Medicine adoption behavior analysis method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of data analysis, and discloses a medicine adoption behavior analysis method and device, computer equipment and a storage medium. The method comprises the following steps: constructing a high-order social network structure based on a heterogeneity risk avoidance consciousness group; according to the high-order social network structure, constructing a drug intervention adoption dynamics model; according to the method, a drug intervention adoption dynamic model is represented by using a microscopic Markov chain theory, and an evolution equation is iterated to obtain a drug intervention adoption level and an epidemic infection range, so that the influence of a heterogeneity risk avoidance consciousness group on the drug intervention adoption level and the epidemic infection range is analyzed, and a drug adoption strategy is optimized. According to the method, group social interaction and behavior evolution can be accurately quantified, a risk perception-based dynamic behavior adjustment mechanism of an individual is disclosed, a scientific basis is provided for formulating a differentiated intervention strategy, the immune coverage rate is remarkably increased, epidemic spread is inhibited, and the public health resource allocation efficiency is optimized.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a drug adoption behavior analysis method, device, computer equipment and storage medium. Background Art

[0002] In recent years, global pandemics such as COVID-19, the 2014 Ebola outbreak, and the 2009 influenza A outbreak have exposed the huge challenges of implementing unified and effective epidemic response measures in a large and diverse country like China. In particular, the elderly population, due to the particularity of their physiology and immune system, has a large difference in risk aversion, resulting in a large disagreement on whether to adopt drug interventions (such as vaccination). In order to study the drug intervention adoption behavior of the elderly population, existing studies first focus on the evolution of drug intervention adoption behavior of the elderly in social networks. For example, Galvani et al. studied influenza vaccination strategies through game theory models, showing that self-interest can promote vaccination among the elderly. The partition model proposed by Nguyen et al. analyzed the impact of risk tolerance on the scale of epidemics, showing that the heterogeneity of risk tolerance has an important impact on epidemics.

[0003] Existing studies mainly consider the static impact of heterogeneous risk aversion awareness of elderly individuals on drug intervention adoption behavior. However, individual behavior is not static and often adjusts with changes in the social environment. Static models cannot reflect the dynamic changes of individual behavior, especially when new information or environmental changes occur, individual behavior will be updated accordingly. Existing static models may cause the prediction results to be out of line with the actual elderly drug intervention adoption behavior.

[0004] In addition, an individual's risk aversion is often driven by the social infection environment, but existing research often simplifies the social infection environment into the number of infections in the entire system. This approach overestimates the infection risk that older individuals can perceive and fails to accurately reflect the behavioral responses of older individuals in real situations.

[0005] Therefore, how to accurately simulate the drug intervention adoption behavior of the elderly in a dynamic environment and take into account changes in social interactions within the group and the external social environment has become an important issue that needs to be solved urgently. Summary of the invention

[0006] In view of this, the embodiments of the present application provide a drug adoption behavior analysis method, apparatus, computer equipment and storage medium, which can effectively solve the problems of insufficient dynamic modeling, incomplete social network interaction modeling and accurate prediction of drug adoption behavior of the elderly group in the prior art.

[0007] In a first aspect, an embodiment of the present application provides a method for analyzing drug adoption behavior, comprising:

[0008] Construct a corresponding high - order social network structure based on the group with heterogeneous risk - aversion awareness;

[0009] Construct a drug - intervention adoption dynamics model according to the high - order social network structure; the drug - intervention adoption dynamics model includes a public drug - intervention stage and an epidemic - transmission stage;

[0010] Use the microscopic Markov chain theory to characterize the drug - intervention adoption dynamics model and iterate the evolution equation to obtain the drug - intervention adoption level and the epidemic - infection range of the group with heterogeneous risk - aversion awareness;

[0011] Analyze the influence of the group with heterogeneous risk - aversion awareness on the drug - intervention adoption level and the epidemic - transmission range according to the drug - intervention adoption level and the epidemic - infection range, and optimize the drug - adoption strategy according to the influence.

[0012] In some embodiments, before constructing the corresponding high - order social network structure, it includes:

[0013] Introduce an elderly group with heterogeneous risk - aversion awareness, divide the elderly group into multiple sub - groups with heterogeneous risk - aversion awareness according to the heterogeneous risk - aversion awareness levels of each elderly individual in the elderly group, and define an endogenous - intervention adoption rate and an exogenous - intervention adoption rate for the heterogeneous risk - aversion awareness level of each elderly individual in the sub - groups.

[0014] Wherein: the endogenous - intervention adoption rate represents the probability that the elderly individual actively chooses to adopt drugs based on their own risk perception and strategy - selection mechanism without the influence of external intervention; the exogenous - intervention adoption rate represents the probability that the elderly individual passively adopts drugs in the high - order social network structure according to the perceived group - infection risk pressure.

[0015] In some embodiments, constructing the corresponding high - order social network structure based on the group with heterogeneous risk - aversion awareness includes:

[0016] Based on the division of the sub - groups with heterogeneous risk - aversion awareness, construct a contact matrix, and use the simplicial complex model to characterize the interaction between each elderly individual in the high - order social network structure. Among them, use the first - order simplex to represent the pairwise interaction between the elderly individuals, and use the second - order and higher - order simplices to represent the high - order interaction between the elderly individuals.

