Domain expert knowledge-driven crisis event evolution context generation method and device

Abstract

The invention discloses a domain expert knowledge-driven crisis event evolution context generation method and device, and belongs to the field of artificial intelligence, and the method comprises the steps: building a basic data model based on domain expert knowledge experience by using a program, and operating the model based on a computer processor to support the whole event evolution context analysis; performing event detection analysis based on Internet data by using a computer to form event sentences, and mining basic crisis clues and important crisis clues by using a computer operation model; meanwhile, the development trend of the crisis event is tracked; a computer is used for carrying out time sequence, cooperation and causal relationship analysis on the basis of basic crisis clues, important crisis clues and development trends, connection among crisis events is enhanced, and crisis event evolution venation generation with different dimensions of event types and target types is realized through crisis event classification by using the computer. According to the invention, a complete and abundant crisis event evolution context can be generated.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and more specifically, to a method and device for generating a crisis event evolution context driven by domain expert knowledge. Background Art

[0002] Crisis refers to a moment of danger and disaster. It is a moment that tests decision-making and problem-solving capabilities. It is a turning point in the development of life, groups, and society, involving life and death, interest transfer, and other aspects. Crisis events are usually sudden or major social events that may directly threaten the basic goals and values ​​of individuals, organizations, or society, such as life, property, safety, and order. In today's era of globalization, the world is facing many problems such as rising temperatures, increasing extreme weather events, environmental pollution, resource shortages, and economic instability, all of which are huge crisis risks. Therefore, it is necessary to strengthen the ability and level of crisis identification, mastery, and control, and take effective and special measures to deal with it.

[0003] Crises are generally significant, complex, dangerous, but difficult to predict, and have strong suddenness and immediacy. With the vigorous development of Internet technology, cyberspace data is increasing day by day, and the amount of data is exploding. Crises are often obscured by boundless information. Simply through human observation and sorting, it is no longer possible to effectively and quickly complete the comprehensive analysis and grasp of crisis events, which is not conducive to the prevention and control of crisis events. Therefore, with the advancement of science and technology and the accumulation of a large amount of knowledge and experience in the field, it is urgent to effectively combine technology and knowledge, conduct evolutionary analysis on the development path of crisis events, generate information such as the previous background and current status of the crisis event process, and assist in improving the comprehensive grasp and understanding of crisis events and reducing the damage and losses caused by crises. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a method and device for generating a crisis event evolution context driven by domain expert knowledge. A method and a device are provided using computer technology, which can be used to generate a complete and rich crisis event evolution context.

[0005] The object of the present invention is achieved through the following solutions:

[0006] A method for generating crisis event evolution context driven by domain expert knowledge includes the following steps:

[0007] S1, based on the knowledge and experience of domain experts, use the program to build a basic data model containing crisis event characteristics and crisis event indicators in the field, and run the model based on a computer processor to support the analysis of the entire event evolution context;

[0008] S2, using computers to perform event detection and analysis based on Internet data to form event sentences, and using computers to run the model to mine basic crisis clues and important crisis clues; and at the same time track the development trend of crisis events;

[0009] S3, using computers to conduct time series, coordination and causal relationship analysis based on the basic crisis clues, important crisis clues and development trends, strengthen the connection between crisis events, and use computers to classify crisis events to achieve the generation of crisis event evolution context in different dimensions of event type and target type.

[0010] Furthermore, the basic crisis clues include relevant meetings, speeches and holidays; the important crisis clues include behaviors and states; and the development trend of the crisis event includes the occurrence, aggravation, weakening and resolution of the crisis event.

[0011] Furthermore, in step S1, based on the domain expert knowledge and experience, a basic data model including crisis event characteristics and crisis event indicators in the domain is constructed using a program, which specifically includes the following sub-steps:

[0012] Using computer programming, data models are constructed from two dimensions: crisis event characteristics and crisis event indicators, and a unified crisis event model template is established. Then, crisis event models in different fields are based on this, and the knowledge of domain experts is used to visualize and instantiate the crisis event model.

