Decision support system construction method and system based on infection prevention and control

Abstract

The invention relates to the technical field of decision support systems, in particular to a decision support system construction method and system based on infection prevention and control. The method comprises the following steps: acquiring infection prevention and control related data and performing data cleaning processing to obtain infection transmission cleaning data and crowd flow cleaning data; performing propagation parameter extraction analysis on the infection propagation cleaning data, performing crowd aggregation statistical analysis on the crowd flow cleaning data, and performing potential association mining analysis on an analysis result to obtain propagation parameter key features and crowd aggregation key features to construct a crowd infection propagation model; performing parameter estimation optimization and infection prediction processing on the crowd infection propagation model to obtain infection propagation rule data; and performing prevention and control decision support analysis on the infection transmission rule data to construct a decision support system so as to execute a corresponding infection prevention and control decision. According to the invention, the decision support system for infection prevention and control can be accurately constructed, so that the quality and efficiency of infection prevention and control decisions are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of decision support systems, and in particular to a method and system for constructing a decision support system based on infection prevention and control. Background Art

[0002] Infection prevention and control is a crucial public health area, playing a crucial role in disease outbreaks and spread. However, due to the complexity of infectious viruses, the development and implementation of prevention and control measures often present challenges. Existing decision support systems suffer from several issues, such as a lack of real-time performance, accuracy, and personalization. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a method for constructing a decision support system based on infection prevention and control to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a method for constructing a decision support system based on infection prevention and control includes the following steps:

[0005] Step S1: Acquire infection prevention and control related data, perform data cleaning on the infection prevention and control related data, and obtain infection prevention and control cleaned data, wherein the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data;

[0006] Step S2: extract and analyze the infection propagation parameters of the cleaned infection propagation data to obtain infection propagation parameter data;

[0007] Step S3: Performing crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data;

[0008] Step S4: Perform potential association mining analysis on the infection transmission parameter data and the crowd gathering data to obtain key features of the transmission parameters and key features of the crowd gathering; and construct a crowd infection transmission model based on the key features of the transmission parameters and the key features of the crowd gathering;

[0009] Step S5: Optimizing the parameters of the population infection propagation model to obtain an infection propagation optimization model; using the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data;

[0010] Step S6: Conduct prevention and control decision support analysis on the infection transmission pattern data to obtain prevention and control decision analysis results; build a decision support system based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

[0011] The present invention first collects data related to infection prevention and control, including infection transmission data and crowd flow data. At the same time, the collected data is cleaned, including data deduplication, missing value processing, outlier processing, etc., to ensure data quality and consistency. After cleaning, infection prevention and control cleaned data is obtained, which can be used for subsequent analysis and modeling. Secondly, the infection transmission cleaned data is subjected to propagation parameter extraction and analysis to analyze the relevant features such as the propagation process, propagation rate, propagation range, etc. in the infection transmission cleaned data to extract parameter data related to infection transmission. These parameters may include parameters such as propagation rate, infection probability, and infection transmission distance for subsequent modeling and analysis. Then, by performing aggregate statistical analysis on the cleaned crowd flow data, and analyzing data such as crowd movement trajectory, crowd density, and crowd flow direction, crowd gathering data can be obtained. These data may include information such as the location, time, and scale of crowd gathering, which is of great significance for understanding crowd flow patterns and clustered infection risks. Next, potential associations are mined and analyzed between infection transmission parameter data and population cluster data. Data mining techniques are used to identify possible correlations and interactions between transmission parameters and population clusters. Key characteristics of transmission parameters and population clusters are extracted from these characteristics and used to construct a population infection transmission model. Machine learning methods, statistical analysis, and other techniques can be used to build this model. Furthermore, parameter estimation and optimization are performed on the constructed population infection transmission model to find parameter values ​​that best align the model's predictions with the actual observed data, thereby improving the model's fitting performance. Parameter optimization yields an adjusted and improved infection transmission optimization model that better aligns with the actual data. Furthermore, the infection transmission optimization model is used to perform infection prediction on infection transmission parameter data and population cluster data. The model's predictive capabilities allow inferences about infection transmission patterns based on the available data and model parameters. This can help predict and understand infection spread trends, the effectiveness of control measures, and provide data support for guiding population behavior and policies. Finally, by conducting prevention and control decision support analysis on infection transmission pattern data, we can analyze infection transmission trends, influencing factors, etc. to obtain analysis results related to prevention and control decisions. These results can help decision makers understand the risks and impacts of infection transmission and formulate corresponding prevention and control strategies. Based on the analysis results, a decision support system is constructed to assist in the process of executing infection prevention and control decisions, thereby achieving the reliability, personalization and accuracy of the decision support system construction to support the formulation and implementation of infection prevention and control decisions in real time.

[0012] Preferably, the present invention further provides a decision support system construction system based on infection prevention and control, which is used to execute the above-mentioned decision support system construction method based on infection prevention and control, and the decision support system construction system based on infection prevention and control includes:

[0013] The infection data processing module is used to obtain infection prevention and control related data and perform data cleaning on the infection prevention and control related data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data;

[0014] A propagation parameter processing module is used to extract and analyze the propagation parameters of the infection propagation cleaning data to obtain infection propagation parameter data;

[0015] The crowd data analysis module is used to perform crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data;

[0016] The infection transmission model construction module is used to conduct potential correlation mining analysis on infection transmission parameter data and crowd gathering data to obtain key characteristics of transmission parameters and crowd gathering; and to construct a crowd infection transmission model based on the key characteristics of transmission parameters and crowd gathering;

[0017] The parameter optimization and prediction module is used to perform parameter estimation and optimization processing on the population infection propagation model to obtain the infection propagation optimization model; the infection propagation optimization model is used to perform infection prediction processing on the infection propagation parameter data and the crowd gathering data to obtain the infection propagation law data;

[0018] The decision support system construction module is used to conduct prevention and control decision support analysis on infection transmission law data to obtain prevention and control decision analysis results; a decision support system is constructed based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

[0019] In summary, the present invention provides a decision support system construction system based on infection prevention and control, which consists of an infection data processing module, a propagation parameter processing module, a crowd data analysis module, an infection propagation model construction module and a decision support system construction module. It can implement any one of the infection prevention and control-based decision support system construction methods described in the present invention, and is used to combine the operations between the computer programs running on each module to implement a decision support system construction method based on infection prevention and control. The internal structures of the system cooperate with each other, and by using infection prevention and control-related data and advanced technical means, according to the characteristics and laws of infection propagation, a decision support system is constructed that can analyze multiple factors such as infection propagation and crowd gathering, providing decision makers with real-time and accurate decision support, and combining the decision makers' needs and background information to generate personalized infection prevention and control decision recommendations, thereby improving the quality and efficiency of infection prevention and control decisions. This can greatly reduce duplication of work and manpower investment, and can quickly and effectively provide a more accurate and efficient decision support system construction process, thereby simplifying the operation process of the infection prevention and control-based decision support system construction system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0021] Figure 1 This is a flowchart of the steps of the method for constructing a decision support system based on infection prevention and control according to the present invention;

[0022] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0023] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0024] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative work are within the scope of protection of the present invention.

[0025] In addition, the accompanying drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor and / or microcontroller approaches.

[0026] It should be understood that although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used solely to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element, without departing from the scope of the exemplary embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the listed associated items.

[0027] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for constructing a decision support system based on infection prevention and control, the method comprising the following steps:

[0028] Step S1: Acquire infection prevention and control related data, perform data cleaning on the infection prevention and control related data, and obtain infection prevention and control cleaned data, wherein the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data;

[0029] Step S2: extract and analyze the infection propagation parameters of the cleaned infection propagation data to obtain infection propagation parameter data;

[0030] Step S3: Performing crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data;

[0031] Step S4: Perform potential association mining analysis on the infection transmission parameter data and the crowd gathering data to obtain key features of the transmission parameters and key features of the crowd gathering; and construct a crowd infection transmission model based on the key features of the transmission parameters and the key features of the crowd gathering;

[0032] Step S5: Optimizing the parameters of the population infection propagation model to obtain an infection propagation optimization model; using the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data;

[0033] Step S6: Conduct prevention and control decision support analysis on the infection transmission pattern data to obtain prevention and control decision analysis results; build a decision support system based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

[0034] In the embodiment of the present invention, please refer to Figure 1 FIG. 1 is a flow chart of the steps of the method for constructing a decision support system based on infection prevention and control according to the present invention. In this example, the steps of the method for constructing a decision support system based on infection prevention and control include:

[0035] Step S1: Acquire infection prevention and control related data, perform data cleaning on the infection prevention and control related data, and obtain infection prevention and control cleaned data, wherein the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data;

[0036] The present invention obtains infection prevention and control-related data by acquiring infection epidemic data, population flow data, medical data, and other data related to infection prevention and control from data sources such as public health departments, medical institutions, and research institutions. This data is then cleaned through repeated screening, anomaly filtering, data interpolation, and standardization to ultimately generate infection prevention and control cleaned data. This cleaned infection prevention and control data includes infection transmission cleaned data and population flow cleaned data.

