Earthquake event emergency response method

Abstract

This invention relates to the field of information processing, and more particularly to an earthquake event emergency response method, comprising: acquiring earthquake operational data; determining a preferred template generation method based on template correlation and report response coefficient, wherein the preferred template generation method is either reorganizing preferred templates based on regional matching averages or generating preferred templates based on information matching amplitudes; determining a report update method based on earthquake impact coefficients and public opinion information volume, wherein the report update method is either adjusting the report update frequency based on comprehensive evaluation values ​​or determining a report supplementation method based on the characteristic correlation coefficient of effective public opinion areas; the report supplementation method includes feature text supplementation and related text supplementation. This invention can improve the accuracy and timeliness of earthquake event response technical reports.

Description

Technical Field

[0001] This invention relates to the field of information processing, and more particularly to an emergency response method for earthquake events. Background Technology

[0002] When an earthquake occurs, a large amount of seismic sensing information is generated. However, processing this data presents numerous challenges, such as the poor matching between report templates and the actual earthquake situation, and the report content failing to fully meet the public's information needs. These problems not only exacerbate the difficulty of integrating disaster information but also reduce the practicality of earthquake event response technical reports. Therefore, how to update earthquake event response technical reports while ensuring the accuracy of the reported information, in order to improve the accuracy and timeliness of earthquake event response technical reports and thus meet the public's information needs, is a technical problem that urgently needs to be solved by those skilled in the art.

[0003] However, existing technologies generate earthquake event response technical reports based solely on earthquake operational data, failing to consider the problem of insufficient report content that cannot meet the public's information needs. Furthermore, they cannot adapt the report generation method to public needs, resulting in poor report usability. Summary of the Invention

[0004] To address this, the present invention provides an earthquake emergency response method to overcome the problems of existing technologies that fail to consider insufficient report content to meet public information needs and cannot adapt the report generation method to public needs, resulting in poor report usability.

[0005] To achieve the above objectives, the present invention provides an earthquake event emergency response method, comprising:

[0006] Acquire earthquake operational data;

[0007] The preferred template generation method is determined based on the template relevance and report response coefficient. The preferred template generation method is to reorganize the preferred template based on the regional matching mean or to generate the preferred template based on the information matching amplitude.

[0008] The report update method is determined based on the earthquake impact coefficient and the amount of public opinion information. The report update method is to adjust the report update frequency based on the comprehensive assessment value or to determine the report supplementation method based on the characteristic correlation coefficient of the effective public opinion area.

[0009] The report supplementation methods include feature text supplementation and related text supplementation.

[0010] Furthermore, the preferred template generation method is determined based on template relevance and report response coefficient, including:

[0011] If the template relevance is less than the preset template relevance or the report response coefficient is less than the preset report response coefficient, the preferred template generation method is to reorganize the preferred template based on the regional matching mean.

[0012] If the template relevance is greater than or equal to the preset template relevance and the report response coefficient is greater than or equal to the preset report response coefficient, the preferred template generation method is to generate the preferred template based on the information matching amplitude.

[0013] Furthermore, template reorganization is performed based on the regional matching mean, including:

[0014] The earthquake event response technical report template with the largest regional matching mean was used as the pre-reconstruction template, and the reconstruction method corresponding to each title to be reconstructed in the pre-reconstruction template was determined according to the replacement difficulty coefficient.

[0015] For a single title to be reorganized,

[0016] If the replacement difficulty coefficient is greater than or equal to the preset replacement difficulty coefficient, the recombination method is to replace the whole;

[0017] If the replacement difficulty coefficient is less than the preset replacement difficulty coefficient, the recombination method is partial recombination;

[0018] The title to be restructured is the title corresponding to the template keywords that do not appear in the earthquake business data in the pre-restructuring template.

[0019] Furthermore, in some reorganizations, the keyword replacement method corresponding to each template keyword to be replaced in the title to be reorganized is determined based on the interaction influence coefficient;

[0020] For a single keyword in the template to be replaced,

[0021] If the interaction impact coefficient is greater than or equal to the preset interaction impact coefficient, the keyword replacement method is template selection replacement;

[0022] If the interaction impact coefficient is less than the preset interaction impact coefficient, the keyword replacement method is data selection replacement;

[0023] The template keywords to be replaced are those that did not appear in the earthquake operational data in the pre-recombined template.

[0024] Furthermore, the report update method is determined based on the earthquake impact coefficient and the amount of public opinion information, including:

[0025] If the earthquake impact coefficient is less than the preset earthquake impact coefficient and the amount of public opinion information is less than the preset amount of public opinion information, the report update method is to increase the report update frequency based on the comprehensive assessment value;

[0026] If the earthquake impact coefficient is greater than or equal to the preset earthquake impact coefficient or the amount of public opinion information is greater than or equal to the preset amount of public opinion information, the report update method will be determined based on the characteristic correlation coefficient of the effective public opinion area to supplement the report.

[0027] The increase in the report update frequency is negatively correlated with the overall evaluation value.

[0028] Furthermore, the selection method for effective public opinion pages is determined based on the redundancy coefficient and the correlation impact threshold, including:

[0029] If the redundancy coefficient is less than the preset redundancy coefficient and the correlation impact threshold is less than the preset correlation impact threshold, the effective public opinion page is selected based on the historical fit coefficient and the risk assessment coefficient.

[0030] If the redundancy coefficient is greater than or equal to the preset redundancy coefficient or the correlation impact threshold is greater than or equal to the preset correlation impact threshold, the effective public opinion page is selected based on the data richness.