[0017] In some embodiments, constructing the drug - intervention adoption dynamics model according to the high - order social network structure includes:

[0018] Define behavioral strategies and states for each of the elderly individuals, where the behavioral strategies include adopting drug intervention and rejecting drug intervention; the states include susceptible state, immune state, infected state, and recovered state;

[0019] According to the behavioral strategies, define corresponding costs for different states of each of the elderly individuals, and define combined state benefits based on the costs; the costs include drug intervention costs and infection costs, where the drug intervention costs represent the price that the elderly individual needs to pay for actively choosing to adopt preventive drug intervention or reactive drug intervention, and the infection costs represent the price that the elderly individual needs to pay for rejecting drug intervention; the combined state benefits are used to drive the elderly individual to update the strategy;

[0020] Based on the behavioral strategies, the states, and the costs, construct a two-stage drug intervention adoption dynamics model.

[0021] In some embodiments, the constructing a two-stage drug intervention adoption dynamics model based on the behavioral strategies, the states, and the costs includes:

[0022] In the stage of public drug intervention, the elderly individual uses the Fermi function to imitate the strategies of other elderly individuals according to the combined state benefits of the previous season to select the preventive drug intervention strategy for the current season and pay the corresponding costs;

[0023] In the stage of epidemic spread, conduct reactive drug intervention on the elderly individuals who have not adopted the preventive drug intervention strategy and pay the corresponding costs; the reactive drug intervention conducts drug intervention through the endogenous intervention adoption rate and the exogenous intervention adoption rate.

[0024] In some embodiments, use the microscopic Markov chain theory to characterize the drug intervention adoption dynamics model and iterate the evolution equation to obtain the drug intervention adoption level and the epidemic infection range of the heterogeneous risk-averse awareness group, including:

[0025] Define state probabilities for the elderly individuals to obtain the state probability distribution of each elderly individual at different times;

[0026] Process the state probability distribution to construct an epidemic spread evolution equation on the high-order social network structure;

[0027] Iterate the epidemic spread evolution equation to obtain the state density distribution of the elderly individuals in the steady state;

[0028] According to the state density distribution, calculate the infection range of the elderly group and define the drug intervention adoption probability of the elderly individuals;

[0029] Construct a drug adoption evolution equation on the high-order social network structure according to the infection density of the elderly individuals, the effectiveness of the drug, and the game rules.

[0030] Iterate the drug adoption evolution equation, calculate the drug intervention adoption level until reaching an equilibrium state, and output the final infection range and the final immunity coverage rate.

[0031] In some embodiments, optimizing the drug adoption strategy according to the drug intervention adoption level and the epidemic transmission range includes:

[0032] Simulate different levels of heterogeneous risk aversion and the proportion of the elderly group with heterogeneous risk aversion awareness, and obtain the impact on the drug intervention adoption level and the epidemic transmission range.

[0033] Study the impact of the enhanced infection effect in the high-order social network structure on the drug intervention adoption behavior of the elderly individuals, where the enhanced infection effect increases the drug intervention adoption level of the elderly individuals through high-order interactions.

[0034] Study the impact of the enhanced risk aversion effect in the high-order social network structure on the drug intervention adoption behavior of the elderly individuals, where the enhanced risk aversion effect increases the drug intervention adoption level of the elderly individuals through high-order interactions.

[0035] In a second aspect, an embodiment of the present application provides a drug adoption behavior analysis device, including:

[0036] A social network structure construction module, configured to construct a corresponding high-order social network structure based on the heterogeneous risk aversion awareness group.

[0037] A dynamics model construction module, configured to construct a drug intervention adoption dynamics model according to the high-order social network structure; the drug intervention adoption dynamics model includes a public drug adoption stage and an epidemic transmission stage.

[0038] A range and coverage rate acquisition module, configured to characterize the drug intervention adoption dynamics model using the microscopic Markov chain theory and iterate the evolution equation to obtain the drug intervention adoption level and the epidemic infection range of the heterogeneous risk aversion awareness group.

[0039] A drug adoption strategy optimization module, configured to analyze the impact of the heterogeneous risk aversion awareness group on the drug intervention adoption level and the epidemic transmission range according to the drug intervention adoption level and the epidemic infection range, and optimize the drug adoption strategy according to the impact.

[0040] In a third aspect, an embodiment of the present application provides a computer device, which includes a processor and a memory. The memory stores a computer program, and the processor is configured to execute the computer program to implement the drug adoption behavior analysis method in the first aspect above.

[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium. When the computer program is executed on a processor, it implements the drug adoption behavior analysis method in the first aspect above.

[0042] The embodiments of the present application have the following beneficial effects:

[0043] The drug adoption behavior analysis method of the present application deeply analyzes the essential differences of different-risk elderly individuals in drug intervention decisions (such as willingness to vaccinate independently, passive response) by dividing different risk aversion awareness levels, and constructs a heterogeneous behavior group that conforms to the characteristics of the actual population. Combining with the high-order social network structure, it quantifies the combined impact of point-to-point contact and group aggregation interaction on infection transmission, breaks through the simplification limitations of traditional models on the complexity of social interaction, and significantly improves the theoretical simulation degree of the disease transmission mechanism.

[0044] Adopting a two-stage dynamics model (public intervention + epidemic spread) and a microscopic Markov chain method, it dynamically tracks the behavior adjustment process of elderly individuals under the influence of the social environment. By driving strategy updates through game rules (such as imitating the behavior of high-income neighbors), it reveals how individual decisions form group behavior patterns through social network interaction, and accurately predicts the dynamic change trends of immunization coverage and infection range under different intervention strategies.

