Infectious disease spread simulation system integrating multi-source heterogeneous data

Abstract

The present invention discloses an infectious disease diffusion simulation system integrating multi-source heterogeneous data, and relates to the field of data processing technology. The system comprises: collecting discovery feature information of a target infectious disease to be diffused; collecting multiple regional feature data sets according to heterogeneous multi-source data; analyzing the correlation between the diffusion of the target infectious disease and the multi-source data based on the historical infectious data of the target infectious disease; conducting simulation prediction of the diffusion of the target infectious disease in multiple potential infectious areas, and obtaining multiple simulation result sets; constructing multiple diffusion Merkel trees; obtaining the similarity coefficient between each diffusion Merkel tree and other multiple diffusion Merkel trees; and calculating and obtaining multiple diffusion simulation results of multiple potential infectious areas. The system solves the technical problem that the existing infectious disease diffusion simulation is often based on limited parameters and assumptions, resulting in deviations between the simulation results and the actual situation, and achieves the technical effect of improving the accuracy and comprehensiveness of the infectious disease diffusion simulation results.

Description

Technical Field

[0001] The present application relates to fields related to data processing technology, and specifically to an infectious disease diffusion simulation system that integrates multi-source heterogeneous data. Background Art

[0002] In the field of public health, infectious diseases have always been a key issue of concern. From the plague and cholera in ancient times to the influenza in modern times, the outbreak of infectious diseases has brought huge challenges to human society. In order to more effectively prevent and control the spread of infectious diseases, scientists have been working hard to find and verify various prediction and simulation methods. The simulation of the spread of infectious diseases has become an important tool for preventing and controlling the spread of diseases. However, traditional simulations of the spread of infectious diseases are usually based on several limited parameters and assumptions, such as the infection rate, recovery rate, mortality rate of the disease, and the contact pattern of the population. Data limitations lead to a certain deviation between the simulation results and the actual development of infectious diseases. Moreover, with the increase in population mobility, traditional simulations of the spread of infectious diseases cannot adapt to the complex, changeable and unpredictable spread of infectious diseases.

[0003] Therefore, at this stage, there are technical problems in the related technologies for simulating the spread of infectious diseases, which are based on limited parameters and assumptions, resulting in deviations between the simulation results and the actual situation. Summary of the invention

[0004] This application provides an infectious disease spread simulation system that integrates multi-source heterogeneous data and adopts technical means such as Merkle tree and similarity analysis to solve the technical problem that existing infectious disease spread simulations are often based on limited parameters and assumptions, resulting in deviations between simulation results and actual conditions, thereby achieving the technical effect of improving the accuracy and comprehensiveness of infectious disease spread simulation results.

[0005] The present application provides an infectious disease diffusion simulation system that integrates multi-source heterogeneous data, the system comprising: a discovery feature information acquisition module, the discovery feature information acquisition module is used to collect discovery feature information of a target infectious disease to be subjected to diffusion simulation, and determine multiple potential infectious areas based on the discovery feature information, wherein the discovery feature information includes discovery location information and discovery scale information; a regional feature data set acquisition module, the regional feature data set acquisition module is used to collect multiple regional feature data sets of the multiple potential infectious areas according to heterogeneous multi-source data, wherein the multi-source data includes population density, population mobility and population structure; a diffusion correlation coefficient acquisition module, the diffusion correlation coefficient acquisition module is used to analyze the correlation between the diffusion of the target infectious disease and the multi-source data based on the historical infection data of the target infectious disease, and obtain multiple diffusion correlation coefficients; a simulation result set acquisition module, the simulation result set acquisition module is used to respectively obtain the diffusion correlation coefficients according to the historical infection data of the target infectious disease and the multi-source data; The multi-source data and discovery scale information in the multiple regional feature data sets are used to simulate and predict the spread of the target infectious disease in the multiple potential infectious areas, and multiple simulation result sets are obtained, wherein each simulation result set includes diffusion parameters of multiple potential infectious areas; a diffusion Merkel tree construction module, the diffusion Merkel tree construction module is used to construct multiple diffusion Merkel trees based on multiple simulation result sets; a diffusion influence coefficient acquisition module, the diffusion influence coefficient acquisition module is used to perform pairwise combination and similarity analysis based on multiple diffusion Merkel trees, obtain the similarity coefficient of each diffusion Merkel tree with other multiple diffusion Merkel trees, calculate and obtain multiple diffusion accuracy coefficients, and calculate and obtain multiple diffusion influence coefficients in combination with the multiple diffusion correlation coefficients; a diffusion simulation result calculation module, the diffusion simulation result calculation module is used to calculate and obtain multiple diffusion simulation results for multiple potential infectious areas based on multiple diffusion influence coefficients and multiple simulation result sets.

