A flood mortality estimation method based on population displacement and purchasing power parity per capita

Abstract

The application discloses a flood mortality rate estimation method based on population displacement and purchasing power parity per capita, and the method comprises the following steps: firstly, collecting historical data of flood events, which at least covers the number of population displacement, the purchasing power parity per capita and the actual number of deaths caused by the flood; checking the number of population displacement and the purchasing power parity per capita; then, according to a preset standardization method, normalizing the death data by using the number of population displacement, so as to obtain a death rate index which eliminates the influence of the event scale; next, taking the number of population displacement and the purchasing power parity per capita as characteristic variables and taking the calculated death rate as a target variable, a prediction model is constructed and trained; finally, inputting the corresponding data of a flood event to be evaluated into the trained model, and outputting the estimated value of the death rate of the event. The application can realize cross-event and cross-scale flood mortality rate estimation. The application can be widely applied in the field of risk assessment.

Description

Technical Field

[0001] This invention relates to the field of risk assessment, and more particularly to a method for estimating flood fatality rates based on population displacement and per capita purchasing power parity. Background Technology

[0002] Flood events often cause large-scale casualties, and rapid and accurate estimation of flood fatality rates is crucial for post-disaster statistics. Current technologies often employ statistical regression or empirical coefficients to directly fit the number of fatalities. Statistical regression models directly use the number of deaths / missing persons in historical flood events as the dependent variable, and flood intensity indicators (such as peak flow, inundation depth, and recurrence interval) and regional macroeconomic indicators (such as population density and GDP) as independent variables to establish a multiple regression equation. Empirical coefficient methods use historical disaster statistics to artificially set a fixed fatality rate coefficient (such as the number of deaths per inundated square kilometer or per 10,000 inundated people) and apply it to the affected population or inundation area of ​​a new event to quickly estimate the number of deaths. However, the number of fatalities is significantly affected by the scale of the event and the size of the exposed population, making direct fitting difficult to achieve comparability across events and years. Furthermore, differences in emergency response capabilities and infrastructure at different levels of economic development significantly impact fatality risk. Due to the lack of a unified, standardized, and interpretable estimation framework for fatality rates, existing methods have shortcomings in cross-scale generalization, extreme sample stability, and engineering deployability. Summary of the Invention

[0003] In view of this, in order to address the technical problem that existing flood mortality estimation methods rely on a single metric, resulting in insufficient generalization, this invention proposes a flood mortality estimation method based on population displacement and per capita purchasing power parity. This method includes the following steps: First, historical data on flood events is collected, including at least population displacement, purchasing power parity (PPP), and the actual number of deaths caused by the floods. The population displacement and PPP data are then validated. Next, using a pre-defined standardization method, the mortality data is normalized using the population displacement figures to obtain a mortality rate indicator that eliminates the influence of event scale. Then, using population displacement and PPP as feature variables and the calculated mortality rate as the target variable, a prediction model is constructed and trained. Finally, the corresponding data for the flood event to be assessed is input into the trained model, and the estimated mortality rate for that event is output.

[0004] Furthermore, a Gaussian process regression model was selected as an auxiliary model, and the uncertainty measure corresponding to the mortality rate estimate was output.

[0005] Based on the above scheme, this invention provides a flood lethality estimation method based on population displacement and per capita purchasing power parity. By standardizing the lethality rate, the lethality data is decoupled from the displacement scale, improving the consistency of cross-event comparisons. By using logarithmic space power law modeling, a parameter interpretable lethality estimation function is realized, which is convenient for engineering deployment and reproduction. An optional Gaussian process regression auxiliary model can be introduced to improve the nonlinear fitting ability while maintaining the interpretability of the main model and supporting uncertainty output, providing a basis for risk classification and decision-making. Attached Figure Description

