A method and device for predicting the spread of infectious diseases based on clustering

Abstract

The present application relates to a clustering-based prediction method and device for the spread of infectious diseases. By acquiring the epidemic situation-related data of the target area within a preset time period in real time, the diagnosis rate index, the mortality rate index and the recovery rate index are constructed, and the Markov state space is constructed based on these three indicators. Then, the state space is divided based on the method of minimizing the sum of squared deviations to obtain a preset number of state subspaces, and then a Markov chain is constructed based on the state subspace to obtain a probability transfer matrix, and the spread state of the infectious disease is predicted based on the probability transfer matrix. The present invention comprehensively considers the diagnosis rate index, mortality rate index and recovery rate index of the infectious disease, and according to the characteristics of the index, uses a clustering method to analyze the development of the infectious disease, constructs and analyzes the stability of the probability transfer matrix, and accurately predicts the state development of the epidemic situation of the infectious disease.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a clustering-based method and device for predicting the spread status of an infectious disease. Background Art

[0002] The evaluation of the epidemic situation of infectious diseases is difficult to quantify. With the development of the times, the virus will produce various mutations, and the government will also introduce corresponding policies at the same time, resulting in the epidemic situation of infectious diseases and related data collection not meeting the independent and identically distributed assumption. Therefore, traditional statistical methods are often difficult to work. In terms of prediction, the traditional SIR infectious disease model only uses initial historical data, and the prediction error increases with time. In view of the shortcomings of the traditional infectious disease model, in order to overcome this deficiency, many scholars have used Markov chain theory to predict the incidence of infectious diseases and established some mathematical models. Some researchers have applied the Markov chain method to conduct a comprehensive analysis and study of the incidence of various infectious diseases, and adopted quantitative analysis methods. Because it has less dependence on historical data, and the Markov chain model has a looser requirement for the independent and identically distributed condition of data, it has a higher accuracy. However, the Markov chain method has limitations. The Markov chain model uses artificially preset intervals to divide the state space. This method is often subjective and its results are affected by certain subjective biases. Secondly, the traditional Markov chain model focuses on the transfer of the state space of the number of infected people. A single indicator is difficult to effectively evaluate the epidemic situation of infectious diseases. When the indicator gradually increases, the division of the infectious disease state space will be more subjective. In addition, the adjustment of the transition probability matrix and the excessive influence of objective factors on the prediction accuracy are also the shortcomings of the traditional Markov chain model.

[0003] The use of Markov chain theory to establish a mathematical model to predict the incidence of infectious diseases is to absorb the advantages of the traditional Markov chain method that there is no aftereffect, relax the assumption that the data are independent and identically distributed, and make full use of the advantages of the correlation analysis method to give full play to the role of historical data. If we can construct relevant evaluation indicators and a method for dividing the state space of the epidemic situation of infectious diseases under multiple indicators, and fully explore the properties of the epidemic situation of infectious diseases such as the speed of transmission, degree of danger, and the impact of policies and major events, we can further evaluate the epidemic situation of infectious diseases. Summary of the invention

[0004] Based on this, it is necessary to provide a multi-indicator clustering-based infectious disease transmission status prediction method and device based on the above-mentioned technical problems, which takes the infectious disease epidemic situation assessment as the starting point.

[0005] A method for predicting the spread of infectious diseases based on clustering, the method comprising:

[0006] Acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate;

[0007] Construct a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data;

[0008] Clustering the infectious disease data based on a method of minimizing the sum of squared deviations and then dividing the Markov state space to obtain a plurality of state subspaces;

[0009] A Markov chain is constructed according to the multiple state subspaces to obtain a state transfer matrix, and the propagation state of the infectious disease is predicted according to the stationarity of the state transfer matrix.

[0010] In one embodiment, the method based on minimizing the sum of squared deviations is used to cluster the infectious disease data and then divide the Markov state space to obtain multiple state subspaces, including:

[0011] The infectious disease data is clustered based on the method of minimizing the sum of squared deviations to obtain an initial clustering result; wherein the sum of squared deviations of the i-th cluster in the initial clustering result is:

[0012]

[0013] Among them, x k，i =[R k，c , R k，d , R k，r ]T, represents the diagnosis rate, mortality rate and recovery rate of the infectious disease data clustered into the i-th cluster at time k, n i represents the total number of steps of clustering of the i-th cluster, represents the center of the i-th cluster;