[0006] In a possible implementation, the discovery feature information collection module also performs the following processing: collecting the location where the target infectious disease was discovered and the number of patients, obtaining the discovery location information and the discovery scale information as discovery feature information; based on the discovery location information, selecting the location where people are active within a preset range to obtain multiple potential infection areas.

[0007] In a possible implementation, the diffusion correlation coefficient acquisition module also performs the following processing: based on the historical infection data of the target infectious disease, multiple historical regional characteristic data sets and historical diffusion ratio sets are obtained; according to the multi-source data, the multiple historical regional characteristic data sets are divided to obtain historical dense data sets, historical flow data sets and historical structure data sets; based on the historical dense data sets, historical flow data sets, historical structure data sets and historical diffusion ratio sets, the correlation between the diffusion of the target infectious disease and the multi-source data is analyzed to obtain multiple diffusion correlation coefficients.

[0008] In a possible implementation, the diffusion correlation coefficient obtaining module further performs the following processing: performing dimensionless processing on the historical diffusion ratio set to construct a diffusion data sequence; performing dimensionless processing on the historical dense data set, the historical flow data set, and the historical structure data set to respectively construct a dense data sequence, a flow data sequence, and a structure data sequence; and calculating and obtaining a plurality of diffusion correlation coefficients based on the grey correlation coefficient calculation formula according to the constructed dense data sequence, flow data sequence, structure data sequence, and diffusion data sequence, as shown in the following formula: ;

[0009] in, is the influence correlation coefficient between the kth data in the ith data sequence and the kth data in the diffusion data sequence in the dense data sequence, the flow data sequence and the structure data sequence, is an adjustable calculation coefficient, is the absolute difference between the kth data in the i-th data sequence and the kth data in the diffusion data sequence, is the minimum difference between the two levels, is the maximum difference between the two levels.

[0010] In a possible implementation, the simulation result set acquisition module also performs the following processing: based on the diffusion monitoring records of the target infectious disease, obtain a sample discovery scale information set, a sample density information set, a sample fluidity information set, a sample structure information set and a sample diffusion parameter set; respectively use the sample density information set, the sample fluidity information set, the sample structure information set in combination with the sample discovery scale information set and the sample diffusion parameter set to construct a dense diffusion simulation branch, a flow diffusion simulation branch and a structural diffusion simulation branch to obtain a diffusion simulator; divide the multiple regional feature data sets according to the multi-source data, and input the discovery scale information into the diffusion simulator to obtain multiple simulation result sets, wherein each simulation result set includes multiple diffusion parameters for diffusion simulation prediction according to a class of data from multiple potential sensing areas.

[0011] In a possible implementation, the diffusion Merkel tree construction module also performs the following processing: using multiple dense diffusion parameters in the dense simulation result set within the multiple simulation result sets to construct multiple leaf nodes of the dense Merkel tree; hashing the dense diffusion parameters within the multiple leaf nodes to construct multiple first-level branch nodes, and hashing the hash value combinations within adjacent first-level branch nodes to construct multiple first-level leaf nodes until a root node is constructed to obtain a dense Merkel tree; constructing a flow Merkel tree and a structure Merkel tree based on the flow simulation result set and the structure simulation result set within the multiple simulation result sets.

[0012] In a possible implementation, the diffusion influence coefficient acquisition module also performs the following processing: based on the dense Merkel tree, calculate the average of the similarities between the flow Merkel tree and the structural Merkel tree and the dense Merkel tree as a dense similarity coefficient; calculate and obtain the flow similarity coefficient and the structural similarity coefficient as multiple similarity coefficients.

[0013] In a possible implementation, the diffusion influence coefficient acquisition module further performs the following processing: according to the sizes of the dense similarity coefficient, the flow similarity coefficient and the structural similarity coefficient, multiple diffusion accuracy coefficients are allocated and calculated; and multiple diffusion influence coefficients are calculated by combining the multiple diffusion accuracy coefficients and multiple diffusion correlation coefficients.