[0031] Furthermore, the initial effective public opinion area in the effective public opinion page is determined based on the effective reference value and the error coefficient, and the initial effective public opinion area is increased and adjusted according to the correlation coefficient;

[0032] The initial effective public opinion area after the adjustment is completed is recorded as the effective public opinion area;

[0033] The initial effective public opinion area is positively correlated with the effective reference value and the error coefficient;

[0034] The increase in the initial effective public opinion region is positively correlated with the correlation coefficient.

[0035] Furthermore, the report supplementation method is determined based on the characteristic correlation coefficient of the effective public opinion area, including:

[0036] If the feature correlation coefficient is greater than or equal to the preset feature correlation coefficient, the report will be supplemented with feature text.

[0037] If the feature correlation coefficient is less than the preset feature correlation coefficient, the report will be supplemented by related text.

[0038] Furthermore, the method for confirming the feature correlation coefficient includes:

[0039] If the proportion of identical keywords is greater than or equal to the preset proportion of identical keywords, the feature correlation coefficient is determined based on the proportion of identical keywords and the proportion of related keywords.

[0040] If the proportion of identical keywords is less than the preset proportion of identical keywords, the feature correlation coefficient is determined based on the frequency of identical keywords.

[0041] Furthermore, the earthquake impact coefficient is determined based on the response level and the earthquake disaster impact value;

[0042] The earthquake influence coefficient is negatively correlated with the response level, and positively correlated with the earthquake disaster impact value.

[0043] Compared with the prior art, the beneficial effects of the present invention are that, in the technical solution of the present invention, the matching degree of the earthquake event response technical report template is effectively reflected by the template correlation degree and the report response coefficient. Then, different preferred template generation methods are adaptively selected according to the template correlation degree and the report response coefficient, ensuring that the generated report content is both accurate and in line with actual needs, reducing information redundancy and misleading information, and improving the overall quality of earthquake event response technical reports.

[0044] Furthermore, this invention effectively reflects the urgency and public attention of earthquake events through the earthquake impact coefficient and the amount of public opinion information. Based on these factors, different report update methods are adaptively selected. This ensures that the determined report update method can promptly update the report content as the earthquake situation develops and public information needs change after the earthquake. This helps generate accurate and focused earthquake event response technical reports, enhances public trust in earthquake emergency response work, and creates favorable conditions for subsequent rescue and recovery efforts. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the earthquake emergency response method of the present invention;

[0046] Figure 2 This is a flowchart illustrating how the preferred template generation method is determined based on template relevance and report response coefficient in this invention.

[0047] Figure 3 This is a flowchart illustrating how the report update method is determined based on the earthquake impact coefficient and the amount of public opinion information in this invention.

[0048] Figure 4 A flowchart for determining the report supplementation method based on the characteristic correlation coefficient of the effective public opinion area. Detailed Implementation

[0049] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0050] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0051] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0052] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0053] Please see Figures 1 to 4 As shown, the present invention provides an earthquake event emergency response method, comprising:

[0054] Acquire earthquake operational data;

[0055] The preferred template generation method is determined based on the template relevance and report response coefficient. The preferred template generation method is to reorganize the preferred template based on the regional matching mean or to generate the preferred template based on the information matching amplitude.

[0056] The report update method is determined based on the earthquake impact coefficient and the amount of public opinion information. The report update method is to adjust the report update frequency based on the comprehensive assessment value or to determine the report supplementation method based on the characteristic correlation coefficient of the effective public opinion area.

[0057] The report supplementation methods include feature text supplementation and related text supplementation.

[0058] The application scenario of this invention is the automatic generation of earthquake event response technical reports when an earthquake occurs. The earthquake operational data is a text containing several monitoring information, including but not limited to magnitude, focal depth, building area in the earthquake-affected area, and number of people in the earthquake-affected area. It should be noted that the earthquake operational data is updated once at the end of a single monitoring cycle. The earthquake-affected area is a circular area with the intersection of the reference line and the ground as the center and a preset length as the radius. The reference line is a straight line passing through the focal point and perpendicular to the ground. The preset length is positively correlated with the magnitude. This is content that is easy for those skilled in the art to understand, and will not be elaborated further.

[0059] This invention includes a report template library containing several earthquake event response technical report templates. Each earthquake event response technical report template contains several titles, each title corresponding to several template keywords and placeholders for each template keyword. The titles include, but are not limited to, basic earthquake information and earthquake impact. The template keywords corresponding to the basic earthquake information include, but are not limited to, earthquake name and occurrence time. The template keywords corresponding to the earthquake impact include, but are not limited to, number of casualties and economic losses. The placeholders for each template keyword are []. This invention also includes a public opinion keyword library containing several earthquake-related keywords, including, but not limited to, magnitude, degree of damage, casualties, and rescue. This is content that is easily understood by those skilled in the art and will not be elaborated further.

[0060] This invention includes several historical records, each recording at least one earthquake event response technical report generation process, including report response coefficient, replacement difficulty coefficient, interaction impact coefficient, and earthquake impact coefficient. Each historical record also has a corresponding qualification mark, which indicates whether the generated earthquake event response technical report meets the user's requirements. The qualification mark can be manually recorded. It is understood that the user can determine whether the earthquake event response technical report generation process meets the requirements based on self-defined indicators. Self-defined indicators can be, but are not limited to, the frequency of error generation, which will not be elaborated here. The frequency of error generation refers to the presence of template keywords in the generated earthquake event response technical report that are not in the earthquake business data.