[0014] Each cluster in the initial clustering result is pre-merged in pairs, and the increment of the sum of squares of deviations within the pre-merged cluster relative to the sum of squares of deviations within the corresponding two clusters before the pre-merger is calculated:

[0015] ΔL=L i∪j -(L i +L j )

[0016] Among them, ΔL is the increment of the sum of squared deviations, L i∪j is the sum of squared deviations within the new cluster obtained after pre-merging the i-th and j-th clusters, L i is the sum of squared deviations within cluster i before pre-merging, L j is the sum of squared deviations within cluster j before pre-merging;

[0017] The pre-merged cluster corresponding to the minimum increment and the pre-merged cluster corresponding to the non-minimum increment are used as the initial Markov state space for the next clustering process until a preset number of clusters are obtained, and the preset number of clusters are used as the clusters of the final clustering, and the clusters of the final clustering correspond to the state subspaces obtained after dividing the Markov state space.

[0018] In one embodiment, constructing a Markov chain according to the multiple state subspaces to obtain a state transfer matrix includes:

[0019] The state transfer matrix obtained by constructing a Markov chain according to the multiple state subspaces is:

[0020]

[0021] Among them, m is the number of state subspaces, P ij is the state subspace S i After 1 step, it is transferred to the state subspace S j The transition probability.

[0022] The calculation methods of the confirmed rate, mortality rate and recovery rate include:

[0023] The confirmed rate is calculated based on the number of confirmed cases in the target area within the preset time period and the total population of the target area:

[0024]

[0025] Among them, R c is the diagnosis rate, N c is the number of confirmed cases within the preset time period, N t is the total population of the target area;

[0026] The mortality rate is calculated based on the number of deaths in the target area within the preset time period and the cumulative number of confirmed cases of infectious diseases in the target area:

[0027]

[0028] Among them, R d is the mortality rate, N d (t) is the number of confirmed cases within the preset time period, The cumulative number of deaths from infectious diseases in the target area;

[0029] The recovery rate is calculated based on the number of people who have recovered within the preset time period in the target area and the cumulative number of confirmed cases of infectious diseases in the target area:

[0030]

[0031] Among them, R r is the cure rate, Nr (t) is the number of people who have recovered within the preset time period.

[0032] A clustering-based infectious disease propagation status prediction device, the device comprising:

[0033] An acquisition module is used to acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate;

[0034] A building block for constructing a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data;

[0035] A partitioning module, used for clustering the infectious disease data based on a method of minimizing the sum of squared deviations and then partitioning the Markov state space to obtain a plurality of state subspaces;

[0036] The prediction module is used to construct a Markov chain according to the multiple state subspaces to obtain a state transfer matrix, and predict the propagation state of the infectious disease according to the stationarity of the state transfer matrix.

[0037] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0038] Acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate;

[0039] Construct a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data;

[0040] Clustering the infectious disease data based on a method of minimizing the sum of squared deviations and then dividing the Markov state space to obtain a plurality of state subspaces;

[0041] A Markov chain is constructed according to the multiple state subspaces to obtain a state transfer matrix, and the propagation state of the infectious disease is predicted according to the stationarity of the state transfer matrix.

[0042] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0043] Acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate;

[0044] Construct a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data;

[0045] Clustering the infectious disease data based on a method of minimizing the sum of squared deviations and then dividing the Markov state space to obtain a plurality of state subspaces;

[0046] A Markov chain is constructed according to the multiple state subspaces to obtain a state transfer matrix, and the propagation state of the infectious disease is predicted according to the stationarity of the state transfer matrix.

[0047] The above-mentioned clustering-based infectious disease propagation status prediction method and device constructs a diagnosis rate index, a mortality rate index and a cure rate index by acquiring relevant data on the epidemic situation of infectious diseases in a preset time period in the target area in real time, and constructs a Markov state space based on these three indicators, and then divides the state space based on the method of minimizing the sum of squared deviations to obtain a preset number of state subspaces, and then constructs a Markov chain based on the state subspace to obtain a probability transfer matrix, and predicts the propagation status of infectious diseases based on the probability transfer matrix. The present invention comprehensively considers the diagnosis rate index, mortality rate index and cure rate index of infectious diseases, and according to the characteristics of the indicators, uses a clustering method to analyze the development of infectious diseases, constructs and analyzes the properties of the probability transfer matrix (mainly stability), and accurately predicts the state development of the epidemic situation of infectious diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a flow chart of a clustering-based infectious disease propagation status prediction method in an embodiment;

[0049] Figure 2 A schematic diagram of a method for evenly dividing a state space in one embodiment;

[0050] Figure 3 The result of constructing and dividing the Markov state space in region A in one embodiment;

[0051] Figure 4 The result of constructing and dividing the Markov state space in region B in one embodiment;

[0052] Figure 5 is a structural block diagram of a clustering-based infectious disease propagation status prediction device in one embodiment;

[0053] Figure 6 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] In one embodiment, Figure 1 As shown, a clustering-based infectious disease transmission status prediction method is provided, comprising the following steps:

[0056] Step 102: Obtain an infectious disease dataset in a target area.