[0014] Through the infectious disease diffusion simulation system that integrates multi-source heterogeneous data proposed in this application, it is planned to collect the discovery feature information of the target infectious disease to be simulated for diffusion; collect multiple regional feature data sets according to heterogeneous multi-source data; analyze the correlation between the diffusion of the target infectious disease and the multi-source data based on the historical infection data of the target infectious disease; simulate and predict the diffusion of the target infectious disease in multiple potential infection areas to obtain multiple simulation result sets; construct multiple diffusion Merkel trees; perform pairwise combination and similarity analysis based on multiple diffusion Merkel trees to obtain the similarity coefficient of each diffusion Merkel tree with other multiple diffusion Merkel trees; calculate and obtain multiple diffusion simulation results for multiple potential infection areas. It solves the technical problem that the existing infectious disease diffusion simulation is often based on limited parameters and assumptions, resulting in deviations between the simulation results and the actual situation, and achieves the technical effect of improving the accuracy and comprehensiveness of the infectious disease diffusion simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0016] Figure 1 A schematic diagram of the structure of an infectious disease spread simulation system that integrates multi-source heterogeneous data provided in an embodiment of the present application;

[0017] Figure 2 A schematic diagram of the execution process of a diffusion Merkle tree construction module in an infectious disease diffusion simulation system that integrates multi-source heterogeneous data provided in an embodiment of the present application.

[0018] Explanation of the reference numerals: discovery feature information acquisition module 10, regional feature data set acquisition module 20, diffusion correlation coefficient acquisition module 30, simulation result set acquisition module 40, diffusion Merkel tree construction module 50, diffusion influence coefficient acquisition module 60, diffusion simulation result calculation module 70. DETAILED DESCRIPTION

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0022] The present application embodiment provides an infectious disease diffusion simulation system that integrates multi-source heterogeneous data, such as Figure 1 As shown, the system comprises:

[0023] The discovery feature information collection module 10 is used to collect the discovery feature information of the target infectious disease to be simulated for diffusion, and determine multiple potential infection areas based on the discovery feature information, wherein the discovery feature information includes discovery location information and discovery scale information. The discovery feature information of the infectious disease is collected, and the personnel activity area is selected as the potential infection area according to the preset range, wherein the discovery feature information includes discovery location information and discovery scale information. Specifically, the discovery location information refers to the specific geographical location information of the first discovery or report of the target infectious disease case, such as streets, buildings, communities, cities, etc.; the discovery scale information describes the number and distribution of infectious disease cases found in the area, the number of cases refers to the total number of infectious disease cases found in a specific location, and the case distribution refers to the distribution of cases in a specific location, for example, the cases are concentrated in a small area or scattered throughout the entire area. Based on the above-mentioned discovery feature information, multiple potential infection areas can be determined. For example, if the number of cases in a certain area suddenly increases, or the distribution of cases shows an obvious aggregation trend, then this area is regarded as a potential infection area.

[0024] Below, the specific configuration of the discovery feature information collection module 10 will be described in detail. The discovery feature information collection module 10 further includes: collecting the location where the target infectious disease was discovered and the number of patients, obtaining the discovery location information and the discovery scale information as discovery feature information. Usually, discovery feature information is collected from multiple data sources, such as reports from public health departments, records from medical institutions, laboratory test results, etc., to form accurate and reliable discovery feature information. Based on the discovery location information, the activity area of ​​​​people is selected within a preset range to obtain multiple potential infection areas. According to the discovery location of known cases, combined with the activity patterns of the population and possible transmission paths, it is inferred and delineated that there may be further cases or areas where the epidemic spreads.

[0025] The regional feature data set acquisition module 20 is used to collect multiple regional feature data sets of the multiple potential infection areas according to heterogeneous multi-source data, wherein the multi-source data includes crowd density, crowd mobility and crowd structure. According to data such as population density, population mobility and population structure, regional characteristic data sets of each potential infection area are collected to obtain multiple regional characteristic data sets. Specifically, population density refers to the degree or density of population aggregation in a certain area, for example, the number of people per square kilometer. Crowd density data can be obtained in a variety of ways, such as surveillance cameras in public places, positioning data of mobile devices, check-in data on social media, etc. In dense crowds, diseases are more likely to spread through contact between people; population mobility refers to the movement and circulation of people in geographical space, which can be obtained through positioning data of mobile devices, travel data of transportation departments, sharing data on social media, etc. The increase in mobile population may lead to the cross-regional spread of diseases, thereby expanding the scope of the epidemic; population structure refers to the composition of the population in a region or group, including age, gender, occupation, etc. The data on population structure can be obtained through censuses, social surveys, administrative records, etc. Different population structures may have different effects on the spread of infectious diseases. For example, the elderly, children or people with weak immune systems may be more susceptible to infection.

[0026] The diffusion correlation coefficient acquisition module 30 is used to analyze the correlation between the spread of the target infectious disease and the multi-source data based on the historical infection data of the target infectious disease, and obtain multiple diffusion correlation coefficients. Collect the historical infection data of the target infectious disease and related multi-source data, including the number of cases, infection time, infection location, population density, population mobility, population structure and other information, and perform gray correlation analysis to reveal how these factors affect or predict the spread speed and range of infectious diseases, and obtain the correlation coefficient between the change of multi-source data and the spread of infectious diseases. Specifically, the diffusion correlation coefficient is used to measure the correlation strength or dependence between the spread of the target infectious disease and various multi-source data factors, reflecting the correlation strength and direction between the multi-source data variables and the spread of infectious diseases. For example, if the correlation coefficient between the population density of a certain area and the spread of infectious diseases is high, it means that the population density of the area has an important impact on the spread of infectious diseases.