[0061] This invention sets target coefficients and relevant thresholds. The correspondence between the target coefficients and relevant thresholds is expressed through a weighting formula: Target Coefficient = Weight Coefficient × Relevant Threshold. Specifically, this invention uses the preset length, the increase in report update frequency, the initial effective public opinion area, the increase in the initial effective public opinion area, the feature correlation coefficient, and the earthquake disaster impact value as target coefficients. It uses magnitude, comprehensive assessment value, assessment threshold, correlation coefficient, frequency of similar events, number of deaths, and danger frequency as relevant thresholds. It is understood that all target coefficients have corresponding relevant thresholds. For example, the preset length and magnitude are positively correlated, and this positive correlation is expressed through a weighting formula. The value of the weighting coefficient can be determined based on the user's historical experience and the degree of influence of magnitude on the preset length. Furthermore, the value of the weighting coefficient can be optimized using a multilayer perceptron combined with the automatically generated historical records of earthquake event response reports from multiple earthquakes. The optimization of the weighting coefficient using a multilayer perceptron is easily understood by those skilled in the art and will not be elaborated upon here. The principle for determining the weighting coefficients corresponding to other target coefficients and relevant thresholds is the same and will not be elaborated upon here.

[0062] This invention establishes a continuous monitoring cycle after an earthquake occurs. At the end of each monitoring cycle, the earthquake impact coefficient and the amount of public opinion information are determined. The duration of the monitoring cycle can be set according to the user's needs. The greater the user's need for the accuracy of monitoring the earthquake impact coefficient and the amount of public opinion information, the shorter the monitoring cycle duration. One possible value for the monitoring cycle duration is 10 minutes.

[0063] Specifically, the preferred template generation method is determined based on template relevance and report response coefficient, including:

[0064] If the template relevance is less than the preset template relevance or the report response coefficient is less than the preset report response coefficient, the preferred template generation method is to reorganize the preferred template based on the regional matching mean.

[0065] If the template relevance is greater than or equal to the preset template relevance and the report response coefficient is greater than or equal to the preset report response coefficient, the preferred template generation method is to generate the preferred template based on the information matching amplitude.

[0066] Specifically, the template keywords that appear in the earthquake operational data in the technical report templates for each earthquake event response are recorded as the first keyword, and the template keywords that do not appear in the earthquake operational data are recorded as the second keyword;

[0067] The template relevance is the maximum value among the sub-relevances corresponding to each earthquake event response technical report template. The sub-relevance of a single earthquake event response technical report template = the number of first keywords in the earthquake event response technical report template / the total number of template keywords in the earthquake event response technical report template;

[0068] The report response coefficient is the average of the sub-response coefficients corresponding to each earthquake event response technical report template. The sub-response coefficient corresponding to a single earthquake event response technical report template is calculated as ln(the minimum value of the title completeness among all titles in the earthquake event response technical report template) × (the average of the emergence coefficients corresponding to each first keyword in the earthquake event response technical report template). The title completeness corresponding to a single title is calculated as (the number of first keywords among all template keywords corresponding to the title) / (the total number of template keywords corresponding to the title). The emergence coefficient corresponding to a single first keyword is the number of times the first keyword appears in the earthquake business data.

[0069] Users can determine the values ​​of preset template relevance and preset report response coefficient according to the actual application scenario. The higher the values ​​of preset template relevance and preset report response coefficient, the greater the user's need to reorganize the report file template according to the regional matching coefficient. One preset template relevance and preset report response coefficient value is provided. The preset template relevance is 70%. The historical records of generating preferred templates based on information matching amplitude are detected, and the average value of the report response coefficients corresponding to the historical records that can meet the user's needs is recorded as the preset report response coefficient.

[0070] The preferred template is generated based on the information matching amplitude, including: the earthquake event response technical report template with the largest information matching amplitude after deleting each first keyword and the placeholder corresponding to each first keyword is used as the preferred template; information matching amplitude = sub-response coefficient × sub-correlation degree.

[0071] Specifically, template recombination is performed based on the mean of regional matching, including:

[0072] The earthquake event response technical report template with the largest regional matching mean was used as the pre-reconstruction template, and the reconstruction method corresponding to each title to be reconstructed in the pre-reconstruction template was determined according to the replacement difficulty coefficient.

[0073] For a single title to be reorganized,

[0074] If the replacement difficulty coefficient is greater than or equal to the preset replacement difficulty coefficient, the recombination method is to replace the whole;

[0075] If the replacement difficulty coefficient is less than the preset replacement difficulty coefficient, the recombination method is partial recombination;

[0076] The title to be restructured is the title corresponding to the template keywords that do not appear in the earthquake business data in the pre-restructuring template.

[0077] The mean of regional matching is the average of the regional matching coefficients corresponding to each title in a single earthquake event response technical report template. The method for confirming the regional matching coefficient is as follows: for a single earthquake event response technical report template, the earthquake event response technical report template is recorded as the target template, and other earthquake event response technical report templates outside the target template are recorded as reference templates. For a single title in the target template, the title is recorded as the target title, and the earthquake event response technical report templates that have the target title are recorded as templates to be analyzed. The regional matching coefficient corresponding to the target title = the number of templates to be analyzed + the number of templates to be analyzed that do not have the first keyword corresponding to the target title.

[0078] The replacement difficulty coefficient for a single title to be restructured = the number of first keywords in the title to be restructured / the region matching coefficient of the title to be restructured;

[0079] In the overall replacement, for a single title to be restructured, the analysis title with the highest matching coefficient with the title to be restructured is selected for replacement. The analysis template containing the analysis title is recorded as the template to be replaced. The template to be replaced with the highest matching evaluation coefficient is used as the replacement template corresponding to the title to be restructured. The keywords and placeholders of the title to be restructured and each template corresponding to the title to be restructured are replaced with the keywords and placeholders of each template in the analysis title and the replacement template corresponding to the analysis title.

[0080] The matching evaluation coefficient of a single analysis title and a single template to be replaced = 1 / (similar reference value × reference value of the number of first keywords). The similarity reference value is the number of template keywords in the template to be replaced whose corresponding analysis title is the same as the template keywords in the pre-recombined template. The reference value of the number of first keywords is the number of first keywords in the template to be replaced whose corresponding analysis title is the number of first keywords in the template to be replaced.