[0057] The infectious disease dataset includes infectious disease data at multiple different times. The attributes of each infectious disease data include diagnosis rate, mortality rate, and recovery rate.

[0058] The data set in this embodiment is dynamic streaming data acquired in real time. According to the characteristics of streaming data, the influence of infectious disease epidemic data not satisfying the independent homogeneous distribution assumption can be avoided, and the infectious disease epidemic situation is evaluated only based on local areas.

[0059] Step 104, constructing a Markov state space according to the diagnosis rate, mortality rate and recovery rate of the infectious disease data.

[0060] Every disease, regardless of its occurrence, development and cure, is affected by many external and internal factors of the human body. These factors are complexly related and difficult to explain with structural causal models. This interdependence between data is the most important and useful feature of the research object. According to the changing laws of infectious disease data over time series, establishing a dynamic model is an effective method.

[0061] This example constructs three indicators to measure the development of infectious disease epidemics (R c , R d , R r ), these three indicators reflect the development status of the epidemic situation of infectious diseases from different perspectives, making the constructed Markov state space contain rich information. For example, when the virus is highly contagious but easy to cure, the indicator can be expressed as R c Larger, R r When the virus is less contagious but has a higher mortality rate, the indicator can represent R c Smaller, R d It can be seen that different scales of indicator combinations can be used to measure different epidemic situations of infectious diseases.

[0062] Step 106, clustering the infectious disease data based on the method of minimizing the sum of squared deviations and then dividing the Markov state space to obtain multiple state subspaces.

[0063] The evaluation of infectious disease epidemics is difficult to quantify. With the development of the times, viruses will produce various mutations, and the government will also introduce corresponding policies at the same time, resulting in the data related to the epidemic of infectious diseases not meeting the independent and identically distributed assumption. Therefore, traditional statistical methods are often difficult to work. By using the method of stream data clustering, the sampling results of the same distribution are set in the same cluster, and multiple state subspaces corresponding to multiple clusters can be obtained.

[0064] In the method of evenly dividing the state space, each indicator is divided into two conditions: high and low. Then, two conditions of the three indicators are combined to obtain a total of eight states, such as Figure 2 However, in reality, the distribution of the epidemic situation of infectious diseases in the same target area at different times is uneven and relatively concentrated. Therefore, a more reasonable state space division method that is more in line with the actual situation is needed, that is, according to the characteristics of historical data (R c , R d , R r ) divides the data into K clusters to minimize the differences between each data sample in each cluster. The sum of squared differences represents the dispersion of the data well, so the sum of the variances of all data in the same cluster represents the similarity of the data attributes in the same cluster well.

[0065] Step 108, constructing a Markov chain according to multiple state subspaces to obtain a state transfer matrix, and predicting the spread state of the infectious disease according to the stationarity of the state transfer matrix.

[0066] Multiple state subspaces are related in time and space, so a Markov chain can be constructed based on multiple state subspaces, and then a Markov chain can be constructed based on the time and space correlation between the state subspaces, and then the state transfer matrix can be calculated. Then, by analyzing the stationarity of the state transfer matrix, it can be determined in which state subspace the three indicators of the epidemic situation of the infectious disease gradually stabilize, so as to predict the spread status of the infectious disease.

[0067] The above-mentioned clustering-based infectious disease propagation status prediction method constructs a diagnosis rate index, a mortality rate index and a cure rate index by acquiring relevant data on the epidemic situation of infectious diseases in a preset time period in the target area in real time, and constructs a Markov state space based on these three indicators, and then divides the state space based on the method of minimizing the sum of squared deviations to obtain a preset number of state subspaces, and then constructs a Markov chain based on the state subspace to obtain a probability transfer matrix, and predicts the propagation status of infectious diseases based on the probability transfer matrix. The present invention comprehensively considers the diagnosis rate index, mortality rate index and cure rate index of infectious diseases, and according to the characteristics of the indicators, uses a clustering method to analyze the development of infectious diseases, constructs and analyzes the properties of the probability transfer matrix (mainly stability), and accurately predicts the state development of the epidemic situation of infectious diseases.