[0037] Step S2: extract and analyze the infection propagation parameters of the cleaned infection propagation data to obtain infection propagation parameter data;

[0038] The embodiment of the present invention analyzes and extracts parameter data such as infection propagation trend path, infection propagation rate, infection propagation distance, infection rate and infection propagation latent rate from the infection propagation cleaning data, and ensures the consistency and accuracy of the data through operations such as data format conversion, outlier processing and missing value filling, and finally obtains infection propagation parameter data.

[0039] Step S3: Performing crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data;

[0040] The embodiment of the present invention performs statistical analysis on the crowd flow cleaning data to analyze the crowd flow behavior patterns, crowd flow frequency and regional crowd flow conditions, and analyzes and infers the crowd gathering conditions by using spatial statistical methods such as cluster analysis or hotspot analysis to identify the crowd gathering patterns in different areas, and finally obtain crowd gathering data.

[0041] Step S4: Perform potential association mining analysis on the infection transmission parameter data and the crowd gathering data to obtain key features of the transmission parameters and key features of the crowd gathering; and construct a crowd infection transmission model based on the key features of the transmission parameters and the key features of the crowd gathering;

[0042] The embodiment of the present invention uses statistical analysis, Pearson correlation coefficient calculation method, topic model, cluster analysis, association rule mining and other methods to perform potential association mining analysis on infection propagation parameter data and crowd aggregation data to identify potential patterns and key features in the infection propagation parameter data and crowd aggregation data, so as to obtain key features of propagation parameters and key features of crowd aggregation. Then, the key features of propagation parameters and key features of crowd aggregation are divided into training set, verification set and test set according to the preset division ratio of 7:2:1, and the SEIR model is used to perform model training, verification and testing based on the division results to construct a model that can accurately simulate and predict the process of infection propagation in the crowd, and finally a crowd infection propagation model is obtained.

[0043] Step S5: Optimizing the parameters of the population infection propagation model to obtain an infection propagation optimization model; using the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data;

[0044] The embodiment of the present invention estimates the parameters of a constructed population infection and propagation model using parameter estimation methods such as maximum likelihood estimation and Bayesian inference to enable the model to fit the observed data. The parameters of the population infection and propagation model are then optimized using parameter optimization methods such as gradient descent and genetic algorithms to minimize the model estimation error loss until the optimization goal is reached, thereby obtaining an optimized infection and propagation model. Then, infection prediction is performed by inputting infection and propagation parameter data and population aggregation data into the optimized infection and propagation optimization model to predict and infer infection propagation patterns such as infection trends and propagation rates, ultimately obtaining infection and propagation pattern data.

[0045] Step S6: Conduct prevention and control decision support analysis on the infection transmission pattern data to obtain prevention and control decision analysis results; build a decision support system based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

[0046] The embodiments of the present invention analyze and mine infection transmission data to reveal infection transmission patterns, trends, and risk factors. Then, based on the infection transmission analysis results, prevention and control decision-making analyses are conducted to formulate prevention and control strategies, allocate resources, and assess risks, thereby obtaining prevention and control decision-making analysis results. Finally, based on the prevention and control decision-making analysis results, a decision support system is constructed to assist in the formulation and implementation of corresponding infection prevention and control decisions, such as vaccination strategies, personal protection publicity, community prevention and control measures, and medical resource allocation.

[0047] The present invention first collects data related to infection prevention and control, including infection transmission data and crowd flow data. At the same time, the collected data is cleaned, including data deduplication, missing value processing, outlier processing, etc., to ensure data quality and consistency. After cleaning, infection prevention and control cleaned data is obtained, which can be used for subsequent analysis and modeling. Secondly, the infection transmission cleaned data is subjected to propagation parameter extraction and analysis to analyze the relevant features such as the propagation process, propagation rate, propagation range, etc. in the infection transmission cleaned data to extract parameter data related to infection transmission. These parameters may include parameters such as propagation rate, infection probability, and infection transmission distance for subsequent modeling and analysis. Then, by performing aggregate statistical analysis on the cleaned crowd flow data, and analyzing data such as crowd movement trajectory, crowd density, and crowd flow direction, crowd gathering data can be obtained. These data may include information such as the location, time, and scale of crowd gathering, which is of great significance for understanding crowd flow patterns and clustered infection risks. Next, potential associations are mined and analyzed between infection transmission parameter data and population cluster data. Data mining techniques are used to identify possible correlations and interactions between transmission parameters and population clusters. Key characteristics of transmission parameters and population clusters are extracted from these characteristics and used to construct a population infection transmission model. Machine learning methods, statistical analysis, and other techniques can be used to build this model. Furthermore, parameter estimation and optimization are performed on the constructed population infection transmission model to find parameter values ​​that best align the model's predictions with the actual observed data, thereby improving the model's fitting performance. Parameter optimization yields an adjusted and improved infection transmission optimization model that better aligns with the actual data. Furthermore, the infection transmission optimization model is used to perform infection prediction on infection transmission parameter data and population cluster data. The model's predictive capabilities allow inferences about infection transmission patterns based on the available data and model parameters. This can help predict and understand infection spread trends, the effectiveness of control measures, and provide data support for guiding population behavior and policies. Finally, by conducting prevention and control decision support analysis on infection transmission pattern data, we can analyze infection transmission trends, influencing factors, etc. to obtain analysis results related to prevention and control decisions. These results can help decision makers understand the risks and impacts of infection transmission and formulate corresponding prevention and control strategies. Based on the analysis results, a decision support system is constructed to assist in the process of executing infection prevention and control decisions, thereby achieving the reliability, personalization and accuracy of the decision support system construction to support the formulation and implementation of infection prevention and control decisions in real time.

[0048] Preferably, step S1 includes the following steps:

[0049] Step S11: Obtain infection prevention and control related data;

[0050] Step S12: Repeatedly screen and process the infection prevention and control related data to obtain infection prevention and control screening data;

[0051] Step S13: using the abnormality degree calculation formula to perform abnormality detection calculation on the infection prevention and control screening data to obtain the infection prevention and control abnormal value;

[0052] Step S14: performing abnormal filtering on the infection prevention and control screening data according to the infection prevention and control abnormal values ​​to obtain prevention and control abnormality filtered data;

[0053] Step S15: Perform data interpolation processing on the prevention and control abnormality filtered data to obtain prevention and control interpolation filling data;

[0054] Step S16: statistically integrate the prevention and control interpolation and filling data to obtain prevention and control integrated data;

[0055] Step S17: Standardize the prevention and control integrated data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data.

[0056] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in FIG. 1 , in this embodiment, step S1 includes the following steps:

[0057] Step S11: Obtain infection prevention and control related data;

[0058] The embodiment of the present invention obtains infection epidemic data, population flow data, medical data and other data related to infection prevention and control from data sources such as public health departments, medical institutions, and research institutions, and ultimately obtains infection prevention and control related data.

[0059] Step S12: Repeatedly screen and process the infection prevention and control related data to obtain infection prevention and control screening data;

[0060] The embodiment of the present invention repeatedly screens the infection prevention and control related data to remove duplicate records or data in the infection prevention and control related data to ensure the accuracy and uniqueness of the data, and ultimately obtains infection prevention and control screening data.

[0061] Step S13: using the abnormality degree calculation formula to perform abnormality detection calculation on the infection prevention and control screening data to obtain the infection prevention and control abnormal value;

[0062] The embodiment of the present invention constructs a suitable abnormality degree calculation formula by combining the observation time parameter of the abnormality detection calculation, the characteristic index value of the infection prevention and control screening data, the amplitude control parameter, the mean parameter, the standard deviation parameter, the abnormality degree curve adjustment parameter, the abnormality degree curve threshold parameter, the abnormal amplitude control parameter and the mean parameter of the abnormal distribution peak and related parameters to perform abnormality detection calculation on the infection prevention and control screening data, so as to identify possible abnormal situations or outliers in the infection prevention and control screening data, and finally obtain the infection prevention and control abnormal value.

