety.

Brief Description of the Drawings

[0033]

Figure 1

Figure 2

Figure 3

Figure 4

Embodiments for Carrying Out the Invention

[0034] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. However, it goes without saying that the scope of the present invention is not limited here.

[0035] In this specification, this embodiment is provided to completely disclose the present invention, and the scope of the present invention is defined only by the claims. Therefore, in some embodiments, well-known components, well-known operations, and well-known technologies are not specifically described in order to avoid the present invention being ambiguously interpreted.

[0036] The terms used in this specification are for the purpose of describing embodiments and are by no means intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated otherwise in the context. Also, components or operations referred to as "including (or comprising)" do not exclude the existence or addition of one or more other components or operations.

[0037] The present invention relates to an AI-based integrated disaster safety knowledge management system that utilizes an intelligent analysis service for disaster safety data sharing and disaster safety data, enables a question-and-answer service for specialized knowledge in the field of disaster safety by AI, and enables an automatic reporting service that supports policy planning and report creation for specific topics.

[0038] Generally classified, the present invention can be regarded as composed of a disaster safety knowledge base part combined with a data network and an artificial intelligence part for realizing high-dimensional information processing of humans by artificial intelligence. (See Figure 1)

[0039] The above-mentioned disaster safety knowledge base part is composed of a data collection part for collecting and aggregating various information from external institutions, a data transmission part for transmitting the aggregated information to a server via LTE, 5G, WIFI, etc., and a big data part for analyzing and storing the transmitted data.

[0040] In the above-mentioned data collection part, general materials such as disaster situations, disaster statistics, disaster policies, situation reports, research reports, etc., unstructured data such as news, SNS, laws, situation reports, research reports, etc., and structured data such as damage restoration, weather, safety indexes, disaster warnings, 119 reports, etc. are provided. In some cases, data collection by IoT is also possible through information exchange by the cooperation of all objects such as machine-machine and machine-human. In the above-mentioned big data part, the data accumulation and analysis ability will be strengthened due to the improvement of information processing ability.

[0041] In the above artificial intelligence unit, by utilizing the data accumulated and analyzed in the above big data unit, human cognitive abilities (such as language, voice, vision, and sensibility) and judgments and inferences based on learning and inference functions are executed. Through rapid learning using data, machines are intelligentized so that new values can be created.

[0042] On the other hand, in the above data collection unit, data may be provided through system linkage or cooperation between institutions. For example, stable inspection information may be provided in linkage with the national security information integrated disclosure system, disaster management information may be provided in linkage with the national disaster management information system, stable statistics may be provided in linkage with the stable information integrated management system, and safety report information may be provided in linkage with the safety newspaper.

[0043] Also, disaster cases may be provided through cooperation with disaster response agencies, disaster case white papers may be provided through cooperation with white paper writing agencies, and research reports may be provided through cooperation with disaster safety research institutions.

[0044] In the above big data unit, the data provided from the above data transmission unit is distinguished, analyzed, and accumulated. Real-time data such as steep slope sensors, water level information, and photos of disaster accident sites, structured data such as past damage history information and safety information of various facilities, and unstructured data such as disaster situation report texts, disaster situation report images, and briefing materials are distinguished and analyzed.

[0045] Then, these distinguished and analyzed data are classified into the damage situation of disaster accidents, the response history of disaster accidents, disaster safety policy information, etc., and are accumulated as archive data, enabling the public disclosure and sharing of data with target persons (such as the general public and relevant institutions) who need the data.

[0046] On the one hand, in order for fragmented data to be utilized as useful information, it is necessary to comprehensively explore data and experiential knowledge, transform them into knowledge, apply them to policy-making and disaster management, and gain insights into potential risk factors. Starting from the state of raw data, which is an objective fact, meaning is derived, relevant / correlative relationships are understood through pattern recognition, the data is informationized, and the resulting information encapsulated as unique knowledge is structured and knowledgeized, ultimately leading to the derivation of wisdom.