[0045] Based on the risk aversion level and intervention cost-benefit analysis, it clarifies the priority intervention threshold for high-risk elderly groups and the environmental response conditions for low-risk groups, and proposes a differential resource allocation plan.

[0046] By quantifying the synergy effect of "risk awareness improvement" and "group size expansion" through equilibrium analysis, it provides adjustment suggestions for policymakers.

[0047] The method of the present application provides quantifiable decision support for public health management, significantly improves the health management efficiency of the elderly group and the epidemic prevention and control effect, and helps to achieve the efficient allocation of social resources and the maximization of health benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and thus should not be regarded as limiting the protection scope of the present application. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 Shows a flowchart of a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0050] Figure 2 Shows a schematic diagram of group division in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0051] Figure 3 Shows a schematic diagram of the state change in the epidemic transmission stage in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0052] Figure 4 Shows a schematic diagram of the change in the immune coverage rate of the elderly population under different proportions of risk-averse awareness groups and risk-averse levels in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0053] Figure 5 Shows a schematic diagram of the impact of expanding the proportion of the risk-averse awareness group on the immune coverage rate when the risk-averse level of the risk-averse awareness group increases in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0054] Figure 6 Shows another schematic diagram of the impact of expanding the proportion of the risk-averse awareness group on the immune coverage rate when the risk-averse level of the risk-averse awareness group increases in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0055] Figure 7 Shows a schematic diagram of the improvement of the immune coverage rate of the entire elderly population when all elderly individuals have risk-averse awareness in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0056] Figure 8 Shows a schematic diagram of the changes in the drug intervention adoption rate and the epidemic spread rate among the elderly population under different drug intervention adoption costs in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0057] Figure 9 Shows another schematic diagram of the changes in the drug intervention adoption rate and the spread rate among the elderly population under different drug intervention adoption costs in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0058] Figure 10 Shows a schematic diagram of the changes in the immune coverage rate and the infection rate of the elderly individuals performing exogenous intervention under different drug intervention adoption costs in a method for analyzing drug adoption behavior according to an embodiment of the present application;

[0059] Figure 11It shows a schematic diagram of the changes in the immune coverage rate and infection rate of endogenous intervention performed by elderly individuals in a drug adoption behavior analysis method according to an embodiment of the present application;

[0060] Figure 12 It shows a schematic structural diagram of a drug adoption behavior analysis device according to an embodiment of the present application. Detailed implementation manners

[0061] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0062] Generally, the components of the embodiments of the present application described and shown in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0063] In the following, the terms "including", "having" and their cognates that can be used in various embodiments of the present application are only intended to represent specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be construed as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or increasing the possibility of one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items.

[0064] In addition, terms such as "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0065] Unless otherwise defined, all terms (including technical terms and scientific terms) used here have the same meaning as commonly understood by those of ordinary skill in the art to which various embodiments of the present application belong. The terms (such as those defined in a general use dictionary) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or being overly formal, unless clearly defined in various embodiments of the present application.

[0066] Next, some embodiments of the present application will be described in detail in conjunction with the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0067] Considering that static models in the prior art cannot adapt to the dynamic adjustment of individual behaviors with the changes in the social environment, and existing research has not fully considered the impact of the social infection environment on the drug intervention adoption behavior of the elderly individuals, a drug adoption behavior analysis method is proposed. A high-order social network structure corresponding to heterogeneous risk-averse awareness groups is constructed; according to the high-order social network structure, a drug intervention adoption dynamics model is constructed; the drug intervention adoption dynamics model is characterized using the microscopic Markov chain theory, and the iterative evolution equation is obtained to get the drug intervention adoption level and the epidemic infection range; according to the drug intervention adoption level and the epidemic infection range, the impact of heterogeneous risk-averse awareness groups on the drug intervention adoption level and the epidemic spread range is analyzed, and the drug adoption strategy is optimized based on the impact. It provides a scientific basis for formulating differentiated intervention strategies, significantly improves the immunization coverage rate and inhibits the spread of the epidemic, and optimizes the allocation efficiency of public health resources.

[0068] Figure 1 FIG. shows a flowchart of the drug adoption behavior analysis method according to an embodiment of the present application. Exemplarily, the method includes the following steps:

[0069] Step S100, based on heterogeneous risk-averse awareness groups, construct a corresponding high-order social network structure.

[0070] Exemplarily, by collecting the real contact data of the elderly individuals, these elderly individuals are classified, and according to their different perceptions of risks, they are divided into multiple subgroups. Each subgroup consists of individuals with similar risk-averse awareness. The behavioral decisions of these individuals are affected by their inherent risk-averse awareness level and the external environment. According to these group divisions, a corresponding high-order social network structure is further constructed to depict the interactions and information flows among group members. This high-order social network takes into account point-to-point interactions and high-order interactions, reflecting the collective influence that individuals may receive during social interactions, thereby affecting the drug intervention adoption behavior.

[0071] In an alternative embodiment, before step S100 of constructing a corresponding high-order social network structure, it includes:

[0072] Exemplarily, introduce an elderly group with heterogeneous risk-averse awareness, such as Figure 2 shown, and according to the heterogeneous risk-averse awareness level of each elderly individual, divide them into multiple subgroups. The elderly individuals in each subgroup have similar risk-averse awareness levels, denoted as L α , where α ∈ {1, …, D}, and D represents the maximum level of risk-averse awareness. To further divide these groups, two variables related to the risk-averse awareness level L α are defined: u α and v α, respectively representing individual behavioral characteristics related to the level of risk aversion awareness.