[0081] The earthquake event response technical report template other than the pre-recombination template is designated as the analysis template. The titles in the pre-recombination template are designated as the first title, the titles in the analysis template are designated as the second title, and the second title that is different from the first title is designated as the analysis title.

[0082] For a single title to be restructured in the pre-restructuring template, the title to be restructured is designated as the target title to be restructured, and the other titles in the pre-restructuring template, excluding the target title to be restructured, are designated as reference titles.

[0083] For a single analysis title, the analysis title is recorded as the target analysis title, and the analysis template with the target analysis title is recorded as the target analysis template. The matching coefficient corresponding to the target analysis title = the average of the proportion of reference titles corresponding to each target analysis template + the average of the proportion of analysis corresponding to each target analysis template.

[0084] The percentage of reference titles corresponding to a single target analysis template = the number of reference titles in that target analysis template / the total number of titles in that target analysis template;

[0085] The percentage of analysis corresponding to a single target analysis template = the number of secondary keywords corresponding to the target analysis title in the target analysis template / the total number of template keywords corresponding to the target analysis title in the target analysis template;

[0086] The user can determine the preset replacement difficulty coefficient value based on the actual application scenario. The higher the preset replacement difficulty coefficient value, the greater the user's need for partial reconstruction. The system provides a preset replacement difficulty coefficient value, detects the historical records of partial reconstruction, and records the average replacement difficulty coefficient corresponding to the historical records that meet the user's needs as the preset replacement difficulty coefficient.

[0087] Specifically, in some reorganizations, the keyword replacement method for each template keyword to be replaced in the title to be reorganized is determined based on the interaction impact coefficient;

[0088] For a single keyword in the template to be replaced,

[0089] If the interaction impact coefficient is greater than or equal to the preset interaction impact coefficient, the keyword replacement method is template selection replacement;

[0090] If the interaction impact coefficient is less than the preset interaction impact coefficient, the keyword replacement method is data selection replacement;

[0091] The template keywords to be replaced are those that did not appear in the earthquake operational data in the pre-recombined template.

[0092] The method for determining the interaction impact coefficient is as follows: for a single keyword to be replaced, the title corresponding to the keyword to be replaced is recorded as the reference title, the analysis template with reference titles is recorded as the reference analysis template, and the maximum value of the reference coefficients corresponding to each reference analysis template is recorded as the interaction impact coefficient.

[0093] The second keywords in each reference analysis template that are different from those in the pre-recombination template are recorded as target words.

[0094] The reference coefficient for a single reference analysis template = (the number of target words in the reference analysis template / the number of template keywords corresponding to the reference titles in the reference analysis template) + (the number of template keywords in the reference analysis template that are the same as those in the pre-restructuring template / the number of template keywords in the pre-restructuring template).

[0095] In the template selection and replacement process, for a single keyword to be replaced, the target word with the highest keyword matching coefficient with that keyword is selected for replacement;

[0096] The keyword matching coefficient between a single target word and a single keyword to be replaced is the average of the sub-matching coefficients corresponding to each reference analysis template that contains the target word. The sub-matching coefficient corresponding to a single reference analysis template that contains the target word is the number of template keywords that are the same as those in the pre-reorganization template.

[0097] In the data selection and replacement process, for a single keyword to be replaced, the data keyword that has been used most frequently in combination with that keyword and has not been replaced is replaced;

[0098] The data keywords that have not been replaced are those that have not been replaced with the keywords to be replaced. The data keywords are those that exist in the earthquake business data. Each keyword in the earthquake business data can be identified through NLP technology. This is something that is easy for those skilled in the art to understand, and will not be elaborated on in detail.

[0099] The number of times a single keyword to be replaced is used in combination with a single data keyword is the number of times that each associated keyword corresponding to the keyword to be replaced and the data keyword appear simultaneously in the same data paragraph, where the data paragraph is a single paragraph of earthquake business data;

[0100] Each associated keyword corresponding to a single keyword to be replaced is a template keyword in the pre-recombined template that is the same as the title corresponding to the keyword to be replaced.

[0101] The user can determine the value of the preset interaction impact coefficient according to the actual application scenario. The larger the value of the preset interaction impact coefficient, the greater the user's need for data selection and replacement. A preset interaction impact coefficient value is provided, and the historical records of template selection and replacement are detected. The average value of the interaction impact coefficients corresponding to the historical records that meet the user's needs is recorded as the preset interaction impact coefficient.

[0102] The template generated through the preferred template generation method is referred to as the preferred template. It should be noted that users can supplement the placeholders in the preferred template themselves based on the earthquake business data obtained at the end of the first monitoring period after the earthquake to generate an earthquake event response technical report. This is something that is easy for those skilled in the art to understand, and will not be elaborated on in detail.

[0103] Specifically, the report update method is determined based on the earthquake impact coefficient and the amount of public opinion information, including:

[0104] If the earthquake impact coefficient is less than the preset earthquake impact coefficient and the amount of public opinion information is less than the preset amount of public opinion information, the report update method is to increase the report update frequency based on the comprehensive assessment value;

[0105] If the earthquake impact coefficient is greater than or equal to the preset earthquake impact coefficient or the amount of public opinion information is greater than or equal to the preset amount of public opinion information, the report update method will be determined based on the characteristic correlation coefficient of the effective public opinion area to supplement the report.

[0106] The increase in the report update frequency is negatively correlated with the overall evaluation value.

[0107] Among them, the amount of public opinion information refers to the number of public opinion pages when the technical report on earthquake response is updated; web pages containing public opinion keywords are recorded as public opinion pages. Public opinion pages are obtained through automated web crawling technology, which is a common technical method in this field and will not be elaborated on in detail.