[0027] The specific configuration of the diffusion correlation coefficient acquisition module 30 will be described in detail below. The diffusion correlation coefficient acquisition module 30 further includes: based on the historical infection data of the target infectious disease, obtaining multiple historical regional feature data sets and historical diffusion ratio sets. Using existing historical data to extract and analyze the characteristics and diffusion patterns of infectious diseases in different regions helps to better understand and predict the spread dynamics of infectious diseases. Specifically, the historical regional feature data set contains various characteristic information when infectious diseases spread in different regions, which may include population density, population mobility, distribution of medical resources, environmental factors (such as climate, geographical conditions), socioeconomic conditions, etc.; the historical diffusion ratio set records the diffusion ratio or diffusion speed of infectious diseases between different regions, which can understand the spread trend and law of the epidemic between different regions and predict the future spread of the epidemic between different regions. According to the multi-source data, the multiple historical regional feature data sets are divided to obtain a historical dense data set, a historical flow data set, and a historical structure data set. According to the historical dense data set, the historical flow data set, the historical structure data set, and the historical diffusion ratio set, the correlation between the spread of the target infectious disease and the multi-source data is analyzed to obtain multiple diffusion correlation coefficients. Based on multi-source data sets such as historical diffusion ratio sets, historical dense data sets, historical flow data sets, and historical structure data sets, the correlation between the spread of the target infectious disease and the multi-source data is analyzed, that is, the correlation between each factor and the spread of the infectious disease is evaluated, and the diffusion correlation coefficient is calculated to indicate the degree of correlation between a certain factor and the spread of the infectious disease.

[0028] The specific configuration of the diffusion correlation coefficient acquisition module 30 will be described in detail below. The diffusion correlation coefficient acquisition module 30 further includes: performing dimensionless processing on the historical diffusion ratio set to construct a diffusion data sequence. When conducting infectious disease diffusion analysis, the historical diffusion ratio set may contain diffusion ratio data at different times, in different regions or under different conditions. These data may have different dimensions or units, making direct comparison or analysis difficult. After dimensionless processing, a diffusion data sequence can be constructed, including historical diffusion ratio data with dimension effects removed. For example, the diffusion data sequence can be used for time series analysis, trend forecasting, or correlation analysis with other data sets. Specifically, the purpose of dimensionless processing is to eliminate the influence of these different dimensions or units on the data, so that all data can be compared and analyzed on the same scale. The historical dense data set, the historical flow data set, and the historical structure data set are dimensionless processed to construct dense data sequences, flow data sequences, and structure data sequences, respectively. For the historical dense data set, the historical flow data set and the historical structure data set, appropriate de-dimensionalization methods are used to eliminate the dimension or unit differences between different data sets, and the processed data are arranged in chronological order or other appropriate order to construct dense data sequences, flow data sequences and structure data sequences. According to the construction of dense data sequences, flow data sequences, structure data sequences and diffusion data sequences, multiple diffusion correlation coefficients are calculated based on the grey correlation coefficient calculation formula, as shown in the following formula: ;

[0029] in, is the influence correlation coefficient between the kth data in the ith data sequence and the kth data in the diffusion data sequence in the dense data sequence, the flow data sequence and the structure data sequence, is an adjustable calculation coefficient, is the absolute difference between the kth data in the i-th data sequence and the kth data in the diffusion data sequence, is the minimum difference between the two levels, is the maximum difference between the two levels.

[0030] The simulation result set acquisition module 40 is used to simulate and predict the spread of the target infectious disease in the multiple potential infectious areas according to the multi-source data and the discovery scale information in the multiple regional feature data sets, and obtain multiple simulation result sets, wherein each simulation result set includes the diffusion parameters of multiple potential infectious areas. The multi-source data and the discovery scale information in the multiple regional feature data sets are used to construct a simulation prediction model, simulate and predict the spread of the target infectious disease in multiple potential infectious areas, and obtain multiple simulation result sets. Specifically, each set represents the spread of the target infectious disease in the potential infectious area under a set of simulation conditions. Each simulation result set contains the diffusion parameters of multiple potential infectious areas, which are used to quantitatively describe the spread process of the infectious disease. The diffusion parameters may include but are not limited to the number of infected people, the transmission speed, the peak value, the duration, etc. Specifically, the simulation prediction will give the number of infected people in each potential infectious area at different time points, which is helpful to understand the severity and transmission speed of the epidemic; the transmission speed describes the rate at which the infectious disease spreads among the population; the peak value refers to the highest point of the number of infected people during the epidemic spread; the duration refers to the length of time from the beginning to the end of the epidemic. By comprehensively considering multi-source data and discovery scale information in multiple regional characteristic datasets, simulation predictions can more comprehensively assess the risk of the target infectious disease spreading in potential infectious areas.