[0013] Further, in step S2, the computer is used to perform event detection analysis based on Internet data to form event sentences, and the computer is used to run the model to mine basic crisis clues and important crisis clues, specifically including text data analysis and spatiotemporal data analysis;

[0014] The text data analysis specifically includes the following sub-steps:

[0015] Taking the crisis event indicators of the crisis event model as the guide, we first use computers to mine background signs, perform event detection on text data based on the Bert pre-trained language model, obtain event sentences, and then classify events to obtain major meeting events and speech events, and analyze the emotional polarity of speech. At the same time, we build an important holiday database based on the knowledge of domain experts to obtain current important holiday information in real time.

[0016] Then, we use computers to mine behavioral signs, perform semantic calculations on text and behavioral indicators based on the Bge pre-trained language model, and obtain corresponding action events and deployment events.

[0017] Finally, computers are used to build a state and plan keyword library based on domain expert knowledge to detect the target's current state and future plans, and obtain signs of crisis events;

[0018] The spatiotemporal data analysis specifically includes the following sub-steps:

[0019] Based on the crisis event indicators of the crisis event model and the characteristics of the crisis event, the distance calculation, area calculation, time calculation and the combination of the three are performed by computer. The distance calculation refers to the distance between the target and the designated area, the area calculation refers to whether it is in a designated area, and the time calculation refers to the target's activity events.

[0020] Single-target events are mainly calculated based on distance and area. For example, if the target is in a certain area, group-target events are mainly calculated based on the three dimensions of distance, area, and time. For example, if a group of targets meet, the final calculation result meets the requirements of the crisis event index, and the corresponding crisis event index is used as a sign of a crisis event.

[0021] Furthermore, in step S2, tracking the development trend of the crisis event specifically includes the following sub-steps:

[0022] First, an expert knowledge base is constructed using computers to specify the rules for crisis occurrence, crisis event aggravation, crisis event mitigation, and crisis event resolution. According to the specified rules, semantic matching calculations are performed with the pre-trained language model and text data.

[0023] Then, use the computer to determine whether the crisis event has occurred. If the crisis event has not occurred, continue to perform semantic matching from the text data. If the crisis event has occurred, further use the computer to calculate whether the crisis event has intensified, weakened, or resolved. If the crisis event has resolved, save the occurrence, intensification, weakening, and resolution analysis results. At the same time, guided by the rules in the expert knowledge base, use the computer to calculate the development status of the crisis event through spatiotemporal data.

[0024] Finally, the crisis event stages obtained by matching text data and calculating spatiotemporal data are sorted and deduplicated in chronological order to obtain the final description of the crisis event development stages of occurrence, aggravation, weakening and elimination.

[0025] Further, in step S3, the computer is used to perform time series, coordination and causal relationship analysis based on the basic crisis clues, important crisis clues and development trends, strengthen the connection between crisis events, and use the computer to classify crisis events to achieve the generation of crisis event evolution context in different dimensions of event type and target type, which specifically includes sub-steps:

[0026] First, computers are used to define the synergistic and causal relationships among crisis events in the field based on expert knowledge and experience;

[0027] Then, based on the crisis event signs and crisis event development stage descriptions, the crisis event set is sorted in time, and the crisis event labels and crisis event relationships are classified based on the Bert pre-training model to form three types of crisis event labels: background, behavior, and status. The synergistic and causal relationships between crisis events are given, and finally the crisis events are constructed according to the six dimensions of background, behavior, status, goal, relationship, and signs to obtain a complete and rich crisis event evolution context.

[0028] Furthermore, the step of calculating the development status of the crisis event through spatiotemporal data specifically includes the following sub-steps:

[0029] First, the type of crisis event is determined by computer. If it is a single-target event, the location of the crisis event is further determined. Then, the crisis event stage calculation is performed to calculate whether the distance and area meet the requirements of the rules. If the calculation result does not meet the requirements for the occurrence of the crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of the crisis event, the crisis event is continued to be calculated to determine whether it is aggravated, weakened, and resolved. If the crisis event is not resolved, the crisis event calculation is continued. If the crisis event is resolved, the above-mentioned occurrence, aggravation, weakening, and resolution analysis results are saved and the results are generated into a text description according to the format;

[0030] If it is a group target event, the location of the crisis event is further determined, and then the time granularity of the spatiotemporal data of different targets is aligned, and the spatiotemporal data of different targets are classified according to intervals. Then the median of each interval of different targets is used as the data value after the alignment of their respective data granularity to achieve the consistency of the time scale of different targets. Then the crisis event stage calculation is performed to calculate whether the distance, area, and time between the group targets meet the requirements of the rules. If the calculation result does not meet the requirements for the occurrence of the crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of the crisis event, the crisis event is continued to be calculated. Whether the crisis event is intensified, weakened, and eliminated, if the crisis event is not eliminated, the crisis event calculation is continued. If the crisis event is eliminated, the occurrence, intensification, weakening, and elimination analysis results are saved and the results are generated into a text description according to the format.