[0108] Users can determine the preset values ​​for the earthquake impact coefficient and the preset public opinion information volume based on the actual application scenario. The greater the user's demand for improving the timeliness of earthquake event response technical reports, the smaller the preset values ​​for the earthquake impact coefficient and the preset public opinion information volume will be. A preset earthquake impact coefficient and preset public opinion information volume value is provided. The historical records of users increasing or decreasing the report update frequency based on the comprehensive evaluation value are detected. The average value of the earthquake impact coefficient corresponding to the historical records that meet the user's needs is recorded as the preset earthquake impact coefficient, and the average value of the public opinion information volume corresponding to the historical records that meet the user's needs is recorded as the preset public opinion information volume.

[0109] The comprehensive assessment value = earthquake impact coefficient / preset earthquake impact coefficient + public opinion information volume / preset public opinion information volume. The report update frequency is the time interval between the generation of earthquake event response technical reports. In this invention, the initial report update frequency is 30 minutes.

[0110] Specifically, the selection method for effective public opinion pages is determined based on redundancy coefficients and correlation impact thresholds, including:

[0111] If the redundancy coefficient is less than the preset redundancy coefficient and the correlation impact threshold is less than the preset correlation impact threshold, the effective public opinion page is selected based on the historical fit coefficient and the risk assessment coefficient.

[0112] If the redundancy coefficient is greater than or equal to the preset redundancy coefficient or the correlation impact threshold is greater than or equal to the preset correlation impact threshold, the effective public opinion page is selected based on the data richness.

[0113] Among them, the redundancy coefficient = 1 - (number of non-redundant public opinion pages / total number of public opinion pages). The method for confirming non-redundant public opinion pages is to record public opinion pages with a page richness greater than the preset page richness and a page timeliness coefficient greater than the preset page timeliness coefficient as non-redundant public opinion pages.

[0114] Page richness = number of links in the public opinion page + number of public opinion keywords in the public opinion page; Page timeliness coefficient is the time between the time the public opinion page was published and the time the earthquake occurred;

[0115] Users can determine the values ​​of preset page richness and preset page timeliness coefficient according to the actual application scenario. The higher the user's selection precision of effective public opinion pages, the higher the values ​​of preset page richness and preset page timeliness coefficient. A preset page richness and preset page timeliness coefficient value is provided. The historical records of effective public opinion page selection methods are detected based on redundancy coefficient and correlation impact threshold. The average page richness corresponding to the historical records that can meet the user's needs is recorded as the preset page richness, and the average page timeliness coefficient corresponding to the historical records that can meet the user's needs is recorded as the preset page timeliness coefficient.

[0116] The correlation impact threshold is the average of the sub-correlation coefficients corresponding to each non-redundant public opinion page. The sub-correlation coefficient corresponding to a single non-redundant public opinion page is the number of public opinion keywords that are the same in the non-redundant public opinion page and other non-redundant public opinion pages.

[0117] Users can determine the values ​​of the preset redundancy coefficient and the preset correlation impact threshold according to the actual application scenario. The smaller the values ​​of the preset redundancy coefficient and the preset correlation impact threshold, the greater the user's demand for selecting effective public opinion pages based on data richness. The system provides a set of preset redundancy coefficient and preset correlation impact threshold values, detects the historical records of selecting effective public opinion pages based on data richness, and records the average value of the redundancy coefficient that meets the user's needs as the preset redundancy coefficient and the average value of the correlation impact threshold that meets the user's needs as the preset correlation impact threshold.

[0118] When selecting effective public opinion pages based on historical fit coefficient and risk assessment coefficient, non-redundant public opinion pages with a historical fit coefficient greater than the preset historical fit coefficient or risk assessment coefficient greater than the preset risk assessment coefficient are selected as effective public opinion pages.

[0119] When selecting valid public opinion pages based on data richness, non-redundant public opinion pages with data richness greater than the preset data richness are selected as valid public opinion pages.

[0120] The historical fit coefficient is determined as follows: for a single public opinion page, the public opinion page is recorded as the target public opinion page, the public opinion keywords appearing in the target public opinion page are recorded as target keywords, and the effective public opinion pages selected from the historical records that can meet the user's needs are detected and recorded as historical public opinion pages. The historical fit coefficient corresponding to the target public opinion page = (number of historical public opinion pages with target keywords) / (total number of historical public opinion pages).

[0121] The risk assessment coefficient for a single public opinion page = the number of views of the public opinion page at the end of the current monitoring period + the number of comments on the public opinion page at the end of the current monitoring period; the data richness for a single public opinion page = 1 / (1 + the sub-association coefficient corresponding to the public opinion page);

[0122] Users can determine the preset values ​​for historical consistency coefficient, preset risk assessment coefficient, and preset data richness based on their actual application scenarios. The greater the user's demand for the effectiveness of selected effective public opinion pages, the higher the values ​​of these preset values. One preset set of values ​​is provided: a preset historical consistency coefficient of 50%. The system detects historical records of effective public opinion pages selected based on the historical consistency coefficient and risk assessment coefficient, and records the average historical consistency coefficient of those historical records that meet the user's needs as the preset historical consistency coefficient. Similarly, the system detects historical records of effective public opinion pages selected based on data richness, and records the average data richness of those historical records that meet the user's needs as the preset data richness.

[0123] Specifically, the initial effective public opinion area in the effective public opinion page is determined based on the effective reference value and the error coefficient, and the initial effective public opinion area is increased or adjusted according to the correlation coefficient.

[0124] The initial effective public opinion area after the adjustment is completed is recorded as the effective public opinion area;

[0125] The initial effective public opinion area is positively correlated with the effective reference value and the error coefficient;

[0126] The increase in the initial effective public opinion region is positively correlated with the correlation coefficient.