[0047] Figure 2 is a diagram showing the outline of the disaster safety knowledge consulting technology according to the present invention. In the data collection part of the disaster safety knowledge base, overseas information such as disaster situations, disaster statistics, and disaster policies, unstructured data such as news, SNS, and laws, and structured data such as damage restoration, meteorology, and safety indices are collected and accumulated in the big data part of the disaster safety knowledge base through the collection / management of knowledge resources or human-mimicking knowledge learning and natural language understanding knowledge learning. In the above-mentioned big data part, in addition to knowledge learning of natural language understanding, knowledge curation and knowledge expansion of composite inference are also provided. In the above-mentioned artificial intelligence part, the data accumulated in the above-mentioned big data part is utilized to execute judgment and inference by human cognitive abilities (language, voice, vision, sensibility, etc.) and learning and inference functions.

[0048] It is composed of the process of understanding the user's intention through query interpretation from externally provided queries and keywords and problem analysis through situation interpretation, the process of searching / inferring answer candidates, the process of selecting / generating answers through self-learning and growth, and finally the process of automatically generating in-depth query responses or analysis reports.

[0049] Figure 3 is a configuration diagram of the integrated disaster safety knowledge management system by AI according to the present invention. The integrated disaster safety knowledge management system by AI according to the present invention relates to a disaster safety knowledge bot that can not only provide factual information such as disaster safety cases, experiences, and statistics by utilizing language, learning, and inference intelligent (AI) technologies but also enable in-depth question answering through question inference.

[0050] The disaster safety knowledge base is disaster safety knowledge data that utilizes big data and AI technologies to support efficient information sharing in disaster situations for database decision-making support. It consists of a base composite inference knowledge expansion section, a natural language understanding knowledge learning section, a human imitation knowledge learning section, and a knowledge resource collection / management section.

[0051] The above-mentioned composite inference knowledge expansion section extracts rules and knowledge relationships from structured and unstructured documents, and based on this, explores / inferences new facts to generate knowledge.

[0052] The above-mentioned natural language understanding knowledge learning section accumulates the results of analyzing the structure, situation, context, and intention of queries input by users, and improves the natural language understanding ability.

[0053] The above-mentioned human imitation knowledge learning section imitates the human recognition and judgment functions, and improves the performance through self-learning based on the data accumulated in the knowledge base.

[0054] The above-mentioned knowledge resource collection / management section is configured to collect and manage unstructured / structured data and various overseas data.

[0055] On the other hand, the artificial intelligence section can be regarded as a disaster safety knowledge bot capable of in-depth question answering through AI. It consists of a user interface, a question / keyword input section, a question analysis section, a user intention understanding section, an answer candidate search / inference section, an answer selection / generation section, and an answer generation section.

[0056] The above-mentioned question / keyword input section is composed of a question / keyword collection section for collecting questions / keywords from users, and a question identification section for identifying contextual errors / typos in the questions / keywords themselves and requesting re-entry or transmitting the questions / keywords to the question analysis section.

[0057] The above-mentioned question analysis section is composed of a question interpretation section for interpreting the sentence structure and words of the input question sentence, and a situation interpretation section for interpreting the situation and context included in the question sentence.

[0058] The above user intention understanding unit is composed of an intention extraction unit for extracting the user's intention included in the question sentence and transmitting it to the intention interpretation unit, and an intention interpretation unit for interpreting the user's intention included in the data received from the intention extraction unit.

[0059] The above answer candidate search / inference unit is composed of an answer candidate search unit for searching for answer candidates based on the analyzed question, and an answer candidate inference unit for inferring and ranking the optimal answer based on the user's intention and context in the answer candidate list.

[0060] Homophones require different answers depending on the user's intention and context.

[0061] The above answer selection / generation unit is composed of an answer selection unit for selecting the optimal answer based on the ranked answer candidates, and an answer generation unit for generating an answer based on the selected answer.

[0062] The above answer generation unit is composed of an answer implementation unit for generating an answer in an easy-to-understand colloquial sentence by the user, and an answer expression unit for displaying the answer on the user interface.