[0031] Below, the specific configuration of the simulation result set acquisition module 40 will be described in detail. The simulation result set acquisition module 40 further includes: based on the diffusion monitoring record of the target infectious disease, obtaining a sample discovery scale information set, a sample density information set, a sample mobility information set, a sample structure information set and a sample diffusion parameter set. Extract and analyze various information related to the spread of the target infectious disease from the monitoring records to form a series of data sets. Specifically, the sample discovery scale information set refers to the number or scale information of the target infectious disease cases recorded during the monitoring process, including the number of new cases, the cumulative number of cases, the number of cases in the infectious disease outbreak area, etc. in each monitoring cycle; the sample density information set refers to the information on the density of the distribution of cases in geographic space; the sample mobility information set includes information on the flow of people, for example, by analyzing the travel history, contact history and other information of the case, the transmission chain of the epidemic can be traced; the sample structure information set refers to the social structure, age structure, gender structure and other aspects of the case; the sample diffusion parameter set includes quantitative indicators such as the rate, range, and direction of the epidemic spread, which are used to describe the dynamic changes and spread trends of the epidemic. The sample density information set, sample mobility information set, sample structure information set are respectively used in combination with the sample discovery scale information set and the sample diffusion parameter set to construct a dense diffusion simulation branch, a flow diffusion simulation branch and a structure diffusion simulation branch to obtain a diffusion simulator. By using the sample density information set, the sample mobility information set, the sample structure information set, in combination with the sample discovery scale information set and the sample diffusion parameter set, three different simulation branches, namely, a dense diffusion simulation branch, a flow diffusion simulation branch and a structure diffusion simulation branch, can be constructed to form a complete diffusion simulator for simulating the diffusion of the target infectious disease. Specifically, the dense diffusion simulation branch combined with the sample discovery scale information set (such as the number of new cases and the cumulative number of cases) can simulate the density and spread speed of the epidemic in different regions; the flow diffusion simulation branch combined with the sample discovery scale information set and the sample diffusion parameter set (such as the spread rate and direction) can simulate how the epidemic spreads through personnel flow and logistics transportation; the structure diffusion simulation branch combined with the sample discovery scale information set and the sample diffusion parameter set can simulate the spread of the epidemic in different social groups. According to the multi-source data, the multiple regional feature data sets are divided and input into the diffusion simulator in combination with the discovery scale information to obtain multiple simulation result sets, wherein each simulation result set includes multiple diffusion parameters for diffusion simulation prediction based on a class of data from multiple potential sensing areas.Multiple regional characteristic data sets are divided and input into the diffusion simulator in combination with the discovery scale information to obtain multiple simulation result fits. Specifically, multiple regional characteristic data sets are divided according to multi-source data (such as historical dense data sets, historical flow data sets, historical structure data sets, etc.), and input into the diffusion simulator in combination with the sample discovery scale information set (such as the number of new cases, the cumulative number of cases, etc.). In combination with the regional characteristic data sets and the discovery scale information, the diffusion simulator simulates the spread of infectious diseases in different regions and outputs multiple simulation results. Each simulation result set includes multiple diffusion parameters predicted by diffusion simulation according to a class of data from multiple potential sensing areas, which are used to describe the spread of the epidemic in different regions.

[0032] The diffusion Merkel tree construction module 50 is used to construct multiple diffusion Merkel trees based on multiple simulation result sets. Multiple diffusion Merkle trees are constructed based on multiple simulation result sets, that is, the Merkle tree structure is used for data analysis in the simulation of infectious disease diffusion. The Merkle tree is mainly used to integrate and manage the simulation results of various potential infectious areas to ensure the integrity and verifiability of the data. Specifically, each simulation result set may contain diffusion parameters of multiple potential infectious areas, such as the number of infected people, transmission speed, peak value and duration, etc. These simulation result sets can be regarded as leaf nodes of the Merkle tree, and each node represents a specific simulation result. Specifically, each simulation result set is used as the leaf node data of the Merkle tree, and each simulation result set (i.e., leaf node) is hashed to generate a unique hash value. Starting from the leaf node, the hash values ​​of the two adjacent nodes are spliced, and the hash calculation is performed again to generate the hash value of their parent node, and the process is recursive until only one node is left, i.e., the root node of the Merkle tree. In the simulation of infectious disease diffusion, constructing multiple diffusion Merkle trees can more conveniently manage and analyze a large amount of simulation result data.