[0127] Among them, the effective reference value is the total number of public opinion keywords in a single effective public opinion page. The method for determining the error coefficient is as follows: for a single effective public opinion page, each public opinion keyword in the effective public opinion page is recorded as the target public opinion keyword. The error coefficient corresponding to the effective public opinion page is the area of ​​the smallest rectangle that can contain each target public opinion keyword / the effective reference value corresponding to the effective public opinion page.

[0128] The initial effective public opinion area is the number of characters selected from the initial position in the reading order from front to back on a single effective public opinion page; the initial position is the first character of the sentence containing the first public opinion keyword that appears on a single effective public opinion page in the reading order from front to back;

[0129] The initial effective public opinion area is positively correlated with the evaluation threshold. Evaluation threshold = effective reference value / average effective reference value corresponding to the historical records that can meet user needs + error coefficient / average error coefficient corresponding to the historical records that can meet user needs;

[0130] Keywords within the initial valid public opinion area are recorded as analyzed public opinion keywords, while keywords outside the initial valid public opinion area are recorded as public opinion keywords to be analyzed. The correlation coefficient = (number of public opinion keywords to be analyzed / number of analyzed public opinion keywords) + (number of historical public opinion pages containing the keywords to be analyzed / number of historical public opinion pages containing the keywords to be analyzed).

[0131] The increment value of the initial effective public opinion area is the number of characters selected after the last character in the initial effective public opinion area according to the reading order from front to back.

[0132] Specifically, the report supplementation method is determined based on the characteristic correlation coefficient of the effective public opinion area, including:

[0133] If the feature correlation coefficient is greater than or equal to the preset feature correlation coefficient, the report will be supplemented with feature text.

[0134] If the feature correlation coefficient is less than the preset feature correlation coefficient, the report will be supplemented by related text.

[0135] The user can determine the value of the preset feature correlation coefficient according to the actual application scenario. The larger the value of the preset feature correlation coefficient, the greater the user's need for supplementing associated text. The system provides a preset feature correlation coefficient value, detects the historical records of feature text supplementation, and records the average value of the feature correlation coefficients corresponding to the historical records that meet the user's needs as the preset feature correlation coefficient.

[0136] In the feature text supplementation, the same keywords that appear more frequently than the preset frequency percentage are used as supplementary keywords;

[0137] In the supplementary related text, the detection and analysis scope includes the historical records of each differential keyword. The historical records that can meet the user's needs are recorded as the analysis history records. The supplementary keywords corresponding to each analysis history record are detected and recorded as supplementary keywords to be selected. The supplementary keywords to be selected with a supplementary frequency ratio greater than the preset supplementary frequency ratio are used as supplementary keywords. The analysis scope is each effective public opinion area corresponding to a single historical record.

[0138] The method for confirming the frequency percentage is as follows: for a single identical keyword, the identical keyword is recorded as the target identical keyword, and the frequency percentage of the target identical keyword is equal to the number of effective public opinion areas where the target identical keyword appears / the number of effective public opinion areas.

[0139] The percentage of times a single candidate keyword is added = the number of analysis history records where that candidate keyword is used as a keyword / the total number of analysis history records;

[0140] Users can determine the preset occurrence percentage and preset supplementation percentage based on the actual application scenario. The higher the user's need to improve the relevance of the supplementary content, the larger the preset occurrence percentage and preset supplementation percentage will be. One preset occurrence percentage and preset supplementation percentage are provided: the preset occurrence percentage is 70% and the preset supplementation percentage is 75%.

[0141] It should be noted that in the abnormal earthquake emergency response, the update frequency of the earthquake event response technical report is 30 minutes. When updating the earthquake event response technical report, users need to supplement the relevant content corresponding to each supplementary keyword at the current update time. The relevant content includes several texts containing supplementary keywords. There is no limit to the specific number of supplementary keywords. Users can supplement them according to their actual needs.

[0142] Specifically, the method for confirming the feature correlation coefficient includes:

[0143] If the proportion of identical keywords is greater than or equal to the preset proportion of identical keywords, the feature correlation coefficient is determined based on the proportion of identical keywords and the proportion of related keywords.

[0144] If the proportion of identical keywords is less than the preset proportion of identical keywords, the feature correlation coefficient is determined based on the frequency of identical keywords.

[0145] Among them, the public opinion keywords appearing in the effective public opinion area corresponding to the selected effective public opinion page are recorded as the first public opinion keyword, the first public opinion keyword that is the same as the template keyword in the earthquake event response technical report is recorded as the same keyword, and the first public opinion keyword that is different from the template keyword corresponding to the earthquake event response technical report is recorded as the different keyword. The percentage of the same keyword = the number of the same keyword / the number of template keywords in the earthquake event response technical report.

[0146] The user can determine the preset value of the same keyword percentage based on the actual application scenario. The larger the preset value of the same keyword percentage, the greater the user's need to determine the feature relevance coefficient based on the matching coefficient. One preset value of the same keyword percentage is provided, with a preset same keyword percentage of 60%.

[0147] If the proportion of identical keywords is greater than or equal to the preset proportion of identical keywords, then the feature correlation coefficient = the proportion of identical keywords + the proportion of related keywords;

[0148] If the proportion of identical keywords is less than the preset proportion of identical keywords, then the feature correlation coefficient and the frequency of identical keywords are positively correlated.

[0149] Relevance ratio = Number of differential keywords appearing in earthquake business data / Total number of differential keywords;

[0150] The frequency of occurrence is the average of the frequency of occurrence of each sub-frequency of the different keywords. The frequency of occurrence of occurrence of each sub-frequency of a single different keyword is the average of the frequency reference values ​​of the different keyword and each identical keyword. The frequency reference value of a single different keyword and a single identical keyword is the number of valid public opinion pages that simultaneously appear for the different keyword and the identical keyword.