[0068] In one embodiment, the calculation method of the confirmed rate, mortality rate and recovery rate includes:

[0069] The confirmed rate is calculated based on the number of confirmed cases in the target area within the preset time period and the total population of the target area:

[0070]

[0071] Among them, R c is the diagnosis rate, N c is the number of confirmed cases within the preset time period, N t is the total population of the target area;

[0072] The mortality rate is calculated based on the number of deaths in the target area within the preset time period and the cumulative number of confirmed cases of infectious diseases in the target area:

[0073]

[0074] Among them, R d is the mortality rate, N d (t) is the number of deaths within the preset time period, The cumulative number of confirmed cases of infectious diseases in the target area;

[0075] The recovery rate is calculated based on the number of people who have recovered within the preset time period in the target area and the cumulative number of confirmed cases of infectious diseases in the target area:

[0076]

[0077] Among them, R r is the cure rate, N r (t) is the number of people who have recovered within the preset time period.

[0078] In one embodiment, the infectious disease data is clustered based on the method of minimizing the sum of squared deviations, and the Markov state space is divided to obtain multiple state subspaces, including:

[0079] The infectious disease data are clustered based on the method of minimizing the sum of squared deviations to obtain the initial clustering results; the sum of squared deviations of the i-th cluster in the initial clustering results is:

[0080]

[0081] Among them, x k，i =[R k，c , R k，d , R k，r ] T , represents the diagnosis rate, mortality rate and recovery rate of the infectious disease data clustered into the i-th cluster at time k, n i represents the total number of steps of clustering of the i-th cluster, represents the center of the i-th cluster;

[0082] Each cluster in the initial clustering result is pre-merged two by two, and the increment of the sum of squared deviations within the pre-merged cluster relative to the sum of squared deviations within the corresponding two clusters before the pre-merger is calculated:

[0083] ΔL=L i∪j -(L i +L j )

[0084] Among them, ΔL is the increment of the sum of squared deviations, L i∪j is the sum of squared deviations within the new cluster obtained after pre-merging the i-th and j-th clusters, L i is the sum of squared deviations within cluster i before pre-merging, L j is the sum of squared deviations within cluster j before pre-merging;

[0085] The pre-merged cluster corresponding to the minimum increment and the pre-merged cluster corresponding to the non-minimum increment are used as the initial Markov state space of the next clustering process until a preset number of clusters are obtained, and the preset number of clusters are used as the clusters of the final clustering. The clusters of the final clustering correspond to the state subspaces obtained after dividing the Markov state space.

[0086] For example, we select the epidemic situation data of region A within a certain period of time to construct and divide the Markov state space. The results are as follows: Figure 2 As shown, we get 5 clusters, that is, 5 state subspaces. We can see that the characteristic of state subspace 1 is high R r , low R d ; The state subspace 2 is characterized by R r , high R c, low R d ; State subspace 3 is characterized by low R r , low R d ; State subspace 4 is characterized by low R r , high R d ; State subspace 5 is characterized by low R r , Medium R d The epidemic situation data of infectious diseases in area B within a certain period of time are selected to construct and divide the Markov state space. The results are as follows Figure 3 As shown in the figure, it can be seen that the distribution of the state subspace obtained by dividing the epidemic situation data in different regions is different, which is in line with the actual situation. Different regions adopt different epidemic prevention and control strategies for infectious diseases, and the corresponding epidemic situation development trends of infectious diseases will also be different.

[0087] In one embodiment, a Markov chain is constructed according to a plurality of state subspaces to obtain a state transfer matrix, including:

[0088] The state transfer matrix obtained by constructing a Markov chain based on multiple state subspaces is:

[0089]

[0090] Among them, m is the number of state subspaces, P ij is the state subspace S i After 1 step, it is transferred to the state subspace S j The constructed state space can be divided into different state subspaces. Of course, different prevention and control strategies adopted by regional managers will cause the epidemic situation of infectious diseases to spread throughout the national space. Therefore, the Markov state transfer matrix will change in different periods, that is, by analyzing the changes in the Markov chain transfer matrix in different periods, we can understand the impact of different prevention and control strategies.