[0063] Also, the artificial intelligence unit can additionally include a report generation unit, a language generation unit based on a knowledge base, and a content processing and generation unit in order to automatically support policy planning and the creation of report materials and provide services for supporting decision-making and reducing time and personnel consumption.

[0064] The above report generation unit is composed of a template generation and content composition unit for generating report content in accordance with a template format, and a template and output interface unit for providing the generated report content in accordance with the template format so that the user can easily understand it.

[0065] The above knowledge base language generation unit is composed of a data extraction and support search unit for extracting and searching necessary data from the above disaster safety knowledge base, and a content composition and data catalog for constructing content based on the data extracted and searched thereby.

[0066] The above content processing and generation unit is composed of a table / graph and text processing unit for generating statistical data (table / graph) based on text, and a title and content automatic generation unit for generating content in a user-friendly format and extracting / summarizing the main content.

[0067] Therefore, in the artificial intelligence unit, it is possible to gradually execute judgments and inferences by human perception abilities such as language, voice, vision, and sensibility, and learning and inference functions, by utilizing the data accumulated and analyzed in the above big data unit.

[0068] Also, in the question / keyword input unit of the above artificial intelligence unit, in-depth question answering is performed by question answering / keyword search by a chatbot driving device. It is composed of a question answering pre-management module, a user intention grasping module, a dialogue agent, and modules of a dialogue management system, together with a chatbot driving device, an input device, a management system, and a machine learning tool. In the above question answering pre-management module, the transmission content of the Intent Finder and the Dialog Agent is managed through session control.

[0069] A session is generated by utilizing four conditions: USER, DEVICE, CHATBOT, and a certain time.

[0070] In the above user intention grasping module, it is configured to grasp the user's intention, send a message to the optimal DIALOG AGENT, and support the search for answers after grasping the dialogue intention based on various features.

[0071] In the above-described dialogue agent system, it is configured to respond to a user's utterance in a dialogue agent platform environment, develop and distribute it in a disaster safety data environment, and register and execute it through ADMI.

[0072] The above-described dialogue management system can conduct various dialogues such as daily conversations and news depending on the type of Q&A engine and knowledge base, and is configured to enable the definition of slots and tasks in the SDS scenario generated by the dialogue modeling tool.

[0073] The present invention includes a function for semi-automatically or automatically generating a report. In the report design and report server part, it is configured to generate a report via a data link and provide procedures for report integration / distribution and report utilization.

[0074] Each report created based on various report creation sources (such as a knowledge base) in conjunction with a report agent is integrated and distributed on the distribution server side, and then the content of the report reflected in the content management system of the application server is finally transferred to the report server and configured to be publicly available to users.

[0075] As described above, in the disaster safety knowledge integration management system by AI according to the present invention, by utilizing disaster safety data sharing and knowledge analysis services for disaster safety data, it becomes possible to provide a question-and-answer service for specialized knowledge in the field of disaster safety by AI, and there is an effect that an automatic report service for supporting policy planning and creation of report materials for a specific theme becomes possible.

[0076] Furthermore, it is possible to dramatically reduce the time, personnel, costs, and workload involved in data exploration, analysis, and report preparation. Even non-experts or those with relatively little experience can be supported by machines to enable analysis and the generation of policy materials, and by supporting comprehensive judgment considering data other than disasters, new technologies, and new problems, there is also an effect of significantly improving the level of support for disaster safety policies through the insights of an integrated knowledge base in the field of disaster safety.

[0077] Although the technical idea of the present invention has been specifically described in the preferred embodiments, please note that the above embodiments are for the purpose of explanation and not for limitation. It is obvious that various modifications and corrections are possible within the scope of the technical idea of the present invention, and therefore, it is natural that these modifications and corrections are included in the claims.