[0151] Specifically, the earthquake impact coefficient is determined based on the response level and the earthquake disaster impact value;

[0152] The earthquake influence coefficient is negatively correlated with the response level, and positively correlated with the earthquake disaster impact value.

[0153] The response levels include Level 1, Level 2, Level 3, and Level 4. Earthquakes with a magnitude of 4.9 or less have a response level of Level 4; earthquakes with a magnitude greater than 4.9 and less than or equal to 5.9 have a response level of Level 3; earthquakes with a magnitude greater than 5.9 and less than or equal to 6.9 have a response level of Level 2; and earthquakes with a magnitude greater than 6.9 have a response level of Level 1.

[0154] Methods for confirming the impact value of earthquake disasters include:

[0155] If the building area ratio of the earthquake-affected area is greater than or equal to the preset building area ratio, the earthquake disaster impact value is determined based on the number of deaths, where the earthquake disaster impact value and the number of deaths are positively correlated.

[0156] If the building area ratio in the earthquake-affected area is less than the preset building area ratio, the earthquake disaster impact value is determined based on the hazard frequency, where the earthquake disaster impact value and the hazard frequency are positively correlated.

[0157] The method for determining the danger frequency is to record historical records with the same magnitude as the current earthquake and with the same building area in the earthquake-affected area as the current earthquake-affected area as reference historical records, and record the number of reference historical records with casualties as danger frequencies.

[0158] The percentage of building area in the earthquake-affected area = the building area in the earthquake-affected area / the area of ​​the earthquake-affected area;

[0159] The user can determine the preset building area percentage based on the actual application scenario. The larger the preset building area percentage, the greater the user's need to determine the impact value of earthquake disasters based on the frequency of danger. One preset building area percentage is provided, which is 30%.

[0160] Earthquake impact coefficient = (1 - response level / average response level of each historical record) + earthquake disaster impact value / average earthquake disaster impact value of each historical record.

[0161] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0162] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An emergency response method for earthquake events, characterized in that, include: Acquire earthquake operational data; The preferred template generation method is determined based on the template relevance and report response coefficient. The preferred template generation method is to reorganize the preferred template based on the regional matching mean or to generate the preferred template based on the information matching amplitude. The report update method is determined based on the earthquake impact coefficient and the amount of public opinion information. The report update method is to adjust the report update frequency based on the comprehensive assessment value or to determine the report supplementation method based on the characteristic correlation coefficient of the effective public opinion area. The report supplementation methods include feature text supplementation and related text supplementation; The preferred template generation method is determined based on template relevance and report response coefficient, including: If the template relevance is less than the preset template relevance or the report response coefficient is less than the preset report response coefficient, the preferred template generation method is to reorganize the preferred template based on the regional matching mean. If the template relevance is greater than or equal to the preset template relevance and the report response coefficient is greater than or equal to the preset report response coefficient, the preferred template generation method is to generate the preferred template based on the information matching amplitude. The report update method is determined based on the earthquake impact coefficient and the amount of public opinion information, including: If the earthquake impact coefficient is less than the preset earthquake impact coefficient and the amount of public opinion information is less than the preset amount of public opinion information, the report update method is to increase the report update frequency based on the comprehensive assessment value; If the earthquake impact coefficient is greater than or equal to the preset earthquake impact coefficient or the amount of public opinion information is greater than or equal to the preset amount of public opinion information, the report update method will be determined based on the characteristic correlation coefficient of the effective public opinion area to supplement the report. The increase in the report update frequency is negatively correlated with the overall evaluation value; The selection method for effective public opinion pages is determined based on redundancy coefficients and correlation impact thresholds, including: If the redundancy coefficient is less than the preset redundancy coefficient and the correlation impact threshold is less than the preset correlation impact threshold, the effective public opinion page is selected based on the historical fit coefficient and the risk assessment coefficient. If the redundancy coefficient is greater than or equal to the preset redundancy coefficient or the correlation impact threshold is greater than or equal to the preset correlation impact threshold, the effective public opinion page is selected based on the data richness. Redundancy coefficient = 1 - (Number of non-redundant public opinion pages / Total number of public opinion pages). Non-redundant public opinion pages are those with a page richness greater than the preset page richness and a page timeliness coefficient greater than the preset page timeliness coefficient. The initial valid public opinion area in the valid public opinion page is determined based on the valid reference value and the error coefficient, and the initial valid public opinion area is increased or adjusted according to the correlation coefficient. The initial effective public opinion area after the adjustment is completed is recorded as the effective public opinion area; The initial effective public opinion area is positively correlated with the effective reference value and the error coefficient; The increase in the initial effective public opinion region is positively correlated with the correlation coefficient. The method for confirming the feature correlation coefficient includes: If the proportion of identical keywords is greater than or equal to the preset proportion of identical keywords, the feature correlation coefficient is determined based on the proportion of identical keywords and the proportion of related keywords. If the proportion of identical keywords is less than the preset proportion of identical keywords, the feature correlation coefficient is determined based on the frequency of identical keywords. The template relevance is the maximum value among the sub-relevances corresponding to each earthquake event response technical report template. The sub-relevance of a single earthquake event response technical report template = the number of first keywords in the earthquake event response technical report template / the total number of template keywords in the earthquake event response technical report template; The report response coefficient is the average of the sub-response coefficients corresponding to each earthquake event response technical report template. The sub-response coefficient corresponding to a single earthquake event response technical report template is calculated as ln(minimum value of title completeness among all titles in the earthquake event response technical report template) × (average of the emergence coefficients corresponding to each first keyword in the earthquake event response technical report template). The title completeness corresponding to a single title is calculated as (number of first keywords among all template keywords corresponding to the title) / (total number of template keywords corresponding to the title). The emergence coefficient corresponding to a single first keyword is the number of times the first keyword appears in the earthquake business data. The regional matching mean is the average of the regional matching coefficients corresponding to each title in a single earthquake event response technical report template. The method for confirming the regional matching coefficient is as follows: for a single earthquake event response technical report template, the earthquake event response technical report template is recorded as the target template, and other earthquake event response technical report templates outside the target template are recorded as reference templates. For a single title in the target template, the title is recorded as the target title, and the earthquake event response technical report templates that have the target title are recorded as templates to be analyzed. The regional matching coefficient corresponding to the target title = the number of templates to be analyzed + the number of templates to be analyzed that do not have the first keyword corresponding to the target title. Information matching amplitude = sub-response coefficient × sub-association degree; Comprehensive assessment value = earthquake impact coefficient / preset earthquake impact coefficient + public opinion information volume / preset public opinion information volume; The historical fit coefficient is determined as follows: for a single public opinion page, the public opinion page is recorded as the target public opinion page, the public opinion keywords appearing in the target public opinion page are recorded as target keywords, and the effective public opinion pages selected from the historical records that can meet the user's needs are detected and recorded as historical public opinion pages. The historical fit coefficient corresponding to the target public opinion page = (number of historical public opinion pages with target keywords) / (total number of historical public opinion pages). The risk assessment coefficient for a single public opinion page = the number of times the public opinion page was viewed at the end of the current monitoring period + the number of comments on the public opinion page at the end of the current monitoring period; The data richness of a single public opinion page = 1 / (1 + sub-association coefficient of the public opinion page); The correlation impact threshold is the average of the sub-correlation coefficients corresponding to each non-redundant public opinion page. The sub-correlation coefficient corresponding to a single non-redundant public opinion page is the number of public opinion keywords that are the same in the non-redundant public opinion page and other non-redundant public opinion pages. Page richness = number of links on the public opinion page + number of public opinion keywords on the public opinion page; The page timeliness coefficient is the time interval between the time the public opinion page was published and the time the earthquake occurred; The effective reference value is the total number of public opinion keywords in a single effective public opinion page. The method for determining the error coefficient is as follows: for a single effective public opinion page, each public opinion keyword in the effective public opinion page is recorded as the target public opinion keyword. The error coefficient corresponding to the effective public opinion page = the area of ​​the smallest rectangle that can contain each target public opinion keyword / the effective reference value corresponding to the effective public opinion page. Correlation coefficient = (Number of keywords to be analyzed / Number of keywords already analyzed) + (Number of historical public opinion pages containing the keyword to be analyzed / Number of historical public opinion pages containing the keyword already analyzed); Percentage of identical keywords = Number of identical keywords / Number of template keywords in the earthquake event response technical report; Relevance ratio = Number of differential keywords appearing in earthquake business data / Total number of differential keywords; The frequency of occurrence is the average of the frequency of occurrence of each sub-frequency of the different keywords. The frequency of occurrence of occurrence of each sub-frequency of a single different keyword is the average of the frequency reference values ​​of the different keyword and each identical keyword. The frequency reference value of a single different keyword and a single identical keyword is the number of valid public opinion pages that simultaneously appear for the different keyword and the identical keyword.