[0063] Step S14: performing abnormal filtering on the infection prevention and control screening data according to the infection prevention and control abnormal values ​​to obtain prevention and control abnormality filtered data;

[0064] The embodiment of the present invention compares and judges the calculated infection prevention and control anomaly value according to a preset anomaly threshold to determine whether the corresponding infection prevention and control screening data has an anomaly. When the calculated infection prevention and control anomaly value is greater than or equal to the preset anomaly threshold, it means that the infection prevention and control screening data corresponding to the infection prevention and control anomaly value has an anomaly, and the anomaly should be filtered out or corrected. When the calculated infection prevention and control anomaly value is less than the preset anomaly threshold, it means that the infection prevention and control screening data corresponding to the infection prevention and control anomaly value does not have an anomaly, and no anomaly filtering or correction is performed, and finally the prevention and control anomaly filtered data is obtained.

[0065] Step S15: Perform data interpolation processing on the prevention and control abnormality filtered data to obtain prevention and control interpolation filling data;

[0066] The embodiment of the present invention interpolates the missing values ​​in the control anomaly filtering data by using methods such as linear interpolation, polynomial interpolation, and Kriging interpolation to fill the missing data in the control anomaly filtering data, and finally obtains the control interpolation filling data.

[0067] Step S16: statistically integrate the prevention and control interpolation and filling data to obtain prevention and control integrated data;

[0068] The embodiment of the present invention performs statistical analysis and integration processing on the interpolated and filled data, including operations such as aggregation, summary, calculation of statistical indicators or generation of new variables, so as to comprehensively analyze the prevention and control interpolation and filling data from multiple dimensions and angles, extract important features and indicators, and finally obtain prevention and control integrated data.

[0069] Step S17: Standardize the prevention and control integrated data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data.

[0070] In an embodiment of the present invention, the integrated control data is standardized using a minimum-maximum normalization method to eliminate the dimensional differences between different data indicators, and ultimately obtain infection control cleaned data, where the infection control cleaned data includes infection transmission cleaned data and crowd flow cleaned data.

[0071] The present invention first collects data related to infection prevention and control, which may include infection epidemic data, population flow data, medical data, etc., to obtain comprehensive and accurate infection prevention and control data, which can provide basic data for subsequent analysis and prediction. By screening the duplicate data in the acquired infection prevention and control related data, duplicate data entries in the infection prevention and control related data can be eliminated to ensure the accuracy and consistency of the data. At the same time, by using a suitable abnormality degree calculation formula to perform anomaly detection calculation on the infection prevention and control screening data, the calculation formula can measure the abnormality degree of the data by calculating abnormal values ​​to identify possible abnormal values, which may contain potential abnormal situations or outliers. Secondly, by performing anomaly filtering processing on the infection prevention and control screening data based on the calculated infection prevention and control abnormal values, abnormal data corresponding to the infection prevention and control abnormal values ​​can be eliminated to reduce the interference of abnormal data, thereby improving the accuracy and reliability of the data. Then, by performing data interpolation processing on the prevention and control data after abnormality filtering, data interpolation is a method of filling missing values. By estimating and filling missing values ​​based on the patterns and rules of existing data, the data set is made more complete. Through data interpolation, the existing prevention and control data can be used to infer missing data, thereby obtaining a prevention and control data set that has been interpolated and filled. Next, we statistically analyze and integrate the interpolated control data to comprehensively analyze the control data from multiple dimensions and perspectives, extracting key features and indicators to form more representative and comprehensive control data. Finally, we standardize the control data to convert it to a unified standard scale for subsequent analysis and comparison. Standardization eliminates dimensional differences between data, making different indicators comparable. Ultimately, we obtain cleaned infection control data, including cleaned infection transmission data and cleaned population flow data. After cleaning and processing, these data are more usable and analyzable.

[0072] Preferably, the abnormality degree calculation formula in step S13 is specifically:

[0073]

[0074] Where E is the infection prevention and control anomaly value, T is the observation time range length for anomaly detection calculation, t is the time parameter for anomaly detection calculation, x(t) is the characteristic index value of the infection prevention and control screening data, α is the amplitude control parameter of the infection prevention and control screening data, exp is the exponential function, μ is the mean parameter of the infection prevention and control screening data, σ is the standard deviation parameter of the infection prevention and control screening data, γ is the anomaly amplitude control parameter of the anomaly distribution peak in the infection prevention and control screening data, β is the mean parameter of the anomaly distribution peak in the infection prevention and control screening data, ρ is the anomaly degree curve adjustment parameter of the infection prevention and control screening data, η is the anomaly degree curve threshold parameter of the infection prevention and control screening data, and ε is the correction value of the infection prevention and control anomaly value.

[0075] The present invention constructs a formula for calculating the degree of abnormality for performing anomaly detection on infection prevention and control screening data. This formula uses an amplitude control parameter to control the amplitude influence of the infection prevention and control screening data in the anomaly detection calculation. By adjusting the value of this parameter, the sensitivity of the anomaly detection results to the data amplitude can be adjusted. The distribution characteristics of the infection prevention and control screening data are described by the mean parameter and the standard deviation parameter. Furthermore, by considering the mean and standard deviation of the data, anomalies relative to the normal range can be identified. The peak position and amplitude control of the anomaly distribution are described by using the anomaly amplitude control parameter and the mean parameter of the anomaly distribution peak. By adjusting the values ​​of these parameters, the position and amplitude of the anomaly peak can be determined, further identifying anomalies. The shape and threshold of the anomaly degree curve can also be adjusted by using an anomaly degree curve adjustment parameter and a threshold parameter. By adjusting the values ​​of these parameters, the anomaly degree calculation process can be flexibly controlled, thereby improving the accuracy of anomaly detection. In addition, correction values ​​can be used to fine-tune the calculation results to suit specific anomaly detection processes. This formula fully considers the infection control anomaly value E, the observation time range length T for anomaly detection calculation, the time parameter t for anomaly detection calculation, the characteristic index value x(t) of the infection control and prevention screening data, the amplitude control parameter α of the infection control and prevention screening data, the exponential function exp, the mean parameter μ of the infection control and prevention screening data, the standard deviation parameter σ of the infection control and prevention screening data, the anomaly amplitude control parameter γ of the anomaly distribution peak in the infection control and prevention screening data, the mean parameter β of the anomaly distribution peak in the infection control and prevention screening data, the anomaly degree curve adjustment parameter ρ of the infection control and prevention screening data, the anomaly degree curve threshold parameter η of the infection control and prevention screening data, and the correction value ε of the infection control and prevention anomaly value. Based on the mutual correlation between the infection control and prevention anomaly value E and the above parameters, a functional relationship is formed:

[0076]

[0077] This formula can realize the anomaly detection calculation process of infection prevention and control screening data. At the same time, by introducing the correction value ε of the infection prevention and control anomaly value, it can be adjusted according to actual conditions, thereby improving the accuracy and stability of the anomaly degree calculation formula.

[0078] Preferably, step S2 includes the following steps:

[0079] Step S21: performing path trend analysis on the infection propagation cleansing data to obtain an infection propagation trend path;

[0080] Step S22: Calculate the infection propagation rate of the cleaned infection propagation data to obtain the infection propagation rate;

[0081] Step S23: performing propagation distance mining analysis based on the infection propagation trend path and the infection propagation rate to obtain the infection propagation distance;

[0082] Step S24: Calculate the infection rate of the infection propagation cleansed data to obtain the infection rate;

[0083] Step S25: performing a latent calculation on the infection transmission cleaned data using a transmission latent rate calculation formula according to the infection transmission rate, the infection transmission distance, and the infection rate to obtain the infection transmission latent rate;

[0084] The formula for calculating the transmission latency rate is as follows:

[0085]

[0086] Where Q(τ) is the infection transmission latent rate within time τ, τ is the time range parameter for latent calculation, and τ ′ is the time variation parameter of the latent calculation, δ is the infection rate, I(τ-τ ′ ) is the time τ-τ ′ The infection spread rate value within, N is the total number of individuals infected in the infection spread cleaning data, A ij The adjacency matrix of the interpersonal contact network composed of individual i and individual j in the infection spread cleaning data, |A ij | is the modulus of the adjacency matrix, i.e., the infection spread distance value of the interpersonal contact network, is the infection recovery rate, f(τ) is the infection incubation period distribution function, S(τ-τ ′ ) is the time τ-τ ′ The number of susceptible people in ,∈ is the correction value of the latent rate of infection transmission;

[0087] Step S26: Integrate and analyze the infection spread trend path, infection spread rate, infection spread distance, infection rate, and infection spread latency rate to obtain infection spread parameter data.