Claims

1. By sharing disaster safety data and utilizing intelligent analysis services, an AI-based question and answer service for expertise in the field of disaster safety becomes possible, and an automatic reporting service that supports policy planning and report creation for specific topics becomes possible. It relates to an AI-based integrated disaster safety knowledge management system. It consists of a disaster safety knowledge base part combined with a data network and an AI part for realizing high-dimensional information processing with AI. The above-mentioned disaster safety knowledge base part is composed of a data collection part for collecting and aggregating various information from external institutions, a data transmission part for transmitting the collected information to a server through LTE, 5G, WIFI, etc., and a big data part for analyzing and accumulating the transmitted data. In the above big data part, in order to classify, analyze, and accumulate the data provided by the data transmission part, real-time data such as steep slope sensors, quantity water level information, and photos of disaster accident sites, regular data such as past damage history information and safety information of various facilities, and unstructured data such as disaster situation report texts, disaster situation report images, and briefing materials are classified and analyzed. The data classified and analyzed in this way is classified into disaster accident damage situations, disaster accident response histories, and disaster safety policy information and stored as archive data. In the above AI part, by utilizing the materials accumulated and analyzed by the big data part, the judgment and inference by human cognitive abilities (language, voice, vision, sensibility, etc.) and learning and inference functions are executed step by step, and the machine is intelligentized. The above AI part is composed of a disaster safety knowledge bot capable of deep question and answer by AI, a user interface, a question / keyword input part, a problem analysis part, a user intention understanding part, a candidate answer search / inference part, an answer selection / generation part, and an answer generation part. In the question / keyword input section of the artificial intelligence unit, deep question answering is performed through question answering / keyword search by the chatbot driving device. Together with the chatbot driving device, input device, management system, and machine learning tool, it is composed of a question answering pre-management module, a user intention understanding module, a dialogue agent, and modules of the dialogue management system. In the above question answering pre-management module, the transmission content of the Intent Finder and Dialog Agent is managed through session control. In the user intention understanding module, the user's intention is grasped and a message is sent to the optimal DIALOG AGENT. After grasping the dialogue intention based on various features, it is configured to support answer search. The above dialogue agent system responds to the user's speech in the dialogue agent (Dialog Agent) platform environment, is developed and distributed in the disaster safety data environment, and is configured to be registered and executed through ADMI. The above dialogue management system can conduct various dialogues such as daily conversations and news according to the types of Q&A engines and knowledge bases, and is characterized in that slot and task definitions can be made for the SDS scenarios generated by the dialogue modeling tool. It is an AI-based disaster safety knowledge integration management system.

2. In the first item, in the artificial intelligence unit, from the state of the raw data, which is an objective fact, through semantic extraction and pattern recognition, the relevant / correlation relationship is understood, informationized, and encapsulated as its own knowledge. The resulting information is structured and knowledgeized, and finally, the structuring of knowledge is derived. It is an AI-based disaster safety knowledge integration management system characterized by this.

3. In the first aspect, in the disaster safety knowledge-based data collection unit, overseas information, unstructured data, and structured data are collected, and after going through knowledge resource collection / management or human-mimicking knowledge learning and natural language understanding knowledge learning, they are accumulated in the big data unit of the disaster safety knowledge base. In the big data unit, in addition to natural language understanding knowledge learning, knowledge curation and composite inference knowledge expansion are both provided. An AI-based disaster safety knowledge integration management system characterized by this.

4. In the first or second aspect, in the artificial intelligence unit, the data accumulated and analyzed in the big data unit is utilized to perform judgment and inference by human cognitive abilities (such as language, voice, vision, and sensibility) and learning and inference functions. It includes the process of analyzing problems through query interpretation and situation interpretation from queries and keywords provided externally to understand the user's intention; the process of searching / inferring answer candidates; The process of self-learning and growth, judgment / prediction, and selection / generation of answers; and finally, an AI-based disaster safety knowledge integration management system characterized by performing deep query responses or automatically generating analysis reports.