2. The earthquake emergency response method according to claim 1, characterized in that, Template reorganization is performed based on the mean of regional matching, including: The earthquake event response technical report template with the largest regional matching mean was used as the pre-reconstruction template, and the reconstruction method corresponding to each title to be reconstructed in the pre-reconstruction template was determined according to the replacement difficulty coefficient. For a single title to be reorganized, If the replacement difficulty coefficient is greater than or equal to the preset replacement difficulty coefficient, the recombination method is to replace the whole; If the replacement difficulty coefficient is less than the preset replacement difficulty coefficient, the recombination method is partial recombination; The title to be restructured is the title corresponding to the template keywords that do not appear in the earthquake business data in the pre-restructuring template.

3. The earthquake emergency response method according to claim 2, characterized in that, In some reorganizations, the keyword replacement method for each template keyword to be replaced in the title to be reorganized is determined based on the interaction impact coefficient; For a single keyword in the template to be replaced, If the interaction impact coefficient is greater than or equal to the preset interaction impact coefficient, the keyword replacement method is template selection replacement; If the interaction impact coefficient is less than the preset interaction impact coefficient, the keyword replacement method is data selection replacement; The template keywords to be replaced are those that did not appear in the earthquake operational data in the pre-recombined template; The interaction impact coefficient is determined as follows: for a single keyword to be replaced, the title corresponding to the keyword to be replaced is recorded as the reference title, the analysis template with reference titles is recorded as the reference analysis template, and the maximum value of the reference coefficients corresponding to each reference analysis template is recorded as the interaction impact coefficient. The reference coefficient corresponding to a single reference analysis template = (the number of target words in the reference analysis template / the number of template keywords corresponding to the reference questions in the reference analysis template) + (the number of template keywords in the reference analysis template that are the same as those in the pre-recombination template / the number of template keywords in the pre-recombination template).

4. The earthquake emergency response method according to claim 1, characterized in that, The report supplementation method is determined based on the characteristic correlation coefficient of the effective public opinion area, including: If the feature correlation coefficient is greater than or equal to the preset feature correlation coefficient, the report will be supplemented with feature text. If the feature correlation coefficient is less than the preset feature correlation coefficient, the report will be supplemented by related text.

5. The earthquake emergency response method according to claim 1, characterized in that, The earthquake impact coefficient is determined based on the response level and the earthquake disaster impact value; The earthquake influence coefficient is negatively correlated with the response level, and positively correlated with the earthquake disaster impact value.

Images