[0088] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 Detailed step flow diagram of step S2 in the embodiment, step S2 includes the following steps:

[0089] Step S21: performing path trend analysis on the infection propagation cleansing data to obtain an infection propagation trend path;

[0090] The embodiment of the present invention uses geographic information system technology to perform path trend analysis on infection propagation cleaned data to detect the propagation path between each infection propagation individual, and analyzes the infection propagation trend path based on the propagation path, and finally obtains the infection propagation trend path.

[0091] Step S22: Calculate the infection propagation rate of the cleaned infection propagation data to obtain the infection propagation rate;

[0092] The embodiment of the present invention calculates the infection propagation rate in different time periods based on the time information and propagation relationship in the infection propagation cleaning data, and finally obtains the infection propagation rate.

[0093] Step S23: performing propagation distance mining analysis based on the infection propagation trend path and the infection propagation rate to obtain the infection propagation distance;

[0094] The embodiment of the present invention analyzes the spread of infection in different paths by using the infection propagation trend path and the infection propagation rate, and calculates the spatial distance between each infected individual based on the analyzed propagation situation, and finally obtains the infection propagation distance.

[0095] Step S24: Calculate the infection rate of the infection propagation cleansed data to obtain the infection rate;

[0096] The embodiment of the present invention calculates the infection frequency according to the number of infected individuals and the total population in the infection propagation cleaned data, and finally obtains the infection rate.

[0097] Step S25: performing a latent calculation on the infection transmission cleaned data using a transmission latent rate calculation formula according to the infection transmission rate, the infection transmission distance, and the infection rate to obtain the infection transmission latent rate;

[0098] The embodiment of the present invention constructs a suitable transmission latent rate calculation formula by combining the time parameters of latent calculation, infection rate, infection transmission rate value, the total number of individuals in the infected population, the adjacency matrix of the interpersonal contact network, the infection transmission distance value, the infection recovery rate, the infection incubation period distribution function, the number of susceptible people and related parameters to perform latent calculation on the infection transmission cleaning data to analyze and predict the potential development trend of the infection, and finally obtain the infection transmission latent rate.

[0099] The formula for calculating the transmission latency rate is as follows:

[0100]

[0101] Where Q(τ) is the infection transmission latent rate within time τ, τ is the time range parameter for latent calculation, and τ ′ is the time variation parameter of the latent calculation, δ is the infection rate, I(τ-τ ′ ) is the time τ-τ ′ The infection spread rate value within, N is the total number of individuals infected in the infection spread cleaning data, A ij The adjacency matrix of the interpersonal contact network composed of individual i and individual j in the infection spread cleaning data, |A ij | is the modulus of the adjacency matrix, i.e., the infection spread distance value of the interpersonal contact network, is the infection recovery rate, f(τ) is the infection incubation period distribution function, S(τ-τ ′ ) is the time τ-τ ′ The number of susceptible people in ,∈ is the correction value of the latent rate of infection transmission;

[0102] The present invention constructs a formula for calculating the transmission latency rate, which is used to calculate the latency of infection transmission cleansed data based on the infection transmission rate, infection transmission distance, and infection rate. By setting appropriate time ranges and time-varying parameters, this formula can analyze and estimate the incubation period of infection transmission. By taking into account the occurrence of infection, the latency calculation results can be adjusted to reflect the actual infection situation. By considering the variation in the infection transmission rate, the transmission latency rate during the incubation period can be more accurately calculated. By using the adjacency matrix of the interpersonal contact network to describe the interpersonal relationships and contact networks in the infection transmission data, and considering the adjacency matrix of the interpersonal contact network and the number of infected people, the impact of the infection transmission distance and the transmission latency rate can be analyzed. By representing the infection transmission distance value of the interpersonal contact network using the modulus of the adjacency matrix, the impact of the infection transmission distance on the transmission latency rate can be considered, thereby better understanding the characteristics of the incubation period during the transmission process. Furthermore, by combining the infection recovery rate with the calculation of the transmission latency rate during the incubation period, the characteristics of the transmission process can be further analyzed. By using the infection latency distribution function to consider the distribution of the infection latency, the transmission latency rate can be more accurately estimated. In addition, the number of susceptible individuals is also considered to further analyze the latent situation of infection transmission. By comprehensively considering the above parameters and the characteristics of infection transmission cleaning data, the transmission latent rate calculation formula can calculate the infection transmission latent rate within a given time range, which is used to analyze the transmission process and characteristics of the infection. This helps to understand the degree and speed of transmission during the incubation period and provide support for the formulation of corresponding infection prevention and control measures and strategies. This calculation formula fully considers the infection transmission latent rate Q(τ) within time τ, the time range parameter τ of the latent calculation, and the time change parameter τ of the latent calculation. ′ , infection rate δ, at time τ-τ ′ The infection spread rate value I(τ-τ ′ ), the total number of infected individuals N in the infection spread clean data, the adjacency matrix A of the interpersonal contact network composed of individuals i and j in the infection spread clean data ij , the modulus of the adjacency matrix |A ij |, that is, the infection transmission distance value of the interpersonal contact network, the infection recovery rate ζ, the infection incubation period distribution function f(τ), at time τ-τ ′ The number of susceptible people S(τ-τ ′ ), the correction value of the infection transmission latent rate ∈, according to the mutual correlation between the infection transmission latent rate Q(τ) within time τ and the above parameters, constitutes a functional relationship:

[0103]

[0104] This formula can realize the latent calculation of infection transmission cleaning data based on the infection transmission rate, infection transmission distance and infection rate. At the same time, by introducing the correction value ∈ of the infection transmission latent rate, it can be adjusted according to actual conditions, thereby improving the accuracy and applicability of the transmission latent rate calculation formula.

[0105] Step S26: Integrate and analyze the infection spread trend path, infection spread rate, infection spread distance, infection rate, and infection spread latency rate to obtain infection spread parameter data.

[0106] The embodiment of the present invention merges and organizes parameter data such as infection transmission trend path, infection transmission rate, infection transmission distance, infection rate and infection transmission latent rate, and performs operations such as data format conversion, outlier processing and missing value filling to ensure data consistency and accuracy, and finally obtains infection transmission parameter data.

[0107] The present invention first analyzes cleaned infection spread data to reveal the temporal and spatial spread trends of the infection. By studying the time series and geographic location of infected cases, the paths and trends of infection spread can be identified, helping to understand the patterns of infection spread, its scope, and possible transmission pathways. Furthermore, by calculating the transmission rate based on the cleaned infection spread data, the infection spread rate can be calculated based on the time series data. The transmission rate represents the change in the number of infected people per unit time and can be used to assess the speed and extent of infection spread. By statistically analyzing the cleaned infection spread data, the infection spread rate can be calculated and key indicators of infection development can be derived. Secondly, by combining the trend path and transmission rate of infection spread, transmission distance analysis is performed. Transmission distance represents the distance an infection spreads from a source to other individuals and is an important indicator of the infection's spread. By analyzing the path and rate, the transmission distance of the infection can be inferred, thereby helping to predict the spread and transmission pathways of the infection. Finally, the infection rate is calculated based on the cleaned infection spread data. The infection rate is the ratio of the number of infected people to the total population during a specific time period, which is used to assess the prevalence and risk level of the infection. By statistically analyzing and calculating the cleaned infection transmission data, we can derive the infection rate, thereby better understanding the spread of infection and the degree of risk. Next, by combining the infection transmission rate, distance, and infection rate, we use an appropriate formula to calculate the latent rate of infection transmission. The latent rate refers to the time delay between infection and the onset of clinical symptoms and is a key parameter for infection control and prediction. By comprehensively considering factors such as transmission rate, distance, and infection rate, we can use a calculation formula to estimate the latent rate, thereby analyzing and predicting infection trends. Finally, we integrate and analyze parameters such as the infection transmission trend path, transmission rate, transmission distance, infection rate, and latent rate. This comprehensive analysis provides a more comprehensive and detailed picture of infection transmission and allows for more accurate prediction and assessment of infection progression. The results of this integrated analysis can provide important reference for decision makers and the public regarding infection control and prevention measures, and provide foundational data for the subsequent development of a decision support system.