[0091] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0092] A specific embodiment is provided here to explain this method in detail:

[0093] The prevention and control strategies for area A are:

[0094] February 23: Restrictions on travel to areas where the epidemic is more serious.

[0095] March 16: Social distancing measures are introduced, and gatherings of more than 50 people are prohibited from March 26 to July 26.

[0096] Therefore, we selected data from February 23 to March 30 for calculation. Considering that the incubation period of infectious diseases is 14 days, we chose March 30 instead of March 16, because even if a new control strategy is issued on March 16, the next 14 days will still show the characteristics of the epidemic spread of infectious diseases under the previous strategy. Figure 4 As shown, a Markov chain is obtained, and the corresponding state transfer matrix is ​​obtained:

[0097]

[0098] By applying the Markov chain stability judgment method, we can know that the Markov chain is stable. Therefore, under this prevention and control strategy, the impact of the spread of infectious diseases will remain stable. It can be obtained:

[0099]

[0100] This means that under this strategy, the epidemic situation will gradually approach state 2, R c Higher, while R r This indicates that if this strategy continues, the epidemic situation of infectious diseases will not improve and the strategy needs to be adjusted.

[0101] From the final stationary distribution, it can be seen that the quarantine control strategy adopted on March 16 was helpful in controlling the epidemic situation. During the stationary distribution, the initial strategy (February 23) will shift from state 1 to state 2.

[0102] Next, we select the data from March 30 to May 26 for calculation and obtain the probability transfer matrix:

[0103]

[0104] This Markov chain is also stable. Under this policy, the impact of the spread of the epidemic situation will also remain stable, and we get:

[0105]

[0106] The measures taken on March 16 reduced the probability of transitioning from state 1 to state 2 to 0.82 and increased the probability of transitioning to state 5 to 0.18. State 5 represents high R r , low R d and R c , that is, low infection rate, low mortality rate and high cure rate. Therefore, the measures taken on March 16 played a positive role in controlling the spread of the epidemic situation.

[0107] Finally, select the data from May 26 to July 4 and get the probability transfer matrix:

[0108]

[0109] The Markov chain is also stationary, and we have:

[0110]

[0111] If this continues, the epidemic situation in region A will be in state 1: a high level of R c and R d , while at a low level of R c state, that is, a completely out-of-control state with high mortality, high diagnosis rate, and low cure rate; or at 59% R c and R r Under low conditions, the impact of disease transmission cannot be controlled. This shows that in the context of the spread of infectious diseases, demonstrations will make the epidemic worse.

[0112] In one embodiment, Figure 5 As shown, a clustering-based infectious disease transmission status prediction device is provided, comprising: an acquisition module, an indicator construction module and a division module, wherein:

[0113] An acquisition module is used to acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate;

[0114] A building block for constructing a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data;

[0115] A partitioning module is used to cluster the infectious disease data based on the method of minimizing the sum of squared deviations and then partition the Markov state space to obtain multiple state subspaces;

[0116] The prediction module is used to construct a Markov chain based on multiple state subspaces to obtain a state transfer matrix, and predict the spread state of the infectious disease based on the stationarity of the state transfer matrix.

[0117] For the specific definition of a clustering-based infectious disease propagation status prediction device, please refer to the definition of a clustering-based infectious disease propagation status prediction method above, which will not be repeated here. Each module in the above-mentioned clustering-based infectious disease propagation status prediction device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0118] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data related to the epidemic situation of infectious diseases. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a clustering-based infectious disease propagation status prediction method is implemented.

[0119] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0120] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method in the above embodiment when executing the computer program.

[0121] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method in the above embodiment are implemented.