[0006] Figure 1 This is a flowchart of the steps of a flood mortality rate estimation method based on population displacement and per capita purchasing power parity of the present invention. Figure 2 This is a structural block diagram of a flood mortality estimation system based on population displacement and per capita purchasing power parity according to the present invention. Figure 3 A comparison chart showing the back-substitution estimation of the mortality rate of Model A (power-law regression) on the training set with the actual mortality rate; Figure 4 A comparison of the estimated case fatality rate and the actual mortality rate of Model A (power-law regression) under the leave-one-out (LOO) prediction condition; Figure 5 This is a scatter plot comparing the consistency between the actual mortality rate and the predicted mortality rate of Model A (power law regression) under the leave-one-out (LOO) prediction condition. The dashed line is the ideal consistency line (1:1 line). Figure 6 A comparison chart showing the back-substitution estimation of the mortality rate of Model B (Gaussian Process Regression GPR) on the training set and the actual mortality rate; Figure 7 A comparison of the estimated case fatality rate and the actual case fatality rate of Model B (Gaussian process regression GPR) under leave-one-out (LOO) prediction conditions; Figure 8 This is a scatter plot comparing the consistency between the actual mortality rate and the predicted mortality rate of Model B (Gaussian process regression GPR) under the leave-one-out (LOO) prediction condition. The dashed line represents the ideal consistency line (1:1 line). Detailed Implementation

[0007] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0008] It should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0009] It should be understood that the terms "system," "apparatus," "unit," and / or "module" used in this application are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0010] As indicated in this application, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. An element defined by the phrase "comprising an..." does not exclude the presence of other identical elements in the process, method, product, or apparatus that includes the element.

[0011] In the description of the embodiments of this application, "a plurality of" refers to two or more. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0012] Furthermore, flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Additionally, other operations can be added to these processes, or one or more steps can be removed from them.

[0013] Reference Figure 1 This is a flowchart illustrating an optional example of the flood mortality rate estimation method based on population displacement and per capita purchasing power parity proposed in this invention. The method can be applied to computer equipment, and the estimation method proposed in this embodiment may include, but is not limited to, the following steps: Step S1: Obtain flood event-related data, including population displacement, per capita purchasing power parity, and flood-related death data; Step S2: Conduct an effectiveness test on the number of people displaced and the per capita purchasing power parity; Step S3: Calculate the mortality rate based on population displacement, flood fatality data, and preset standardized rules; Step S4: Generate a prediction model based on population displacement, per capita purchasing power parity, and mortality rate. Step S5: Input the data to be estimated into the prediction model and output the estimated mortality rate.

[0014] In some feasible embodiments, step S2 specifically includes: The validity of the population displacement data D and the flood-related death data P is tested. When D≤0 or P≤0, the data is replaced by a very small positive number ε or marked as an outlier. ε satisfies 0<ε≤10. -6 .

[0015] In some feasible embodiments, step S3, the formula for calculating the lethality rate is as follows: in, Indicates the mortality rate; This represents data on deaths caused by floods; Indicates the number of people displaced; This represents the standardization factor, which can be taken as 10. 5 Let R be expressed as "the number of deaths per 100,000 population displacement".

[0016] In some feasible embodiments, step S4 specifically includes: Construct the logarithmic space eigenvectors X=[ln(D),ln(P)], and construct the logarithmic space labels Y=ln(R); A mortality rate prediction model is trained based on feature vector X and label Y. The mortality rate prediction model includes at least a power-law regression model, the expression of which is: in, , , This represents the corresponding model parameters.

[0017] In some feasible embodiments, step S5 specifically includes: During the prediction phase, the inputs are D and P, which are to be estimated, and the output is the estimated mortality rate. ,in: Furthermore, if an estimated number of deaths is required, it can be converted using the following formula: In some feasible embodiments, step S4 may also replace log-space modeling with a Gaussian process regression model: in Obtained through Gaussian process regression training; in the prediction phase, except for the output In addition, it outputs uncertainty measures (such as variance or confidence intervals) for risk classification or to trigger manual review.