[0108] Preferably, step S3 includes the following steps:

[0109] Step S31: using kernel density estimation technology to perform density estimation processing on the crowd flow cleaned data to obtain the crowd flow estimated density;

[0110] The embodiment of the present invention performs density estimation on crowd flow cleaning data by using a kernel density estimation technology composed of a Gaussian kernel function, converts the crowd flow cleaning data into point density in space, generates a crowd flow density map based on the converted density results, so as to intuitively display the crowd flow density in different areas, and estimates the density around each data point in the crowd flow density map to infer the flow density of the entire area, and finally obtains the estimated crowd flow density.

[0111] Step S32: analyzing the crowd flow cleansing data based on the crowd flow estimated density to obtain the crowd flow behavior pattern;

[0112] The embodiment of the present invention analyzes the flow behavior patterns of crowd flow cleaning data by estimating the crowd flow estimated density, so as to analyze and study the flow patterns and trends of the crowd between different areas, and presents the patterns and trends of the crowd flow by using streamline diagrams, and finally obtains the crowd flow behavior patterns.

[0113] Step S33: Calculate the crowd flow cleaning data using the crowd flow frequency calculation formula to obtain the crowd flow frequency;

[0114] The embodiment of the present invention constructs a suitable crowd flow frequency calculation formula by combining the observation time variable, crowd segmentation, population size, average crowd flow distance, time attenuation control parameter of crowd flow, stability control parameter, shape adjustment scaling factor and related parameters of crowd flow calculation to perform flow calculation on the crowd flow cleaning data to calculate the flow frequency between different areas and finally obtain the crowd flow frequency.

[0115] Step S34: performing regional flow distribution analysis on the crowd flow cleansing data to obtain regional crowd flow conditions;

[0116] The embodiment of the present invention divides the crowd flow cleaning data into different areas or grids to facilitate the analysis of crowd flow, and counts the number of people entering and leaving each area or the crowd flow volume, analyzes the flow patterns and trends between different areas, and finally obtains the regional crowd flow situation.

[0117] Step S35: Perform spatial aggregation analysis based on crowd flow behavior patterns, crowd flow frequency, and regional crowd flow conditions to obtain crowd aggregation data.

[0118] The embodiments of the present invention use spatial statistical methods such as cluster analysis or hotspot analysis to analyze and infer crowd aggregation by combining the patterns of crowd flow behavior, crowd flow frequency, and regional crowd flow conditions, so as to identify crowd aggregation patterns in different areas, determine the degree of aggregation and distribution characteristics, and draw aggregation hotspot maps to intuitively understand the crowd aggregation conditions and distribution patterns, and finally obtain crowd aggregation data.

[0119] The present invention first processes cleaned crowd flow data using kernel density estimation, a nonparametric statistical method used to estimate density distribution within a dataset. By converting crowd flow data into spatial point density, kernel density estimation can be applied within geographic information systems (GIS), revealing dense areas and hotspots of crowd flow. Secondly, the crowd flow data is analyzed for behavioral patterns based on the estimated crowd flow density. By observing the spatial distribution and flow density of crowd flow, trends and patterns of crowd flow can be identified. For example, the main starting and ending points of crowd flow, as well as the direction and intensity of flow, can be analyzed. These results are important for understanding crowd flow and making infection prevention and control decisions. Then, a suitable crowd flow frequency calculation formula is used to calculate the flow of cleaned crowd flow data. The calculated crowd flow frequency represents the amount of population movement within a specific time frame, typically expressed as person-times per time unit. By statistically analyzing information such as crowd flow segmentation, population size, and average crowd flow distance, the crowd flow frequency can be derived. This can be used to optimize infection flow prediction, providing basic data for the subsequent construction of decision support systems. Next, by performing regional flow distribution analysis on the cleaned crowd flow data and associating the crowd flow data with geographical regions, we can understand the flow of people between different regions, including the number of people flowing into and out of different regions and the direction of flow. This helps to understand the population flow pattern, population flow distribution characteristics, and possible population flow bottlenecks in cities or regions during the infection prevention and control period. Finally, by combining the behavioral patterns, flow frequency, and regional flow distribution of crowd flow to conduct spatial aggregation analysis, we can count and analyze the geographical areas and spatial scope of crowd aggregation, and obtain data on densely populated areas, community hotspots, and crowd flow gathering centers during the infection prevention and control period. This will help the subsequent construction of the infection prevention and control decision support system to better consider the aggregation of people and provide more accurate decision support.

[0120] Preferably, the crowd flow frequency calculation formula in step S33 is specifically:

[0121]

[0122] Where F is the crowd flow frequency, T′ is the observation period for crowd flow calculation, t ′ is the observation time variable for crowd flow calculation, n is the total number of crowd segments for crowd flow cleaning data, and λ k The population size of the kth crowd segment in the crowd flow cleaning data, D k is the average flow distance of the kth crowd segment in the crowd flow cleaning data, θ1 is the time attenuation control parameter of the crowd flow, θ2 is the stability control parameter of the crowd flow, The scaling factor is adjusted for the shape of the crowd flow, and ξ is the correction value for the frequency of the crowd flow.

[0123] The present invention constructs a crowd flow frequency calculation formula for performing flow calculation on crowd flow cleansing data. This formula calculates the flow intensity of a crowd under different time segmentation variables to understand the frequency of crowd flow. By selecting an appropriate observation period, the flow frequency of a crowd can be more accurately estimated. By using the observation time variable in the crowd flow calculation to control the change in flow frequency over time, the temporal characteristics of crowd flow can be analyzed. By considering the population size and flow distance of different crowd segments, the formation mechanism of crowd flow frequency can be better understood. By using the population size parameter to consider the impact of population size on flow frequency of different crowd segments, the magnitude and distribution of crowd flow frequency can be evaluated. By also considering the average flow distance of crowd segments, the relationship between crowd flow frequency and spatial distance can be analyzed. The time decay control parameter can be used to adjust the decay rate of crowd flow over time, and the stability control parameter can be used to control the stability of crowd flow. By adjusting the stability control parameter, the stability and volatility of crowd flow can be analyzed. In addition, by incorporating the shape adjustment scaling factor into the calculation formula, the characteristics of crowd flow can be better simulated and described. By comprehensively considering the above parameters and the characteristics of crowd flow cleaning data, the crowd flow frequency calculation formula can calculate the flow frequency of the crowd in a specific observation period, which is used to analyze the flow behavior patterns and frequency distribution of the crowd. This helps to understand the flow activities of the crowd during the infection prevention and control period, predict the behavior trends of the crowd, and plan prevention and control decisions and measures related to crowd flow. This formula fully considers the crowd flow frequency F and the observation period T of the crowd flow calculation. ′ , the observation time variable t for crowd flow calculation ′ , the total number of crowd segments n in the crowd flow cleaning data, the population size λ of the kth crowd segment in the crowd flow cleaning data k , the average flow distance D of the kth crowd segment in the crowd flow cleaning data k , the time decay control parameter θ1 of the crowd flow, the stability control parameter θ2 of the crowd flow, the shape adjustment scaling factor of the crowd flow The correction value of crowd flow frequency ξ forms a functional relationship based on the mutual correlation between crowd flow frequency F and the above parameters The formula can realize the flow calculation process of crowd flow cleaning data. At the same time, by introducing the correction value ξ of the crowd flow frequency, it can be adjusted according to actual conditions, thereby improving the accuracy and applicability of the crowd flow frequency calculation formula.

[0124] Preferably, step S2 includes the following steps:

[0125] Step S41: performing a first association mining analysis on the infection propagation parameter data and the crowd gathering situation data to obtain propagation parameter characteristics and crowd gathering characteristics;

[0126] The embodiment of the present invention uses an association analysis method to perform the first association mining on the infection propagation parameter data and the crowd gathering data to mine the initial correlation and association rules between the infection propagation parameter data and the crowd gathering data, and finally obtains the propagation parameter characteristics and the crowd gathering characteristics.

[0127] Step S42: performing potential pattern recognition analysis on the infection propagation parameter data and the crowd gathering data to obtain a potential pattern of the propagation parameters and a potential pattern of the crowd gathering;

[0128] The embodiment of the present invention performs potential pattern recognition analysis on infection propagation parameter data and crowd gathering data by using statistical analysis, Pearson correlation coefficient calculation method, topic model, cluster analysis and other methods to identify potential patterns and structures in the infection propagation parameter data and crowd gathering data, and finally obtain the potential pattern of propagation parameters and the potential pattern of crowd gathering.