[0073] Within the group, all elderly individuals with the same level of risk aversion awareness L α are regarded as a subgroup Ω α ; Ω α ={u α , v α ,…}, and the members in this subgroup have similar risk perceptions and decision-making methods. Based on this, two intervention adoption rates are further defined:

[0074] The endogenous intervention adoption rate ζ α : It represents the probability that an elderly individual actively chooses to adopt a drug based on their own risk perception and personal strategy selection mechanism without the influence of external intervention. This probability reflects the decision-making behavior of an individual without external social or environmental influence. The exogenous intervention adoption rate represents the probability that an elderly individual passively adopts a drug under the influence of a social network or group behavior. In this context, the decision of an individual is affected by the behaviors of other members within the group and the external social environment, leading to possible adjustments to their drug intervention adoption decisions.

[0075] Through this method, a group with heterogeneous risk aversion awareness is introduced, and based on the level of risk aversion awareness and intervention adoption rate of each subgroup, the behavioral patterns of the elderly group when facing drug intervention can be simulated and analyzed more accurately. It not only considers the autonomous decision-making of individuals but also the mutual influence of behaviors within the group. Therefore, it can provide a more precise optimization plan for public health strategies, thereby improving the effectiveness and accuracy of drug intervention strategies.

[0076] In an alternative embodiment, in step S100, based on the group with heterogeneous risk aversion awareness, a corresponding high-order social network structure is constructed, including:

[0077] For example, according to the actual contact data between elderly individuals in the group, a contact matrix W is constructed, where w ij represents the contact probability between elderly individuals i and j. Then, according to the contact probability w ij , a simplicial complex is constructed to further describe the interaction relationship between elderly individuals. To accurately describe these interactions, a high-order social network structure is adopted, which can not only reflect pairwise interactions but also capture the influence of high-order interactions. In the construction of the high-order social network structure, a k-order simplicial complex is introduced, which is used to simulate the interaction between k + 1 elderly individuals, denoted as σ = [v0,…,v kThis model can characterize the interaction relationships between different elderly individuals through simplicial complex graphs. Eventually, the constructed high-order social network structure can more accurately simulate the behavior patterns of different individuals in the group and their decision-making on drug intervention adoption under the influence of the group.

[0078] Step S200: Construct a drug intervention adoption dynamics model based on the high-order social network structure.

[0079] Based on the constructed high-order social network structure, a drug intervention adoption dynamics model is further constructed. This model includes two key stages: In the stage of public drug intervention, each elderly individual calculates the drug intervention adoption strategy for the current cycle based on the benefit in the previous cycle and decides whether to adopt drug intervention. The main purpose of this stage is to simulate the decision-making process of individuals in collective behavior through game theory. In the stage of epidemic spread, reactive drug intervention is carried out on elderly individuals who have not adopted drug intervention, and infection spread is carried out for pair-wise interactions and high-order interactions with other elderly individuals (such as neighbors) in the high-order social network structure.

[0080] Through the construction of this dynamics model, the interactions of elderly individuals in the social network and their behavior decisions in the face of an epidemic can be simulated, thus providing support for public health decision-making.

[0081] In an optional embodiment, in step S200, constructing a drug intervention adoption dynamics model based on the high-order social network structure includes:

[0082] Different behavior strategies and states are defined for each elderly individual. The behavior strategy of each elderly individual includes the drug intervention adoption strategy §1 or the drug intervention rejection strategy §2. And the state of each elderly individual includes: drug intervention rejection and susceptible state S D 、drug intervention adoption and susceptible state S C 、drug intervention adoption and immune state V, infected state I, and recovered state R. Under this state framework, corresponding costs are further defined according to the differences in each state, specifically including: drug intervention cost C V and infection cost C I。The cost of drug intervention represents the price that elderly individuals need to bear when choosing to adopt a drug intervention strategy §1. For example, (1) Preventive drug intervention refers to drug intervention measures taken to prevent the occurrence and spread of diseases without obvious infections or symptoms, or before the outbreak of diseases or epidemics. For example, elderly individuals take drugs in advance to reduce the risk of infection before an epidemic breaks out or after being exposed to potential risk sources. (2) Reactive drug intervention: Drug intervention measures taken after the outbreak of diseases or the emergence of epidemics to alleviate the condition, accelerate recovery, control the spread of diseases, and prevent further spread. For example, elderly individuals take drugs after being infected to treat or relieve the condition, or prevent the condition from deteriorating. The infection cost represents the price that needs to be paid due to infection after refusing drug intervention. For example, if an elderly individual who adopts drug intervention strategy §1 unfortunately gets infected due to reasons such as vaccine failure, the total cost to be paid is C V +C I 。

[0083] In this way, a two-stage drug intervention adoption dynamics model including multiple behavioral strategies, states, and cost factors is constructed. This model can simulate how elderly individuals make decisions under the influence of different environments and strategies, and evaluate the effect of drug intervention on the spread of epidemics.

[0084] In an alternative implementation, in the stage of public drug intervention, elderly individuals obtain benefits based on the cost calculation of the previous season to select the preventive drug intervention strategy for this season and pay the corresponding cost. Specifically, the goal of the public drug intervention stage is for elderly individuals to evaluate whether to continue preventive drug intervention based on the relevant benefits of the previous season.