[0122] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0124] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A clustering-based infectious disease transmission status prediction method, characterized in that: The method comprises: Acquire an infectious disease data set of a target area; the infectious disease data set includes infectious disease data at multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate; The calculation method of the confirmed rate, mortality rate and recovery rate includes: calculating the confirmed rate according to the number of confirmed cases in the target area within a preset time period and the total population of the target area is: in, The diagnosis rate is is the number of confirmed cases within the preset time period. is the total population of the target area; The mortality rate is calculated based on the number of deaths in the target area within the preset time period and the cumulative number of confirmed cases of infectious diseases in the target area: in, For mortality, is the number of deaths within the preset time period, The cumulative number of confirmed cases of infectious diseases in the target area; The recovery rate is calculated based on the number of people who have recovered within the preset time period in the target area and the cumulative number of confirmed cases of infectious diseases in the target area: in, For the cure rate, The number of people who have recovered within a preset time period; Construct a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data; Clustering the infectious disease data based on a method of minimizing the sum of squared deviations and then dividing the Markov state space to obtain a plurality of state subspaces; The infectious disease data are clustered based on the method of minimizing the sum of squared deviations to obtain an initial clustering result; The sum of squared deviations of the clusters is: in, , indicating that it is clustered as The infectious disease data of clusters is The diagnosis rate, mortality rate and recovery rate at each moment, Indicates The total number of steps for clustering, Indicates The center of the cluster; Each cluster in the initial clustering result is pre-merged in pairs, and the increment of the sum of squares of deviations within the pre-merged cluster relative to the sum of squares of deviations within the corresponding two clusters before the pre-merger is calculated: in, is the increment of the sum of squared deviations, For pre-merge The sum of squared deviations within the new clusters obtained after two clusters, Before the merge i The sum of squared deviations within the cluster, Before the merge j The sum of squared deviations within the cluster; The pre-merged cluster corresponding to the minimum increment and the pre-merged cluster corresponding to the non-minimum increment are used as the initial Markov state space of the next clustering process until a preset number of clusters are obtained, and the preset number of clusters are used as the clusters of the final clustering, and the clusters of the final clustering correspond to the state subspaces obtained after the Markov state space is divided; A Markov chain is constructed according to the multiple state subspaces to obtain a state transfer matrix, and the propagation state of the infectious disease is predicted according to the stationarity of the state transfer matrix.

2. The method according to claim 1, characterized in that The step of constructing a Markov chain according to the multiple state subspaces to obtain a state transfer matrix includes: The state transfer matrix obtained by constructing a Markov chain according to the multiple state subspaces is: in, is the number of state subspaces, is the state subspace After 1 step, transfer to the state subspace The transition probability.

3. A clustering-based infectious disease transmission status prediction device, characterized in that: The device comprises: The acquisition module is used to acquire an infectious disease data set of a target area, wherein the infectious disease data set includes infectious disease data of multiple different times; the attributes of each infectious disease data include a confirmed rate, a mortality rate, and a cure rate; the calculation method of the confirmed rate, the mortality rate, and the cure rate includes: calculating the confirmed rate according to the number of confirmed cases within a preset time period of the target area and the total population of the target area: in, The diagnosis rate is is the number of confirmed cases within the preset time period. is the total population of the target area; the mortality rate is calculated based on the number of deaths within the preset time period of the target area and the cumulative number of confirmed cases of infectious diseases in the target area: in, For mortality, is the number of deaths within the preset time period, is the cumulative number of confirmed cases of infectious diseases in the target area; the recovery rate is calculated based on the number of recovered people within the preset time period in the target area and the cumulative number of confirmed cases of infectious diseases in the target area: in, For the cure rate, The number of people who have recovered within a preset time period; A building block for constructing a Markov state space based on the diagnosis rate, mortality rate, and recovery rate of infectious disease data; A partitioning module is used to cluster the infectious disease data based on the method of minimizing the sum of squared deviations and then partition the Markov state space to obtain multiple state subspaces; cluster the infectious disease data based on the method of minimizing the sum of squared deviations to obtain an initial clustering result; wherein the first The sum of squared deviations of the clusters is: in, , indicating that it is clustered as The infectious disease data of clusters is The diagnosis rate, mortality rate and recovery rate at each moment, Indicates The total number of steps for clustering, Indicates The centers of the clusters are pre-merged in pairs, and the increment of the sum of squares of deviations within the pre-merged clusters relative to the sum of squares of deviations within the two corresponding clusters before the pre-merger is calculated: in, is the increment of the sum of squared deviations, For pre-merge The sum of squared deviations within the new clusters obtained after two clusters, Before the merge i The sum of squared deviations within the cluster, Before the merge j The sum of squared deviations within the cluster; using the pre-merged cluster corresponding to the minimum increment and the pre-merged cluster corresponding to the non-minimum increment as the initial Markov state space of the next clustering process, until a preset number of clusters are obtained, and using the preset number of clusters as the clusters of the final clustering, wherein the clusters of the final clustering correspond to the state subspaces obtained after dividing the Markov state space; The prediction module is used to construct a Markov chain according to the multiple state subspaces to obtain a state transfer matrix, and predict the propagation state of the infectious disease according to the stationarity of the state transfer matrix.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images