[0129] Step S43: performing a second association mining analysis on the propagation parameter features using the propagation parameter potential pattern to obtain key features of the propagation parameters;

[0130] The embodiment of the present invention uses the features and associations in the potential patterns of the propagation parameters to perform deep association mining analysis on the propagation parameter features again, further discovers the association rules and key features between the propagation parameters, and finally obtains the key features of the propagation parameters.

[0131] Step S44: performing a second association mining analysis on the crowd gathering characteristics using the crowd gathering potential pattern to obtain the crowd gathering key characteristics;

[0132] The embodiment of the present invention uses the features and associations in the potential pattern of crowd gathering to conduct deep association mining analysis on the crowd gathering features again, further discovers the association rules and key features between crowd gathering situations, and finally obtains the key features of crowd gathering.

[0133] Step S45: Construct a crowd infection propagation model based on the key features of the propagation parameters and the key features of the crowd gathering.

[0134] In this embodiment of the present invention, the key features of propagation parameters and key features of crowd aggregation are divided into a training set, a validation set, and a test set according to a preset division ratio of 7:2:1. Then, the SEIR model is used for model training, verification, and testing based on the division results to construct a model that can accurately simulate and predict the spread of infection in the crowd, and finally a crowd infection propagation model is obtained.

[0135] The present invention first performs association mining analysis on infection transmission parameter data and crowd aggregation data. By statistically analyzing these two types of data, transmission parameter characteristics and crowd aggregation characteristics can be obtained. Transmission parameter characteristics may include infection rate, transmission incubation period, transmission distance, etc., while crowd aggregation characteristics may include densely populated areas, social venues, transportation hubs, etc. These characteristics can help better understand the laws of infection transmission and the impact of crowd aggregation on infection transmission. Secondly, by performing potential pattern recognition analysis on the infection transmission parameter data and crowd aggregation data, the hidden patterns and association patterns behind the data can be revealed. By identifying and extracting potential patterns in the transmission parameter data and crowd aggregation data, transmission parameter potential patterns and crowd aggregation potential patterns can be obtained. These potential patterns are important for understanding the infection transmission mechanism and the impact of crowd behavior patterns on infection transmission. Then, by using the transmission parameter potential patterns to perform a second association mining analysis on the transmission parameter characteristics, further data mining and association analysis can identify key characteristics of the transmission parameters related to infection transmission. These key characteristics may be parameters related to transmission speed, transmission range, and transmission path, which helps to better understand the infection transmission process and influencing factors. At the same time, a second correlation mining analysis of crowd cluster characteristics is conducted using latent crowd cluster patterns. Further data mining and correlation analysis can identify key characteristics of crowd clusters that are relevant to infection spread. These characteristics may be related to crowd density, interpersonal connections, and venues, helping to better understand the role and impact of crowd clusters on infection spread. Finally, a crowd infection spread model is constructed based on the mined key characteristics of transmission parameters and crowd clusters. This model can be used to simulate and predict the spread of infection in a population using methods such as statistics, mathematical modeling, or machine learning. This model can be used to assess the impact of different transmission parameters and crowd cluster conditions on the speed and range of transmission, thereby guiding the development and implementation of infection prevention and control measures and providing data support for the subsequent development of a decision support system.

[0136] Preferably, step S42 includes the following steps:

[0137] Step S421: performing data distribution exploration analysis on the infection propagation parameter data and the crowd gathering situation data to obtain propagation parameter distribution data and crowd gathering distribution data;

[0138] The embodiments of the present invention explore and analyze the infection transmission parameter data and the crowd gathering data by using statistical analysis and visualization methods, such as by calculating descriptive statistics, drawing histograms, box plots, scatter plots, etc., to reveal the distribution characteristics of the infection transmission parameter data and the crowd gathering data, and finally obtain the transmission parameter distribution data and the crowd gathering distribution data.

[0139] Step S422: performing correlation calculations on the propagation parameter distribution data and the crowd gathering distribution data to obtain propagation parameter correlation data and crowd gathering correlation data;

[0140] The embodiment of the present invention performs correlation calculations on the propagation parameter distribution data and the crowd gathering distribution data by using the Pearson correlation coefficient calculation method to calculate the correlation between the propagation parameter distribution data and the crowd gathering distribution data, and visualizes the correlation between the propagation parameter distribution data and the crowd gathering distribution data by drawing correlation visualization charts such as correlation matrix, heat map, scatter plot, etc., so as to better understand the correlation between the data and finally obtain the propagation parameter related data and the crowd gathering related data.

[0141] Step S423: performing correlation mining analysis on the propagation parameter related data and the crowd gathering related data to obtain the propagation parameter correlation relationship and the crowd gathering correlation relationship;

[0142] The embodiment of the present invention mines association relationships on propagation parameter related data and crowd gathering related data by using association rule mining algorithms such as the Apriori algorithm and the FP-Growth algorithm to discover frequent association patterns between propagation parameters or between aggregation situations. According to the association patterns obtained by mining, the propagation parameter association patterns and crowd gathering association patterns are visualized by drawing association rule graphs, network diagrams, etc., so as to better understand the structure and pattern of the association relationships, and finally obtain the propagation parameter association relationships and crowd gathering association relationships.

[0143] Step S424: Perform potential mining analysis on the correlation between propagation parameters and the correlation between crowd gatherings to obtain potential patterns of propagation parameters and potential patterns of crowd gatherings.

[0144] The embodiments of the present invention perform potential pattern mining on the correlation between propagation parameters and the correlation between crowd gathering by using potential pattern mining algorithms such as topic models, cluster analysis, and factor analysis to discover the potential patterns and structures hidden in the data. By visualizing and interpreting the mining results, the discovered potential patterns are presented in the form of charts or text to provide in-depth understanding and analysis, and finally the potential patterns of propagation parameters and the potential patterns of crowd gathering are obtained.

[0145] The present invention first conducts data distribution exploration and analysis on infection transmission parameter data and crowd aggregation data, including statistical description and visualization analysis of the data, to understand the distribution of transmission parameters and crowd aggregation. By analyzing the data distribution, useful information about transmission parameters and crowd aggregation characteristics can be obtained. For example, the distribution range of infection rate, the distribution pattern of transmission distance, and the distribution of crowd density can be determined. Then, by performing correlation calculations on the transmission parameter distribution data and the crowd aggregation distribution data to calculate the correlation between transmission parameter distributions or between crowd aggregation distributions, data related to transmission parameters and crowd aggregation can be obtained. Correlation analysis can reveal the degree of correlation between transmission parameters and crowd aggregation, for example, whether there is a correlation between infection rate and crowd density, or whether there is a correlation between transmission distance and interpersonal connections. Next, by using correlation mining analysis methods, further analysis of the transmission parameter data and crowd aggregation data can reveal potential correlations between transmission parameters or crowd aggregation characteristics. By mining correlations, the mutual influence between infection transmission parameters or the relationship between crowd aggregation characteristics can be discovered, thereby better understanding the impact of transmission mechanisms and human behavior patterns on infection transmission. Finally, potential mining analysis of the correlations between transmission parameters and crowd aggregation can further reveal patterns and regularities hidden in the data. Potential mining analysis methods, such as cluster analysis, principal component analysis, and association rule mining, can help identify potential patterns in transmission parameters and crowd aggregation. These potential patterns can provide deeper information, contributing to our understanding of infection transmission mechanisms and crowd aggregation behaviors.

[0146] Preferably, step S5 includes the following steps:

[0147] Step S51: performing parameter estimation on the population infection propagation model to obtain model parameter estimation values;

[0148] The embodiment of the present invention estimates the parameters of the constructed population infection and propagation model by using parameter estimation methods such as maximum likelihood estimation and Bayesian inference, so that the model can fit the observed data, and finds the parameter values ​​that make the model's prediction results most consistent with the actual observed data, and finally obtains the model parameter estimation values.

[0149] Step S52: using the estimated values ​​of the model parameters to estimate and align the population infection propagation model to obtain an infection propagation estimation model;

[0150] In an embodiment of the present invention, the population infection propagation model is registered using the estimated model parameter values, and the model parameters are adjusted through registration to minimize the difference between the model prediction estimate and the actual observation data, thereby achieving registration between the model and the observation data, and finally obtaining an infection propagation estimation model.

[0151] Step S53: Calculating the error loss of the infection spread estimation model using the model estimation error loss calculation formula to obtain a model estimation error loss value;

[0152] The embodiment of the present invention constructs a suitable model estimation error loss calculation formula by combining the infection propagation estimation model parameters, parameter space, estimated values, real observation data, integral parameter variables, weight functions, probability density functions and related parameters to perform error loss calculation on the infection propagation estimation model to calculate the error loss value between the model prediction and the real observation data, and finally obtain the model estimation error loss value.