[0085] If individual u α chooses to continue to adopt preventive drug intervention §1, they need to pay the drug intervention cost C V , which represents the cost that elderly individuals need to bear to obtain drug protection. According to the effectiveness probability ε of drug intervention, if the drug intervention is effective, the state of individual u α will change from the susceptible state to the immune state V; if the drug intervention is ineffective, the individual will remain in the susceptible state S C . If individual u α chooses to reject drug intervention strategy §2, they do not need to pay the drug intervention cost and are classified as the susceptible state S D .

[0086] Through this process, it simulates how elderly individuals choose preventive measures according to their own interests under public drug intervention.

[0087] In the stage of epidemic spread, such as Figure 3As shown, elderly individuals who do not adopt preventive drug intervention strategies respond with reactive drug intervention based on the risk of infection:

[0088] (1) Two state changes brought about by the adoption of reactive drug intervention:

[0089] First, S-state individuals with a risk aversion awareness level of L α may adopt endogenous drug intervention, that is, they receive reactive vaccination with probability ε in an attempt to protect themselves from infection. D Second, in a small group characterized by a k-dimensional simplex, if there is an S-state elderly individual u

[0090] and k - 1 I-state elderly individuals, then the S-state elderly individual receives reactive vaccination with probability D α . D The effect of reactive drug intervention also depends on the effectiveness probability ε of the drug intervention. Therefore, the following state transitions may occur:

[0091] If the drug intervention is completely effective with probability ε, S-state individuals transition to the immune state V, and the cost change is 0 → C

[0092] ; D If the drug intervention is ineffective with probability 1 - ε, S-state individuals transition to the susceptible state S, and the cost change is 0 → C V ;

[0093] If the drug intervention is ineffective with probability 1 - ε, S-state individuals transition to the susceptible state S, and the cost change is 0 → C D If the drug intervention is ineffective with probability 1 - ε, S-state individuals transition to the susceptible state S, and the cost change is 0 → C C ; V ;

[0094] (2) State changes due to the spread of the epidemic:

[0095] Pairwise interaction: S-state individuals interact pairwise with I-state neighbors based on a first-order simplex, and are infected with probability λ1, transitioning to the I state, and the cost change is 0 → C D ; i ;

[0096] Higher-order interaction: S-state individuals interact with k - 1 I-state individuals based on a k-dimensional simplex, and are infected with probability λ D and transition to the I state, and the cost change is 0 → C k ; I ;

[0097] For S-state elderly individuals, their state transitions are similar to those of S-state individuals, being infected through pairwise or higher-order interactions with I-state neighbors and transitioning to the I state, and the cost change is C C → C D ; V → C V ​​+C I 。

[0098] For elderly individuals in the I state, they transition to the R state with a recovery probability γ, and the cost payment remains unchanged.

[0099] In addition, during the epidemic transmission stage, the cost changes of elderly individuals mainly include:

[0100] Drug intervention cost C V : If an individual chooses to conduct reactive drug intervention, regardless of whether the intervention is effective, they need to pay the drug intervention cost C V 。

[0101] Infection cost C I : If an individual is infected, they need to pay the infection cost C I 。

[0102] Total cost: If an individual is still infected after adopting drug intervention, they need to pay the total cost C V +C I 。

[0103] In addition, define the relative drug intervention adoption cost C = C V / C I , which is used to scale the drug intervention cost and the infection cost.

[0104] By combining the endogenous intervention adoption probability, the exogenous intervention adoption probability, the drug intervention effectiveness probability, the transmission probability, and the recovery probability, it is possible to dynamically simulate the decision-making and behavioral responses of elderly individuals during the epidemic transmission process. This process not only reflects how elderly individuals adjust their drug intervention strategies according to their own risk perception and social network environment, but also reveals the inhibitory effect of reactive drug intervention on epidemic transmission.

[0105] Furthermore, after the end of the epidemic transmission stage, elderly individuals will adjust their drug intervention adoption strategies for the next season according to their own benefits and the strategies of other elderly individuals (e.g., neighbors). This process is achieved by defining the combined states and corresponding costs and combining the imitation rules of evolutionary game:

[0106] By defining the combined states and corresponding costs, it is possible to quantify the benefits of elderly individuals in different strategies and states, and thus provide a basis for the imitation rules of evolutionary game. Through the imitation rules, the drug intervention strategies of elderly individuals are further dynamically adjusted, enabling them to gradually form group behaviors in long-term evolution.

[0107] That is, mark the healthy state as H and the infected state as NH. Based on the individual's drug intervention strategy and health state, four combined states are defined:

[0108] CH: Adopt drug intervention and be healthy (immune state V or susceptible state SC )

[0109] CNH: Adopt drug intervention but already infected (infected state I)

[0110] DH: Refuse drug intervention and healthy (susceptible state S D )

[0111] DNH: Refuse drug intervention but already infected (infected state I)

[0112] According to the principle that the higher the cost, the lower the benefit, the benefits in each combined state are defined as follows:

[0113] π CH = -C: The benefit of an individual who adopts drug intervention and is healthy is the negative value of the drug intervention cost C V (relative to the cost C = C V / C I )

[0114] π CNH = -C - 1: The benefit of an individual who adopts drug intervention but is already infected is the negative value of the drug intervention cost C v and the infection cost C I of the negative value.

[0115] π DH = 0: An individual who refuses drug intervention and is healthy does not need to pay any cost, and the benefit is 0.

[0116] π DNH = -1: The benefit of an individual who refuses drug intervention but is already infected is the negative value of the infection cost C I of the negative value.