[0033] The specific configuration of the diffusion Merkle tree construction module 50 will be described in detail below. Figure 2As shown, the diffusion Merkel tree construction module 50 further includes: using multiple dense diffusion parameters in the dense simulation result set in the multiple simulation result sets to construct multiple leaf nodes of the dense Merkel tree. The dense diffusion parameters in the multiple leaf nodes are hashed to construct multiple first-level branch nodes, and the hash value combinations in adjacent first-level branch nodes are hashed to construct multiple first-level leaf nodes until a root node is constructed to obtain a dense Merkel tree. The leaf node is the bottom node of the Merkle tree. Each leaf node stores a dense diffusion parameter. The dense diffusion parameter in each leaf node is hashed to generate the hash value of the leaf node. The hashing process refers to the process of converting data into a fixed-length hash value using a hash algorithm. When all leaf nodes are hashed, the first-level branch node is constructed. The first-level branch node is the node on the upper layer of the leaf node. The hash values ​​of adjacent leaf nodes are combined (such as connecting them in series), and then the combination is hashed to generate the hash value of the first-level branch node. If the number of leaf nodes is an odd number, the hash value of the last leaf node may need to be copied once to match the hash value of another leaf node; the root node is the top node of the Merkle tree, which contains a summary of the data of the entire tree. According to the flow simulation result set and the structure simulation result set in multiple simulation result sets, the flow Merkle tree and the structure Merkle tree are constructed.

[0034] The diffusion influence coefficient obtaining module 60 is used to perform pairwise combination and similarity analysis based on multiple diffusion Merkel trees, obtain the similarity coefficient between each diffusion Merkel tree and other multiple diffusion Merkel trees, calculate and obtain multiple diffusion accuracy coefficients, and calculate and obtain multiple diffusion influence coefficients in combination with the multiple diffusion association coefficients. Multiple diffusion Merkel trees are combined in pairs, and the similarity between each pair of Merkel trees is calculated, for example, their root hash values, leaf node data, etc. are compared. For each pair of Merkel trees, a similarity coefficient is obtained through similarity analysis. The similarity coefficient can be a value between 0 and 1, where 0 indicates complete dissimilarity and 1 indicates complete similarity, reflecting the degree of similarity between two sets of simulation results. The Merkel trees constructed by the same multiple simulation results should be consistent, and the Merkel trees constructed by similar multiple simulation results have greater similarity. The diffusion accuracy coefficient is used to evaluate the accuracy of each diffusion Merkel tree (i.e., each set of simulation results). For example, the diffusion accuracy coefficient is obtained by comparing the degree of fit between the simulation results and the actual data, and then the diffusion influence coefficient is calculated by combining the obtained multiple diffusion correlation coefficients. Specifically, based on the three similarities and combined with the corresponding diffusion correlation coefficients, three diffusion accuracy coefficients are assigned to evaluate the importance and influence of each diffusion Merkel tree (i.e., each set of simulation results) on the overall infectious disease diffusion simulation, for example, the ratio of each similarity coefficient to the sum of the three similarity coefficients.

[0035] The specific configuration of the diffusion influence coefficient obtaining module 60 will be described in detail below. The diffusion influence coefficient obtaining module 60 further includes: based on the dense Merkel tree, calculating the average of the similarities between the flow Merkel tree and the structural Merkel tree and the dense Merkel tree as a dense similarity coefficient. The dense Merkle tree is a Merkle tree constructed based on dense diffusion parameters, the flow Merkle tree is a Merkle tree constructed based on flow diffusion parameters (such as personnel flow, logistics transportation, etc.), and the structural Merkle tree is a Merkle tree constructed based on structural diffusion parameters (such as social structure, age structure, etc.). The similarity of the hash values ​​of the corresponding nodes of the two Merkle trees at the same level is calculated to determine the similarity between the flow Merkle tree and the structural Merkle tree and the dense Merkle tree. For example, by comparing whether the hash values ​​are the same or calculating a certain distance between the hash values ​​(such as the Hamming distance), the mean similarity between the flow Merkle tree and the structural Merkle tree and the dense Merkle tree is calculated to obtain a dense similarity coefficient, which indicates the extent to which the flow Merkle tree and the structural Merkle tree are similar to the dense Merkle tree. The flow similarity coefficient and the structural similarity coefficient are calculated as multiple similarity coefficients. Similarly, the flow similarity coefficient and the structural similarity coefficient are calculated and used together as multiple similarity coefficients.