[0031] A device for generating crisis event evolution context driven by domain expert knowledge, comprising a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded by the processor, any of the methods described above is executed.

[0032] The beneficial effects of the present invention include:

[0033] (1) Domain-specificity: The device and method of the present invention focus on the importance and indispensability of domain knowledge in the crisis event analysis process. Domain knowledge is both the foundation and the core. Compared with analysis methods that are mainly data-driven or data-based and knowledge-supplemented, the device and method are more domain-specific and the analysis results in the fields of military, politics, economy, diplomacy, security, science and technology are also reliable.

[0034] (2) Completeness. The device and method of the present invention focuses on the analysis of the previous background and current status of the crisis event, that is, considering the continuous evolution of the current status, and also deeply explores the historical background of the crisis event, comprehensively analyzes the various logical relationships between events, and forms a complete and rich evolution context of the crisis event;

[0035] (3) Practicality. The device and method of the present invention are based on the actual needs of crisis analysis in different fields, take the knowledge of field experts as the starting point, and set up an overall analysis idea in the program flow. Compared with general methods, it is more focused and practical. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1 is a flowchart of the overall steps of the method according to the embodiment of the present invention;

[0038] Figure 2 Build a framework for crisis event modeling;

[0039] Figure 3 A schematic diagram of the principles of crisis event mining;

[0040] Figure 4 Tracking workflow diagrams for crisis events;

[0041] Figure 5 This is a schematic diagram of the principle of crisis event evolution analysis;

[0042] Figure 6 This is a schematic diagram of the evolution of crisis events. DETAILED DESCRIPTION

[0043] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0044] In view of the current situation in the background, the specific implementation process of the present invention is as follows:

[0045] like Figure 1 As shown, according to the present invention, a computer device is provided and a running program is executed to perform the following steps: first, based on the knowledge and experience of domain experts, a set of basic data models including crisis event characteristics and crisis event indicators in the domain is constructed to support the analysis of the entire event evolution context; then, event detection and analysis is performed based on Internet data to form event sentences, and basic crisis clues such as relevant major meetings, speeches of politicians, holidays, and important crisis clues such as behaviors and states are mined; at the same time, the development trend of crisis events is continuously tracked, including the occurrence, aggravation, weakening and elimination of crisis events; finally, based on the previous basic crisis clues, important crisis clues and development trends, time series, coordination and causal relationship analysis are performed to strengthen the connection between crisis events, and through crisis event classification, the generation of crisis event evolution context in different dimensions such as event type and target type is realized.

[0046] like Figure 2As shown, firstly, a crisis event model is constructed as a basic data model to support subsequent crisis event mining and evolution analysis. The present invention constructs data models from two dimensions: crisis event characteristics and crisis event indicators, establishes a unified crisis event model template, and then crisis event models in different fields are based on this, driven by expert knowledge, to visualize and instantiate the crisis event model. Specifically, crisis event characteristics are divided into four categories: participating countries, participating targets, occurrence regions, and crisis types. Among them, participating countries are subdivided into country groups and single countries, and country groups are constructed based on national interests; participating targets are subdivided into people, ships and aircraft, and occurrence areas are subdivided into cities and sea areas; crisis types are subdivided into group target events and single target events. Group target events use people, ships and aircraft as analysis objects and are constrained by country groups. If the occurrence area is a city, the analysis of people's activities is the research focus. If the occurrence area is a sea area, the analysis of ships and aircraft activities is the research focus. Single target events use people, ships and aircraft as analysis objects and are constrained by a single country. If the occurrence area is a city, the analysis of people's activities is the research focus. If the occurrence area is a sea area, the analysis of ships and aircraft activities is the research focus. Crisis event indicators are divided into three categories: background indicators, behavior indicators and status indicators. Among them, background indicators mainly include important political speeches, major social activities, and important holidays. Important political speeches refer to major statements made by major political figures in different countries on potential crisis risk issues such as environment, resources, economy, society, and territory. Different fields can build a database of important political leaders and a risk database of crisis issues based on expert knowledge according to needs and focus, and maintain political leaders and crisis risk points. Major social activities include visits, meetings and other activities. Important holidays include major traditional holidays in different countries, such as the Spring Festival and Thanksgiving. Behavioral indicators are divided into two categories: action and deployment. Action mainly refers to a specific activity, and deployment mainly refers to prior arrangements, which can be a specific activity or a comprehensive reflection of multiple activities. By subdividing actions and deployments, a clearer evolution context can be obtained. Different fields can also be unified as behavior based on actual needs, and no specific distinction is made between the two. Status indicators are divided into two categories: current status and future plans. Current status refers to the current activity of the target, reflecting a static feature, such as the target being in a certain area, and future plans refer to the future activity of the target, reflecting a dynamic feature, such as visiting a certain country in the next few days.