[0153] Among them, the model estimation error loss calculation formula is as follows:

[0154]

[0155] Where, is the model estimation error loss value, is the infection spread estimation model parameter, Ω is the parameter space of the infection spread estimation model parameter, is the estimated value of the infection spread estimation model, R is the real observation data, X is the integral parameter variable in the parameter space, w(X) is the weight function, p(X) is the probability density function, and ι is the correction value of the model estimation error loss value;

[0156] The present invention constructs a model estimation error loss calculation formula for calculating the error loss of an infection spread estimation model. This model estimation error loss calculation formula evaluates the accuracy and goodness of fit of the estimated model by calculating the degree of difference between the infection spread model estimate and the actual observed data, thereby identifying the model parameters that best match the actual data. The infection spread estimation model parameters are used to describe various factors influencing the spread process. By estimating and adjusting the parameters, model performance can be optimized and prediction results improved. By defining a parameter space, the parameter search range can be limited, thereby improving the efficiency of parameter estimation. The estimated values ​​are used to represent the prediction results of the infection spread estimation model under given parameters to evaluate the model's goodness of fit. Furthermore, by using actual observed infection spread data to represent the actual observed infection spread and comparing them with the estimated model's prediction results, the deviation between the model and the actual data can be measured. By using a weighting function to adjust the contribution of different parameter values ​​to the error loss, the impact of different parameter values ​​can be emphasized or weakened. Furthermore, by using a probability density function to describe the distribution of parameters within the parameter space, the values ​​within the parameter space can be appropriately weighted and normalized. In addition, the error loss value can be fine-tuned by using the correction value of the error loss value to consider the impact of other factors on the error or adjust the evaluation criteria. By introducing the correction value, the calculation results of the error loss value can be better controlled. By comprehensively considering the above parameters to construct the model estimation error loss calculation formula for error loss calculation, it is possible to optimize the parameters of the infection propagation model by minimizing the model estimation error loss value, thereby improving the model's fit and prediction accuracy, and providing a reliable basis for further analysis and prediction of infection propagation laws. This formula fully takes into account the model estimation error loss value. Infection spread estimation model parameters The parameter space Ω of the infection spread estimation model parameters, the estimated value of the infection spread estimation model Real observation data R, integral parameter variable X in parameter space, weight function w(X), probability density function p(X), correction value ι of model estimation error loss value, and error loss value estimated by model The mutual correlation between the above parameters constitutes a functional relationship This formula can realize the error loss calculation process of the infection spread estimation model. At the same time, by introducing the correction value ι of the model estimation error loss value, it can be adjusted according to actual conditions, thereby improving the accuracy and robustness of the infection spread estimation model.

[0157] Step S54: performing parameter optimization processing on the infection propagation estimation model according to the model estimation error loss value to obtain an infection propagation optimization model;

[0158] The embodiment of the present invention optimizes the parameters of the infection propagation estimation model by using parameter optimization methods such as gradient descent method and genetic algorithm according to the calculated model estimation error loss value to minimize the model estimation error loss value, and iteratively adjusts the parameters of the model and continuously updates the model parameter values ​​until the optimization goal is reached, that is, the estimation error loss value of the model is minimized, and finally an infection propagation optimization model is obtained.

[0159] Step S55: Use the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data.

[0160] The embodiment of the present invention performs infection prediction by inputting infection propagation parameter data and crowd gathering data into the optimized infection propagation optimization model to predict and infer infection propagation laws such as infection trends and propagation rates, and finally obtain infection propagation law data.

[0161] The present invention first performs parameter estimation on a constructed population infection and spread model. The goal of parameter estimation is to find parameter values ​​that best align the model's predictions with actual observed data, using optimization or statistical inference methods. Parameter estimation yields estimated model parameter values, which are used in subsequent infection and spread estimation and optimization processes. Next, the population infection and spread model is estimated and registered using the estimated model parameter values. The goal of registration is to compare the model's predictions with actual observed data to verify the model's accuracy and adaptability. Registration with actual data confirms the model's effectiveness and allows for further adjustments to the model's parameters. The accuracy of the infection and spread estimation model is then evaluated using a suitable model estimation error loss calculation formula. This formula measures the difference between the estimated model's predictions and actual observed data; a smaller model estimation error loss indicates a better fit for the model. By calculating the model's estimation error loss, model performance can be evaluated and used as the objective function for subsequent optimization. The calculated model estimation error loss is then used to adjust and optimize the parameters of the infection and spread estimation model. The optimization goal is to minimize the estimation error loss, thereby improving the model's fit. Through parameter optimization, we can obtain an adjusted and improved infection spread optimization model that better matches actual data. Finally, by using the infection spread optimization model to process infection spread parameter data and crowd aggregation data for infection prediction, the model's predictive capabilities can be used to infer infection spread patterns based on existing data and model parameters. This can help predict and understand infection spread trends, the effectiveness of control measures, and guide population behavior and policies.

[0162] Preferably, the present invention further provides a decision support system construction system based on infection prevention and control, which is used to execute the above-mentioned decision support system construction method based on infection prevention and control, and the decision support system construction system based on infection prevention and control includes:

[0163] The infection data processing module is used to obtain infection prevention and control related data and perform data cleaning on the infection prevention and control related data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data;

[0164] A propagation parameter processing module is used to extract and analyze the propagation parameters of the infection propagation cleaning data to obtain infection propagation parameter data;

[0165] The crowd data analysis module is used to perform crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data;

[0166] The infection transmission model construction module is used to conduct potential correlation mining analysis on infection transmission parameter data and crowd gathering data to obtain key characteristics of transmission parameters and crowd gathering; and to construct a crowd infection transmission model based on the key characteristics of transmission parameters and crowd gathering;

[0167] The parameter optimization and prediction module is used to perform parameter estimation and optimization processing on the population infection propagation model to obtain the infection propagation optimization model; the infection propagation optimization model is used to perform infection prediction processing on the infection propagation parameter data and the crowd gathering data to obtain the infection propagation law data;

[0168] The decision support system construction module is used to conduct prevention and control decision support analysis on infection transmission law data to obtain prevention and control decision analysis results; a decision support system is constructed based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

[0169] In summary, the present invention provides a decision support system construction system based on infection prevention and control, which consists of an infection data processing module, a propagation parameter processing module, a crowd data analysis module, an infection propagation model construction module and a decision support system construction module. It can implement any one of the infection prevention and control-based decision support system construction methods described in the present invention, and is used to combine the operations between the computer programs running on each module to implement a decision support system construction method based on infection prevention and control. The internal structures of the system cooperate with each other, and by using infection prevention and control-related data and advanced technical means, according to the characteristics and laws of infection propagation, a decision support system is constructed that can analyze multiple factors such as infection propagation and crowd gathering, providing decision makers with real-time and accurate decision support, and combining the decision makers' needs and background information to generate personalized infection prevention and control decision recommendations, thereby improving the quality and efficiency of infection prevention and control decisions. This can greatly reduce duplication of work and manpower investment, and can quickly and effectively provide a more accurate and efficient decision support system construction process, thereby simplifying the operation process of the infection prevention and control-based decision support system construction system.

[0170] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0171] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a decision support system based on infection prevention and control, characterized in that: The following steps are involved: Step S1: Acquire infection prevention and control related data, perform data cleaning on the infection prevention and control related data, and obtain infection prevention and control cleaned data, wherein the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data; Step S2: extract and analyze the infection propagation parameters of the cleaned infection propagation data to obtain infection propagation parameter data; Step S3: Performing crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data; Step S4: Perform potential association mining analysis on the infection transmission parameter data and the crowd gathering data to obtain the key features of the transmission parameters and the key features of the crowd gathering; Construct a population infection propagation model based on the key characteristics of transmission parameters and key characteristics of population gathering; Step S5: Optimizing the parameters of the population infection propagation model to obtain an infection propagation optimization model; using the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data; Step S6: Conduct prevention and control decision support analysis on the infection transmission pattern data to obtain prevention and control decision analysis results; build a decision support system based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

2. The method for constructing a decision support system based on infection prevention and control according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Obtain infection prevention and control related data; Step S12: Repeatedly screen and process the infection prevention and control related data to obtain infection prevention and control screening data; Step S13: using the abnormality degree calculation formula to perform abnormality detection calculation on the infection prevention and control screening data to obtain the infection prevention and control abnormal value; Step S14: performing abnormal filtering on the infection prevention and control screening data according to the infection prevention and control abnormal values ​​to obtain prevention and control abnormality filtered data; Step S15: Perform data interpolation processing on the prevention and control abnormality filtered data to obtain prevention and control interpolation filling data; Step S16: statistically integrate the prevention and control interpolation and filling data to obtain prevention and control integrated data; Step S17: Standardize the prevention and control integrated data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data.