5. In the first or third aspect, the disaster safety knowledge base aims to support decision-making in the database and, as disaster safety knowledge data that utilizes big data and AI technology to support efficient information sharing in disaster situations, consists of a composite inference knowledge enhancement unit; a natural language understanding knowledge learning unit; a human-mimicking knowledge learning unit; and a cognitive resource collection / management unit. An AI-based disaster safety knowledge integration management system.

6. In the fifth aspect, the above-mentioned composite inference knowledge enhancement unit is configured to extract rules and knowledge relationships from structured and unstructured documents and, based on that, explore / infer new facts to generate knowledge. An AI-based disaster safety knowledge integration management system characterized by this.

7. In claim 5, the above natural language understanding knowledge learning unit is characterized by accumulating the results of analyzing the structure, situation, context, and intention of the question sentence input by the user and being configured to improve the natural language understanding ability. An AI-based integrated disaster safety knowledge management system

8. In claim 1, the above question / keyword input unit is composed of a question / keyword collection unit for collecting questions / keywords from the user, and a question identification unit for identifying contextual errors and typos in the question / keyword itself and requesting re-entry or transmitting the question / keyword to the question analysis unit. An AI-based integrated disaster safety knowledge management system.

9. In claim 1, the above question analysis unit is composed of a question interpretation unit for interpreting the sentence structure and words of the input question sentence, and a situation interpretation unit for interpreting the situation and context included in the question sentence. An AI-based integrated disaster safety knowledge management system.

10. In claim 1, the above user intention understanding unit is composed of an intention extraction unit for extracting the user's intention included in the question sentence and transmitting it to the intention interpretation unit, and an intention interpretation unit for interpreting the user's intention included in the data received from the intention extraction unit. An AI-based integrated disaster safety knowledge management system.

11. In claim 1, the above answer candidate search / inference unit is composed of an answer candidate search unit for searching for answer candidates based on the analyzed question, and an answer candidate inference unit for inferring and ranking the optimal answer based on the user's intention and context in the answer candidate list. An AI-based integrated disaster safety knowledge management system.

12. In claim 1, the above answer selection / generation unit is composed of an answer selection unit for selecting the optimal answer based on the ranked answer candidates, and an answer generation unit for generating an answer based on the selected answer. An AI-based integrated disaster safety knowledge management system.

13. In the first aspect, the response generation unit of the integrated disaster safety knowledge management system using AI is characterized by being composed of a response implementation unit for generating a response in an easy-to-understand colloquial sentence and a response display unit for transmitting the response to the user interface.

14. In the first aspect, the artificial intelligence unit of the integrated disaster safety knowledge management system using AI additionally comprises a report generation unit, a language generation unit based on a knowledge base, and a content processing and generation unit to provide services for assisting in decision-making by automatically supporting policy planning and report materials and reducing the consumption of time and personnel.

15. In the 14th aspect, the report generation unit of the integrated disaster safety knowledge management system using AI is characterized by being composed of a template generation and content composition unit for generating report content in accordance with a template format and a template and output interface unit for providing the generated report content in a template format that is easy for the user to understand.

16. In the 14th aspect, the language generation unit based on the knowledge base of the integrated disaster safety knowledge management system using AI is characterized by being composed of a data extraction and support search unit for extracting and searching for necessary data from the disaster safety knowledge base and a content composition and data catalog for constructing content based on the extracted and searched data.

17. In the 14th aspect, the content processing and generation unit of the integrated disaster safety knowledge management system using AI is characterized by being composed of a table / graph and text processing unit for generating statistical data (table / graph) in text form and a title and content automatic generation unit for generating content in a form that is easy for the user to understand and extracting and summarizing the main content.

18. In the fourth item, in the report design and report server part, it is configured to be able to generate reports through data links and provide procedures for report integration / distribution and report utilization. However, for each individual report created based on various report creation sources (such as knowledge bases) linked with the report agent, after being integrated and distributed on the distribution server side, the content of the report reflected in the content management system of the application server is finally transferred to the reporting server and configured to be publicly available to users by an AI-based disaster safety knowledge integration management system.

Images