[0047] like Figure 3As shown in the figure, based on the above crisis event model, the text data and spatiotemporal data obtained from the Internet are analyzed for signs, and the background before the crisis event, that is, the crisis signs, is mined. Specifically, the text data analysis is guided by the crisis event indicators of the crisis event model. First, the background signs are mined, and the text data is detected based on the pre-trained language model such as Bert to obtain event sentences. Then, the event classification is performed to obtain major meeting events and speech events, and the emotional polarity of the speech is analyzed, such as positive, negative, etc. At the same time, an important holiday library is built based on expert knowledge to obtain the current important holiday information in real time; then the behavior signs are mined, and the text and behavior indicators are semantically calculated based on the pre-trained language model such as Bge to obtain the corresponding action events and deployment events; finally, the state and plan keyword library is built based on the domain expert knowledge to detect the current state and future plans of the target and obtain the signs of crisis events. Spatiotemporal data analysis is guided by the crisis event indicators of the crisis event model and based on the characteristics of the crisis event. It performs distance calculation, area calculation, time calculation and a combination of the three. Distance calculation refers to the distance between the target and the specified area, area calculation refers to whether it is in a specified area, and time calculation refers to the target's activity events. Single-target events are mainly based on distance calculation and area calculation. If the target is in a certain area, group target events are mainly based on a comprehensive calculation of the three dimensions of distance, area and time. If the group targets meet, and the final calculation result meets the requirements of the crisis event indicators, the corresponding crisis event indicators will be used as signs of crisis events.

[0048] like Figure 4As shown, based on the above crisis event model, the text data and spatiotemporal data obtained from the Internet are tracked and analyzed to determine the development status of the crisis event. Specifically, first, an expert knowledge base is constructed to specify the rules for crisis occurrence, crisis event aggravation, crisis event reduction, and crisis event resolution. According to the specified rules, a semantic matching calculation is performed between the pre-trained language model and the text data to determine whether the crisis event has occurred. If the crisis event has not occurred, semantic matching is continued from the text data. If the crisis event has occurred, similarly, whether the crisis event has intensified, reduced, or resolved is calculated. If the crisis event has resolved, the aforementioned analysis results of occurrence, aggravation, reduction, and resolution are saved. At the same time, guided by the rules in the expert knowledge base, the development status of the crisis event is calculated through spatiotemporal data. First, the type of crisis event is determined. If it is a single-target event, the location of the crisis event is further determined, and then the crisis event stage calculation is performed to calculate whether the distance, area, etc. meet the rule requirements. If the calculation result does not meet the requirements for the occurrence of the crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of the crisis event, the crisis event is continued to be calculated. If the crisis event is not resolved, the crisis event calculation is continued. If the crisis event is resolved, the above-mentioned occurrence, aggravation, weakening, and elimination analysis results are saved and the results are generated in the format of "time + target + crisis event name + occurrence / aggravation / weakening / dissolution" to generate a text description (such as a certain year, month, and day, a certain target went out for a visit); similarly, if it is a group target event, the location of the crisis event is further determined, and then the time granularity of the spatiotemporal data of different targets is aligned, specifically using the time specified by expert knowledge and experience (such as 30 The time scale of different targets is consistent. Then, the crisis event stage calculation is performed to calculate whether the distance, area, time, etc. between group targets meet the rule requirements. If the calculation result does not meet the requirements for the occurrence of a crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of a crisis event, the crisis event calculation is continued to calculate whether the crisis event is intensified, weakened, and eliminated. If the crisis event is not eliminated, the crisis event calculation is continued. If the crisis event is eliminated, the above-mentioned occurrence, intensification, weakening, and elimination analysis results are saved and the results are generated into a text description in the format of "time + target + crisis event name + occurrence / intensification / weakening / elimination" (such as a certain year, a certain month, and a certain day, a certain target meets a certain target). Finally, the crisis event stages obtained by matching the above text data and calculating the spatiotemporal data are sorted in chronological order and deduplicated to obtain the final description of the crisis event development stages such as occurrence, aggravation, weakening, and resolution.