[0117] Before the epidemic spreads in each season, the elderly individuals will adjust their drug intervention adoption strategies according to their own benefits and the benefits of their neighbors. Specifically, the elderly individual u randomly selects a neighbor node v and compares the overall benefits π u and π v . The probability that the individual u adopts the strategy of node v is given by the Fermi function:

[0118]

[0119] where β represents the imitation intensity, and β ∈ [0, ∞). When β is large, individuals are more inclined to imitate the strategies of neighbors with higher benefits; when β is small, the strategy choices of individuals are more random.

[0120] Through this imitation rule, the elderly individuals can dynamically adjust their drug intervention adoption strategies according to their own and their neighbors' benefits, thus forming group behaviors in the long-term evolution.

[0121] Step S300: Characterize the drug intervention adoption kinetics model using the microscopic Markov chain theory, and iterate the evolution equation to obtain the drug intervention adoption level and the epidemic infection range of the heterogeneous risk-averse awareness group.

[0122] In the modeling of epidemic spread, the microscopic Markov chain theory is used to describe the transition probabilities of individuals between different health states. First, for each elderly individual, the probabilities of health states are defined for them, and the state probability distribution of the individual at different times is obtained. The changes in these states are affected by the epidemic spread process, and the health state transition probability of each individual is closely related to the contact situation in their social network. By constructing the evolution equation of epidemic spread and using the contact matrix to describe the interaction between individuals, the model can calculate the probability that an individual transfers from one state to another during the epidemic spread process. These probabilities are iteratively calculated through the microscopic Markov chain theory to reflect the changes in the group's health state over time. Finally, after sufficient iterations, the stable density distribution of different health states in the group is obtained, providing the necessary data support for the drug intervention adoption model.

[0123] The evolution process of drug intervention adoption is further analyzed based on the results of epidemic spread. Based on the state density distribution obtained from the epidemic spread evolution, the adoption probability of drug intervention is defined for each elderly individual. The decision of drug adoption is not only related to the current health state of the individual but also affected by the epidemic spread situation and other individuals in the social network. The drug adoption probability usually depends on the infection state of the individual, and infected individuals may be more inclined to take drug intervention. By establishing the evolution equation of drug adoption and combining the effectiveness of the drug and the game rules, the model can calculate the drug adoption probability of an individual at each stage. The behavior of each elderly individual will be adjusted according to the epidemic spread dynamics and the behavior of their group, so the adoption of drug intervention is also constantly changing. In this process, the decisions of individuals will reflect the behavior patterns of the group, and over time, the group will gradually tend to a balanced state. Finally, through the iterative calculation of the drug adoption evolution equation, the final level of drug intervention adoption in the group is obtained, and based on this, the epidemic infection range and immunity coverage rate are predicted, providing a theoretical basis for optimizing public health intervention strategies.

[0124] In an alternative embodiment, in step S300, the drug intervention adoption kinetics model is iteratively characterized using the microscopic Markov chain theory and iteratively evolved to obtain the drug intervention adoption level and the epidemic infection range of the heterogeneous risk-averse awareness group, including:

[0125] For the elderly individual i with a risk aversion level of L α whose risk aversion level is L α ,define that at time t in season T, they are in state X ∈ {SD , S C , the probability of the state of V, I, R Among them, S D represents the susceptible state with refusal of intervention, S C represents the susceptible state with acceptance of intervention, V is the immune state, I is the infected state, and R is the recovered state.

[0126] Based on the microscopic Markov chain method, combined with the point-to-point interaction probability λ1, the high-order interaction probability λ2, the recovery probability γ, and the endogenous and exogenous intervention rates corresponding to the risk aversion level of the elderly individuals, an epidemic propagation evolution equation for the state probability of the elderly individuals on the high-order network is established. The specific form is:

[0127] S D Evolution equation of the state (refusal of intervention and susceptible):

[0128]

[0129] S C Evolution equation of the state (susceptible after vaccination):

[0130]

[0131] Evolution equation of the V state (immune):

[0132]

[0133] Evolution equation of the I state (infected):

[0134]

[0135] Evolution equation of the R state (recovered):

[0136]

[0137] Among them, A ij Contact matrix, indicating the contact rate between individuals i and j, I j (t) represents the number of infected individuals j at time t, λ1∑ j A ij I j (t) represents the probability that the healthy elderly individual i α is infected by the neighbor based on the point-to-point interaction at time t. V i α (t) represents the state of the immune population at time t, represents the state of the infected population at time t, represents the elderly individual i α The probability of transitioning to the I state under the influence of the environmental enhanced infection effect through high-order interaction. ζ αS representing the selection strategy §2 D The elderly individual i in state α The probability of spontaneously switching to strategy §1 Representing the elderly individual i affected by the enhanced risk aversion effect of the environment α Undergoing reactive drug intervention adoption, that is, the probability of switching to strategy §1. Considering the effectiveness of drug intervention, the elderly individual i adopting strategy §1 α Switches to S with probability 1 - ε C The probability calculation is Or switches to state V with probability ε, written as For the recovery process, γ represents the probability that the elderly individual i in state I α Recovers to state R

[0138] After establishing the epidemic propagation evolution equation, by iteratively calculating the state probability when t → ∞ for the epidemic propagation evolution equation Determine that the density distribution of elderly individuals with a risk aversion level of L within season T α In the steady state is While the steady-state probability that any individual in the entire elderly population is in state X is Through the state density distribution, the infection range R(T) of the elderly population in season T can be obtained as R(T) = ∑ i R i (T) / N, that is, the proportion of infected individuals V(T) obtained from the epidemic propagation equation is V(T) = ∑ i V(T) / N, where N is the total number of the elderly population