[0018] In some feasible embodiments, it also includes: Simultaneously train a power-law regression model and a Gaussian process regression model, and use leave-one-out cross-validation or K-fold cross-validation to evaluate the error. Select the final output model based on the minimum error criterion, or fuse the outputs of the two models according to preset weights.

[0019] Based on the overall process described above, this invention also provides relevant simulation data. In the embodiments, the actual flood fatality rate R (number of deaths per 100,000 displaced people) is calculated based on historical annual sample data, and model A (power-law regression) and model B (Gaussian process regression GPR) are used to estimate the fatality rate, respectively. To evaluate the model's back-substitution fitting effect on the training set, [the following is a process description] is generated. Figure 3 and Figure 6 To evaluate the model's generalization and predictive ability, leave-one-out (LOO) method was used to predict the annual samples one year at a time, generating... Figure 4 , Figure 7 and consistency scatter comparison Figure 5 , Figure 8 The dashed line represents the ideal congruence line (1:1 line). Simultaneously, the error indices RMSE and coefficient of determination R² are calculated under both back-substitution and leave-one-out prediction conditions to quantify the degree of deviation between the estimated results and the actual mortality rate.

[0020] like Figure 2 As shown, a flood mortality estimation system based on population displacement and per capita purchasing power parity includes: The data acquisition module is used to execute step S1; The data verification and processing module is used to execute step S2; The lethality standardization module is used to perform step S3; The model building module is used to execute step S4; The prediction output module is used to execute step S5.

[0021] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0022] A flood mortality rate estimation device based on population displacement and per capita purchasing power parity: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described above.

[0023] The content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0024] A storage medium storing processor-executable instructions, which, when executed by a processor, are used to implement a flood mortality estimation method based on population displacement and per capita purchasing power parity as described above.

[0025] The content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0026] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for estimating flood fatality rate based on population displacement and per capita purchasing power parity, characterized in that, Includes the following steps: Acquire flood event-related data, including population displacement, per capita purchasing power parity, and flood-related death data; The mortality rate is calculated based on the population displacement, flood fatality data, and preset standardized rules. A prediction model is generated based on the population displacement, the per capita purchasing power parity, and the mortality rate. The data to be estimated is input into the prediction model, and the estimated mortality rate is output.

2. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 1, characterized in that, The calculation formula for the standardization rule is as follows: in, Indicates the mortality rate. This indicates the number of deaths caused by the floods. Indicates the number of people displaced. This represents the standardization coefficient.

3. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 1, characterized in that, The step of generating a prediction model based on the mortality rate, the number of people displaced, and the mortality rate specifically includes: Construct a logarithmic spatial feature vector based on the population displacement and the per capita purchasing power parity; Construct logarithmic space labels based on the aforementioned lethality rate; The power-law regression model is trained based on the logarithmic space feature vector and the logarithmic space label to obtain the prediction model.

4. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 3, characterized in that, The power-law regression model is as follows: in, , , Indicates the corresponding model parameters. Indicates the mortality rate. Indicates the number of people displaced. It represents purchasing power parity per capita.

5. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 4, characterized in that, The prediction model is as follows: in, This represents the estimated mortality rate.

6. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 1, characterized in that, Also includes: A Gaussian process regression model is selected as the auxiliary model, and the uncertainty measure corresponding to the mortality rate estimate is output.

7. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 6, characterized in that, The Gaussian process regression model satisfies the following in logarithmic space: in, Indicates the mortality rate. Indicates the number of people displaced. It represents purchasing power parity per capita.

8. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 1, characterized in that, Also includes: The validity of the population displacement and the per capita purchasing power parity are tested and anomaly detection is performed.

9. The flood mortality rate estimation method based on population displacement and per capita purchasing power parity as described in claim 1, characterized in that, Also includes: The prediction model is evaluated for error, and the model parameters are updated based on the error evaluation results.

10. A device for estimating flood fatality rate based on population displacement and per capita purchasing power parity, characterized in that, include: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a flood mortality estimation method based on population displacement and per capita purchasing power parity as described in any one of claims 1-9.

Images