[0049] like Figure 5 As shown in the figure, firstly, based on the expert knowledge and experience, the collaborative relationship (such as the coordination and cooperation relationship between a certain goal and a certain goal) and causal relationship (such as the occurrence of type A event will lead to the occurrence of type B event) of crisis events in the field are defined. Then, based on the above crisis event signs and the description of the crisis event development stage, the crisis event set is sorted in time and the crisis event label classification and crisis event relationship classification are performed based on the Bert pre-training model to form three types of crisis event labels: background, behavior, and state. The collaborative relationship and causal relationship between crisis events are given. Finally, the crisis event is constructed according to the six dimensions of background, behavior, state, goal, relationship, and sign to obtain a complete and rich crisis event evolution context. For the specific result diagram, please refer to Figure 6 .

[0050] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0051] According to one aspect of an embodiment of the present invention, a computer program product or a computer program is provided, the computer program product or the computer program includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in the above various optional implementations.

[0052] As another aspect, an embodiment of the present invention further provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the above embodiment.

Claims

1. A method for generating crisis event evolution context driven by domain expert knowledge, characterized in that: The following steps are involved: S1, based on the knowledge and experience of domain experts, use the program to build a basic data model containing crisis event characteristics and crisis event indicators in the field, and run the model based on a computer processor to support the analysis of the entire event evolution context; S2, using a computer to perform event detection and analysis based on Internet data to form event sentences, and using a computer to run the model to mine basic crisis clues and important crisis clues; and at the same time track the development trend of crisis events; S3, using computers to conduct time series, coordination and causal relationship analysis based on the basic crisis clues, important crisis clues and development trends, strengthen the connection between crisis events, and use computers to classify crisis events to achieve the generation of crisis event evolution context in different dimensions of event type and target type.

2. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 1 is characterized in that: The basic crisis clues include relevant meetings, speeches and holidays; the important crisis clues include behaviors and status; the development trend of the crisis event includes the occurrence, aggravation, weakening and elimination of the crisis event.

3. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 1 is characterized in that: In step S1, based on the domain expert knowledge and experience, a basic data model including crisis event characteristics and crisis event indicators in the domain is constructed using a program, which specifically includes the following sub-steps: Using computer programming, data models are constructed from two dimensions: crisis event characteristics and crisis event indicators, and a unified crisis event model template is established. Then, crisis event models in different fields are based on this, and the knowledge of domain experts is used to visualize and instantiate the crisis event model.

4. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 1 is characterized in that: In step S2, the computer is used to perform event detection and analysis based on Internet data to form event sentences, and the computer is used to run the model to mine basic crisis clues and important crisis clues, specifically including text data analysis and spatiotemporal data analysis; The text data analysis specifically includes the following sub-steps: Taking the crisis event indicators of the crisis event model as the guide, we first use computers to mine background signs, perform event detection on text data based on the Bert pre-trained language model, obtain event sentences, and then classify events to obtain major meeting events and speech events, and analyze the emotional polarity of speech. At the same time, we build an important holiday database based on the knowledge of domain experts to obtain current important holiday information in real time. Then, we use computers to mine behavioral signs, perform semantic calculations on text and behavioral indicators based on the Bge pre-trained language model, and obtain corresponding action events and deployment events. Finally, computers are used to build a state and plan keyword library based on domain expert knowledge to detect the target's current state and future plans, and obtain signs of crisis events; The spatiotemporal data analysis specifically includes the following sub-steps: Based on the crisis event indicators of the crisis event model and the characteristics of the crisis event, the distance calculation, area calculation, time calculation and the combination of the three are performed by computer. The distance calculation refers to the distance between the target and the designated area, the area calculation refers to whether it is in a designated area, and the time calculation refers to the target's activity events. Single-target events are mainly calculated based on distance and area. For example, if the target is in a certain area, group-target events are mainly calculated based on the three dimensions of distance, area, and time. For example, if a group of targets meet, the final calculation result meets the requirements of the crisis event index, and the corresponding crisis event index is used as a sign of a crisis event.

5. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 1 is characterized in that: In step S2, tracking the development trend of the crisis event specifically includes the following sub-steps: First, an expert knowledge base is constructed using computers to specify the rules for crisis occurrence, crisis event aggravation, crisis event mitigation, and crisis event resolution. According to the specified rules, semantic matching calculations are performed with the pre-trained language model and text data. Then, use the computer to determine whether the crisis event has occurred. If the crisis event has not occurred, continue to perform semantic matching from the text data. If the crisis event has occurred, further use the computer to calculate whether the crisis event has intensified, weakened, or resolved. If the crisis event has resolved, save the occurrence, intensification, weakening, and resolution analysis results. At the same time, guided by the rules in the expert knowledge base, use the computer to calculate the development status of the crisis event through spatiotemporal data. Finally, the crisis event stages obtained by matching text data and calculating spatiotemporal data are sorted and deduplicated in chronological order to obtain the final description of the crisis event development stages of occurrence, aggravation, weakening and elimination.

6. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 1 is characterized in that: In step S3, the computer is used to perform time series, coordination and causal relationship analysis based on the basic crisis clues, important crisis clues and development trends, strengthen the connection between crisis events, and classify crisis events by computer to achieve the generation of crisis event evolution context in different dimensions of event type and target type, which specifically includes sub-steps: First, computers are used to define the synergistic and causal relationships among crisis events in the field based on expert knowledge and experience; Then, based on the crisis event signs and crisis event development stage descriptions, the crisis event set is sorted in time, and the crisis event labels and crisis event relationships are classified based on the Bert pre-training model to form three types of crisis event labels: background, behavior, and status. The synergistic and causal relationships between crisis events are given, and finally the crisis events are constructed according to the six dimensions of background, behavior, status, goal, relationship, and signs to obtain a complete and rich crisis event evolution context.

7. The method for generating crisis event evolution context driven by domain expert knowledge according to claim 5 is characterized in that: The step of calculating the development status of the crisis event through spatiotemporal data specifically includes the following sub-steps: First, the type of crisis event is determined by computer. If it is a single-target event, the location of the crisis event is further determined. Then, the crisis event stage calculation is performed to calculate whether the distance and area meet the requirements of the rules. If the calculation result does not meet the requirements for the occurrence of the crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of the crisis event, the crisis event is continued to be calculated to determine whether it is aggravated, weakened, and resolved. If the crisis event is not resolved, the crisis event calculation is continued. If the crisis event is resolved, the above-mentioned occurrence, aggravation, weakening, and resolution analysis results are saved and the results are generated into a text description according to the format; If it is a group target event, the location of the crisis event is further determined, and then the time granularity of the spatiotemporal data of different targets is aligned, and the spatiotemporal data of different targets are classified according to intervals. Then the median of each interval of different targets is used as the data value after the alignment of their respective data granularity to achieve the consistency of the time scale of different targets. Then the crisis event stage calculation is performed to calculate whether the distance, area, and time between the group targets meet the requirements of the rules. If the calculation result does not meet the requirements for the occurrence of the crisis event, the crisis event occurrence calculation is continued. If the calculation result meets the requirements for the occurrence of the crisis event, the crisis event is continued to be calculated. Whether the crisis event is intensified, weakened, and eliminated, if the crisis event is not eliminated, the crisis event calculation is continued. If the crisis event is eliminated, the occurrence, intensification, weakening, and elimination analysis results are saved and the results are generated into a text description according to the format.

8. A device for generating crisis event evolution context driven by domain expert knowledge, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded by the processor, the method according to any one of claims 1 to 7 is executed.

Images