[0139] That is, the elderly individuals may be in four different combined states: CH, CNH, DH, and DNH. These combined states represent the distribution of elderly individuals in different health or living states. The probabilities of elderly individuals in different combined states can be expressed by the following formulas

[0140] (1) Represents the probability that the elderly individual i is in state CH in season T

[0141]

[0142] (2) Represents the probability that the elderly individual i is in state CNH in season T

[0143]

[0144] (3) Represents the probability that the elderly individual i is in state DH in season T

[0145]

[0146] (4) represents the probability that the elderly individual i is in the DNH state in season T:

[0147]

[0148] Where: V i (T): The probability of the immune state of individual i in season T (such as the immune coverage rate after vaccination); and The breakdown state probability of individual i in the susceptible state (such as susceptible after vaccination or susceptible without vaccination); R i (T): The probability of the recovery state of individual i; ε: The drug effectiveness proportion factor (for example, ε = 0.9 means the drug effectiveness is 90%);

[0149] Through these formulas, the joint distribution probability of elderly individuals in different seasons and states can be calculated, so as to understand the distribution of their health or living status in different time periods.

[0150] Based on the state density and the health status of the elderly individuals, the drug intervention adoption probability of the elderly individuals is defined. For the drug intervention adoption strategy of elderly individual i in season T, where, is the probability that individual i adopts drug intervention at time T, V i (T) is the vaccination probability of the individual. The drug intervention adoption level in season T

[0151] Wherein, the drug intervention adoption probability is usually associated with the infection probability and the health status of the individual (for example, susceptible state, infected state, etc.).

[0152] Then, based on the infection density of the elderly individuals, the effectiveness of the drug, and the game rules, a drug adoption evolution equation on the high-order social network structure is constructed. The game rules consider how individuals in the group adjust their drug intervention strategies according to the information in the social network and the behaviors of others. These rules describe the decision-making mechanism of individuals, especially how individuals make choices according to the drug intervention adoption level of the group under different risk aversion awareness.

[0153] That is, define the probability change of the elderly individuals after the T-th round of game Then the probability that elderly individual i chooses to adopt strategy §1 at the beginning of the (T + 1)-th epidemic season can be written as represents the probability that individual i chooses drug intervention at the next moment T + 1, based on the current infection risk and the effectiveness of the drug. It can be calculated as:

[0154]

[0155] Where, F(πx ←π Y ), where \(X, Y\in\{CH, CNH, DH, DNH\}\) describes the probability that an elderly individual in the \(X\) state imitates the strategy of an elderly individual in the \(Y\) state, and \(F(\pi X ←π Y ) can be expressed as:

[0156]

[0157] When \(T ightarrow \infty\), the drug intervention adoption strategy of the elderly individuals reaches an evolutionary equilibrium, and the final infection range \(R\) of the elderly population is output * =\lim T→∞ R(T), the immunization coverage rate \(V * =\lim T→∞ V(T) and the drug intervention adoption level

[0158] Step S400, according to the drug intervention adoption level and the epidemic infection range, analyze the impact of the heterogeneous risk aversion awareness group on the drug intervention adoption level and the epidemic transmission range, and optimize the drug adoption strategy according to the impact.

[0159] Exemplarily, consider how elderly individuals with different levels of risk aversion awareness in the group affect the drug intervention adoption level and the epidemic transmission range. By analyzing the drug intervention adoption level and the epidemic infection range, the impact of different risk aversion awareness groups on the epidemic transmission and immunization coverage is revealed. Specifically, as the proportion of individuals with high risk aversion awareness in the group increases, the drug intervention adoption level usually increases, and the epidemic transmission range will decrease accordingly. According to these analysis results, the drug intervention strategy can be optimized to ensure more targeted drug intervention promotion measures, maximize the immunization coverage rate of the group, and thus effectively control the spread of the epidemic.

[0160] In an alternative embodiment, in step S400, according to the drug intervention adoption level and the epidemic transmission range, optimize the drug adoption strategy, including:

[0161] By simulating different risk aversion levels and the proportion of elderly individuals with risk aversion awareness in the group, analyze their impact on the drug intervention adoption level and the epidemic transmission range. Specifically, if the proportion of individuals with high risk aversion awareness in the group increases, the overall drug intervention adoption level of the group tends to increase, thereby reducing the epidemic transmission range. This result can be used to optimize the group's health status by adjusting the drug intervention strategy, ensuring that the vaccine promotion more effectively covers the groups most in need.