[0036] The specific configuration of the diffusion influence coefficient acquisition module 60 will be described in detail below. The diffusion influence coefficient acquisition module 60 further includes: according to the size of the dense similarity coefficient, the flow similarity coefficient and the structural similarity coefficient, multiple diffusion accuracy coefficients are allocated and calculated. The accuracy of the diffusion process is evaluated or predicted according to different types of similarity coefficients (such as dense similarity coefficient, flow similarity coefficient and structural similarity coefficient). Specifically, according to the size of the above similarity coefficients, multiple diffusion accuracy coefficients can be allocated and calculated to quantify the accuracy and reliability of different parameters or models in predicting or simulating the diffusion process. For example, if the dense similarity coefficient is the highest, it can be considered that the dense parameter has a higher accuracy in predicting the diffusion process, and a higher diffusion accuracy coefficient is allocated accordingly. Combine the multiple diffusion accuracy coefficients and the multiple diffusion correlation coefficients to calculate and obtain multiple diffusion influence coefficients. Combine the multiple diffusion accuracy coefficients and the multiple diffusion correlation coefficients to calculate the multiple diffusion influence coefficients in order to quantify the degree of influence of different factors or parameters on the diffusion process. Specifically, the multiple diffusion accuracy coefficients and the multiple diffusion correlation coefficients are weighted and combined to obtain the corresponding multiple diffusion influence coefficients.

[0037] Diffusion simulation result calculation module 70, the diffusion simulation result calculation module 70 is used to calculate multiple diffusion simulation results for multiple potential infectious areas according to multiple diffusion influence coefficients and multiple simulation result sets. According to the calculated multiple diffusion influence coefficients, the importance or credibility of each simulation result set is evaluated, and the multiple simulation result sets are integrated, for example, by weighted average, fuzzy comprehensive evaluation, etc. The diffusion influence coefficient will be used as a weight factor to weight different simulation result sets to reflect their contribution to the final result. By integrating multiple simulation result sets and combining the weight adjustment of the diffusion influence coefficient, multiple diffusion simulation results for multiple potential infectious areas can be obtained, which more comprehensively reflects the spread of infectious diseases in different regions, including key indicators such as the number of infected people, transmission speed, and duration, so as to obtain more accurate and comprehensive diffusion simulation results.