3. The method for constructing a decision support system based on infection prevention and control according to claim 2, characterized in that: The abnormality degree calculation formula in step S13 is specifically: Where E is the infection prevention and control anomaly value, T is the observation time range length for anomaly detection calculation, t is the time parameter for anomaly detection calculation, x(t) is the characteristic index value of the infection prevention and control screening data, α is the amplitude control parameter of the infection prevention and control screening data, exp is the exponential function, μ is the mean parameter of the infection prevention and control screening data, σ is the standard deviation parameter of the infection prevention and control screening data, γ is the anomaly amplitude control parameter of the anomaly distribution peak in the infection prevention and control screening data, β is the mean parameter of the anomaly distribution peak in the infection prevention and control screening data, ρ is the anomaly degree curve adjustment parameter of the infection prevention and control screening data, η is the anomaly degree curve threshold parameter of the infection prevention and control screening data, and ε is the correction value of the infection prevention and control anomaly value.

4. The method for constructing a decision support system based on infection prevention and control according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing path trend analysis on the infection propagation cleansing data to obtain an infection propagation trend path; Step S22: Calculate the infection propagation rate of the cleaned infection propagation data to obtain the infection propagation rate; Step S23: performing propagation distance mining analysis based on the infection propagation trend path and the infection propagation rate to obtain the infection propagation distance; Step S24: Calculate the infection rate of the infection propagation cleansed data to obtain the infection rate; Step S25: performing a latent calculation on the infection transmission cleaned data using a transmission latent rate calculation formula according to the infection transmission rate, the infection transmission distance, and the infection rate to obtain the infection transmission latent rate; The formula for calculating the transmission latency rate is as follows: Where Q(τ) is the infection transmission latent rate within time τ, τ is the time range parameter for latent calculation, and τ ′ is the time variation parameter of the latent calculation, δ is the infection rate, I(τ-τ ′ ) is the time τ-τ ′ The infection spread rate value within, N is the total number of individuals infected in the infection spread cleaning data, A ij The adjacency matrix of the interpersonal contact network composed of individual i and individual j in the infection spread cleaning data, |A ij | is the modulus of the adjacency matrix, i.e., the infection spread distance value of the interpersonal contact network, is the infection recovery rate, f(τ) is the infection incubation period distribution function, S(τ-τ ′ ) is the time τ-τ ′ The number of susceptible people in ,∈ is the correction value of the latent rate of infection transmission; Step S26: Integrate and analyze the infection spread trend path, infection spread rate, infection spread distance, infection rate, and infection spread latency rate to obtain infection spread parameter data.

5. The method for constructing a decision support system based on infection prevention and control according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: using kernel density estimation technology to perform density estimation processing on the crowd flow cleaned data to obtain the crowd flow estimated density; Step S32: analyzing the crowd flow cleansing data based on the crowd flow estimated density to obtain the crowd flow behavior pattern; Step S33: Calculate the crowd flow cleaning data using the crowd flow frequency calculation formula to obtain the crowd flow frequency; Step S34: performing regional flow distribution analysis on the crowd flow cleansing data to obtain regional crowd flow conditions; Step S35: Perform spatial aggregation analysis based on crowd flow behavior patterns, crowd flow frequency, and regional crowd flow conditions to obtain crowd aggregation data.

6. The method for constructing a decision support system based on infection prevention and control according to claim 5, characterized in that: The formula for calculating the crowd flow frequency in step S33 is specifically: Where F is the crowd flow frequency, T ′ is the observation period for crowd flow calculation, t ′ is the observation time variable for crowd flow calculation, n is the total number of crowd segments for crowd flow cleaning data, and λ k The population size of the kth crowd segment in the crowd flow cleaning data, D k is the average flow distance of the kth crowd segment in the crowd flow cleaning data, θ1 is the time attenuation control parameter of the crowd flow, θ2 is the stability control parameter of the crowd flow, The scaling factor is adjusted for the shape of the crowd flow, and ξ is the correction value for the frequency of the crowd flow.

7. The method for constructing a decision support system based on infection prevention and control according to claim 1, characterized in that: Step S2 includes the following steps: Step S41: performing a first association mining analysis on the infection propagation parameter data and the crowd gathering situation data to obtain propagation parameter characteristics and crowd gathering characteristics; Step S42: performing potential pattern recognition analysis on the infection propagation parameter data and the crowd gathering situation data to obtain a potential pattern of the propagation parameters and a potential pattern of the crowd gathering; Step S43: performing a second association mining analysis on the propagation parameter features using the propagation parameter potential pattern to obtain key features of the propagation parameters; Step S44: performing a second association mining analysis on the crowd gathering characteristics using the crowd gathering potential pattern to obtain the crowd gathering key characteristics; Step S45: Construct a crowd infection propagation model based on the key features of the propagation parameters and the key features of the crowd gathering.

8. The method for constructing a decision support system based on infection prevention and control according to claim 7, characterized in that: Step S42 includes the following steps: Step S421: performing data distribution exploration analysis on the infection propagation parameter data and the crowd gathering situation data to obtain propagation parameter distribution data and crowd gathering distribution data; Step S422: performing correlation calculations on the propagation parameter distribution data and the crowd gathering distribution data to obtain propagation parameter correlation data and crowd gathering correlation data; Step S423: performing correlation mining analysis on the propagation parameter related data and the crowd gathering related data to obtain the propagation parameter correlation relationship and the crowd gathering correlation relationship; Step S424: Perform potential mining analysis on the correlation between propagation parameters and the correlation between crowd gatherings to obtain potential patterns of propagation parameters and potential patterns of crowd gatherings.

9. The method for constructing a decision support system based on infection prevention and control according to claim 1, characterized in that: Step S5 includes the following steps: Step S51: performing parameter estimation on the population infection propagation model to obtain model parameter estimation values; Step S52: using the estimated values ​​of the model parameters to estimate and align the population infection propagation model to obtain an infection propagation estimation model; Step S53: Calculating the error loss of the infection spread estimation model using the model estimation error loss calculation formula to obtain a model estimation error loss value; Among them, the model estimation error loss calculation formula is as follows: Where, is the estimated error loss value for the model, is the infection spread estimation model parameter, Ω is the parameter space of the infection spread estimation model parameter, is the estimated value of the infection spread estimation model, R is the real observation data, X is the integral parameter variable in the parameter space, w(X) is the weight function, p(X) is the probability density function, and ι is the correction value of the model estimation error loss value; Step S54: performing parameter optimization processing on the infection propagation estimation model according to the model estimation error loss value to obtain an infection propagation optimization model; Step S55: Use the infection propagation optimization model to perform infection prediction processing on the infection propagation parameter data and the crowd gathering situation data to obtain infection propagation law data.

10. A decision support system construction system based on infection prevention and control, characterized in that: The method for constructing a decision support system based on infection prevention and control according to claim 1 is configured to execute the method, wherein the decision support system construction system based on infection prevention and control comprises: The infection data processing module is used to obtain infection prevention and control related data and perform data cleaning on the infection prevention and control related data to obtain infection prevention and control cleaned data, where the infection prevention and control cleaned data includes infection transmission cleaned data and crowd flow cleaned data; A propagation parameter processing module is used to extract and analyze the propagation parameters of the infection propagation cleaning data to obtain infection propagation parameter data; The crowd data analysis module is used to perform crowd gathering statistical analysis on the crowd flow cleaning data to obtain crowd gathering data; The infection transmission model construction module is used to conduct potential correlation mining analysis on infection transmission parameter data and crowd gathering data to obtain key characteristics of transmission parameters and crowd gathering; and to construct a crowd infection transmission model based on the key characteristics of transmission parameters and crowd gathering; The parameter optimization and prediction module is used to perform parameter estimation and optimization processing on the population infection propagation model to obtain the infection propagation optimization model; the infection propagation optimization model is used to perform infection prediction processing on the infection propagation parameter data and the crowd gathering data to obtain the infection propagation law data; The decision support system construction module is used to conduct prevention and control decision support analysis on infection transmission law data to obtain prevention and control decision analysis results; a decision support system is constructed based on the prevention and control decision analysis results to execute corresponding infection prevention and control decisions.

Images