[0038] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0039] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. An infectious disease diffusion simulation system integrating multi-source heterogeneous data, characterized by: The system comprises: A discovery feature information collection module, the discovery feature information collection module is used to collect discovery feature information of the target infectious disease to be subjected to diffusion simulation, and determine multiple potential infection areas based on the discovery feature information, wherein the discovery feature information includes discovery location information and discovery scale information; A regional feature data set acquisition module, the regional feature data set acquisition module is used to collect multiple regional feature data sets of the multiple potential infection areas according to heterogeneous multi-source data, wherein the multi-source data includes crowd density, crowd mobility and crowd structure; A diffusion correlation coefficient acquisition module, the diffusion correlation coefficient acquisition module is used to analyze the correlation between the diffusion of the target infectious disease and the multi-source data based on the historical infection data of the target infectious disease, and obtain multiple diffusion correlation coefficients, wherein the diffusion correlation coefficients are used to measure the correlation strength or dependence degree between the diffusion of the target infectious disease and the multi-source data; A simulation result set acquisition module, the simulation result set acquisition module is used to perform simulation prediction of the spread of the target infectious disease in the multiple potential infectious areas according to the multi-source data and discovery scale information in the multiple regional feature data sets, and obtain multiple simulation result sets, wherein each simulation result set includes diffusion parameters of multiple potential infectious areas, and the diffusion parameters of the potential infectious areas are used to quantify the diffusion process of the target infectious disease in the potential infectious areas; A diffusion Merkle tree construction module, wherein the diffusion Merkle tree construction module is used to construct multiple diffusion Merkle trees based on multiple simulation result sets, wherein the diffusion Merkle trees are used to integrate and manage simulation results of potential infection areas; A diffusion influence coefficient acquisition module, the diffusion influence coefficient acquisition module is used to perform pairwise combination and similarity analysis based on multiple diffusion Merkel trees, obtain the similarity coefficient between each diffusion Merkel tree and other multiple diffusion Merkel trees, calculate and obtain multiple diffusion accuracy coefficients, and calculate and obtain multiple diffusion influence coefficients in combination with the multiple diffusion association coefficients, wherein the similarity coefficient reflects the similarity between each diffusion Merkel tree and other diffusion Merkel trees, the diffusion accuracy coefficient is used to evaluate the accuracy of each diffusion Merkel tree, and the diffusion influence coefficient is the importance of a simulation result set to the diffusion simulation; A diffusion simulation result calculation module, wherein the diffusion simulation result calculation module is used to calculate and obtain multiple diffusion simulation results for multiple potential infection areas according to multiple diffusion influence coefficients and multiple simulation result sets; The diffusion correlation coefficient obtaining module performs the following steps: Based on the historical infection data of the target infectious disease, a plurality of historical regional feature data sets and a historical diffusion ratio set are obtained, wherein the historical diffusion ratio set is the diffusion ratio of the target infectious disease in a plurality of potential infection areas; According to the multi-source data, the plurality of historical regional characteristic data sets are divided to obtain a historical dense data set, a historical flow data set and a historical structure data set; Analyzing the correlation between the spread of the target infectious disease and the multi-source data based on the historical dense data set, the historical flow data set, the historical structure data set, and the historical diffusion ratio set to obtain a plurality of diffusion correlation coefficients; The diffusion correlation coefficient obtaining module performs the following steps: De-dimensionalizing the historical diffusion ratio set to construct a diffusion data sequence; De-dimensionalizing the historical dense data set, the historical flow data set, and the historical structure data set to construct a dense data sequence, a flow data sequence, and a structure data sequence, respectively; According to the construction of dense data sequence, flow data sequence, structure data sequence and diffusion data sequence, based on the grey correlation coefficient calculation formula, multiple diffusion correlation coefficients are calculated, as shown in the following formula: Among them, i (k) is the influence correlation coefficient between the kth data in the ith data sequence and the kth data in the diffusion data sequence in the dense data sequence, the flow data sequence and the structure data sequence, ρ is the adjustable calculation coefficient, |y(k)-x i (k)| is the absolute difference between the kth data in the i-th data sequence and the kth data in the diffusion data sequence, minmin|y(k)-x i (k)| is the minimum difference between the two levels, maxmax|y(k)-x i (k)| is the maximum difference between the two levels; The simulation result set obtaining module performs the following steps: Based on the diffusion monitoring records of the target infectious disease, obtaining a sample discovery scale information set, a sample density information set, a sample mobility information set, a sample structure information set, and a sample diffusion parameter set; The sample density information set, the sample mobility information set, the sample structure information set are respectively used in combination with the sample discovery scale information set and the sample diffusion parameter set to construct a dense diffusion simulation branch, a flow diffusion simulation branch and a structure diffusion simulation branch to obtain a diffusion simulator; According to the multi-source data, the plurality of regional feature data sets are divided, and combined with the discovery scale information, they are input into the diffusion simulator to obtain a plurality of simulation result sets, wherein each simulation result set includes a plurality of diffusion parameters predicted by diffusion simulation according to a class of data of a plurality of potential sensing areas; The diffusion Merkle tree construction module executes the following steps: constructing a plurality of leaf nodes of a dense Merkle tree using a plurality of dense diffusion parameters in a dense simulation result set in the plurality of simulation result sets; The dense diffusion parameters in multiple leaf nodes are hashed to construct multiple first-level branch nodes, and the hash value combinations in adjacent first-level branch nodes are hashed to construct multiple first-level leaf nodes until a root node is constructed to obtain a dense Merkle tree; Constructing a flow Merkle tree and a structure Merkle tree according to a flow simulation result set and a structure simulation result set in a plurality of simulation result sets; The diffusion influence coefficient obtaining module performs the following steps: Based on the dense Merkle tree, the average of the similarities between the flow Merkle tree and the structural Merkle tree and the dense Merkle tree is calculated as a dense similarity coefficient; Calculate and obtain the flow similarity coefficient and the structure similarity coefficient as multiple similarity coefficients; The diffusion influence coefficient obtaining module performs the following steps: According to the sizes of the dense similarity coefficient, the flow similarity coefficient and the structural similarity coefficient, multiple diffusion accuracy coefficients are obtained by allocation calculation, wherein the corresponding diffusion accuracy coefficient is obtained by respectively calculating the ratio of each similarity coefficient to the sum of the dense similarity coefficient, the flow similarity coefficient and the structural similarity coefficient; A plurality of diffusion influence coefficients are calculated by combining the plurality of diffusion accuracy coefficients and the plurality of diffusion correlation coefficients.

2. The infectious disease diffusion simulation system integrating multi-source heterogeneous data according to claim 1 is characterized in that: The discovery feature information collection module performs the following steps: Collect the locations where the target infectious disease was discovered and the number of patients, and obtain the discovery location information and discovery scale information as discovery feature information; Based on the discovery location information, a selection is made of the areas where people move around within a preset range to obtain multiple potential infection areas.

Images