[0162] As Figure 4 shown, for example, Figure 4In (a)-(f), it shows that the increase in the proportion of the risk-averse group will, to a certain extent, increase the immunization coverage rate of the entire elderly population. However, simply relying on expanding the proportion of the risk-averse group to increase the drug intervention coverage rate of the entire elderly group has limited effects. Figure 4 As shown in (c) and (e), when the risk-averse group has a low risk-averse probability, that is, a low probability of adopting drug intervention, then further expanding the proportion of the risk-averse group has a relatively less obvious impact on the immunization coverage rate. Figure 4 In (a)-(f), Figure 5 In (g)-(l), Figure 6 The three groups of results in (m)-(r) show that the promotion effect of a single increase in the risk-averse level on the immunization coverage rate of the elderly population is very limited. In particular, when the proportion of the risk-averse group is very small ( Figure 4 as shown in (a)(b) in, Figure 5 in (g)(h), and Figure 6 in (m)(n)), a single increase in the risk-averse level has almost no impact on the immunization coverage rate of the elderly population. However, when the risk-averse level of the risk-averse group increases and at the same time the proportion of the risk-averse group is expanded, it can completely cover the entire elderly population with drug intervention, Figure 5 as shown in (g)-(l) and Figure 6 in (m)-(r). That is to say, the simultaneous increase in the proportion of the risk-averse group and the individual's risk-averse level can more effectively promote the adoption of drug intervention by the elderly group and inhibit infection. Finally, Figure 7 (s)-(x) shows that as long as all individuals in the elderly population have risk aversion, even if the risk-averse levels of most elderly individuals are very low, it can greatly increase the immunization coverage rate of the entire elderly population and control the spread of the epidemic. This result indicates that by adjusting the drug intervention strategy and optimizing the group's health status, it is possible to ensure that the vaccine promotion can more effectively cover the groups in greatest need.

[0163] The drug intervention adoption behavior of elderly individuals is not only affected by personal risk perception but also by the behavior of others in the social network. By introducing a high-order social network, the enhanced infection effect is simulated. This effect means that when an individual has high-order interactions with multiple social network members (such as close family members or friends), the perceived risk of infection will increase, thereby promoting their adoption of drug intervention. For example, in an environment with a high risk of infection, individuals may increase their willingness to be vaccinated through communication and mutual influence in the social network, thus increasing the immunization coverage rate of the group.

[0164] For example, as Figure 8 shown, Figure 8In (a)(b) and (c)(d), when the relative cost of drug intervention adoption is low, the enhanced infection effect brought by higher-order interactions instead promotes the adoption of drug intervention among the elderly population to a certain extent and inhibits the spread of the epidemic. As the relative cost of drug intervention adoption continues to rise, the enhanced infection effect no longer promotes the adoption of drug intervention and instead leads to an increase in the infection range among the elderly population. This is because the increase in the enhanced infection effect leads to an increase in the perceived infection risk of elderly individuals, and coupled with the low relative cost of drug intervention adoption, it thus promotes the adoption of drug intervention. However, when the relative cost of drug intervention adoption rises to a sufficiently large extent, although the enhanced infection effect brings a great infection risk, the high cost of drug intervention causes elderly individuals to abandon the adoption of drug intervention, thereby leading to the spread of the epidemic. Figure 9 In (e)(f) and (g)(h), that is, when the avoidance awareness level of elderly individuals rises, the same experimental results are obtained.

[0165] The enhanced risk aversion effect is also an important factor affecting the adoption of drug intervention. In a higher-order social network, an individual's risk aversion awareness will be strengthened through social interactions. Especially when faced with a relatively high cost of drug intervention, an individual may adjust their decision based on the behavior of other members within the group. For example, when the proportion of individuals with a high risk aversion awareness increases, they may increase the adoption rate of drug intervention through exogenous interventions (such as listening to health advice within the group), thereby increasing the immune coverage rate and effectively controlling the spread of the epidemic. The modeling of this effect provides strong support for the collective behavior response brought about by higher-order interactions.

[0166] For example, as Figure 10 shown, for instance, Figure 10 in (a)-(d), when elderly individuals refuse to adopt endogenous intervention, the enhanced risk aversion effect of the environment enables a relatively high immune coverage rate to still exist among the elderly population when the relative cost C of drug intervention adoption is very high, thereby reducing the infection range to a certain extent, especially when there are a large number of elderly individuals with risk aversion awareness. This phenomenon demonstrates the result of elderly individuals with risk aversion awareness implementing the adoption of exogenous intervention and thus getting vaccinated under the influence of the enhanced risk aversion effect. When elderly individuals adopt endogenous intervention, for example, Figure 11 in (e)-(h), the results are consistent with those in Figure 8 in (a)-(d). Through these studies, we can accurately simulate and analyze the behavior of different groups with risk aversion awareness in drug intervention, and further provide a theoretical basis for the optimization of public health policies.

[0167] The method of the embodiments of the present application dynamically simulates the drug intervention adoption behaviors of elderly groups with different risk aversion levels in the spread of epidemics by dividing the risk aversion awareness levels of elderly individuals and combining high-order social networks and a drug intervention adoption dynamics model. By quantifying the risk preferences of individuals and the social interaction effects, it can accurately reveal the differences in individual behaviors and provide a theoretical basis for formulating more targeted public health strategies. In particular, considering the influence of the collective environment on the individual infection probability and decision-making mechanism, it expands the microscopic Markov chain method and effectively tracks the evolution trajectories of individual behaviors and social networks, providing strong quantitative support for optimizing vaccine promotion strategies and public health policies.

[0168] Figure 12 FIG. shows a schematic structural diagram of a drug adoption behavior analysis device according to an embodiment of the present application. Exemplarily, the device includes:

[0169] A social network structure construction module 121, configured to construct a corresponding high-order social network structure based on a heterogeneous risk aversion awareness group;

[0170] A dynamics model construction module 122, configured to construct a drug intervention adoption dynamics model according to the high-order social network structure; the drug intervention adoption dynamics model includes a public drug adoption stage and an epidemic spread stage;

[0171] A range and coverage acquisition module 123, configured to iteratively evolve the drug intervention adoption dynamics model by using the microscopic Markov chain theory to obtain the drug intervention adoption level and the epidemic infection range of the heterogeneous risk aversion awareness group;

[0172] A drug adoption strategy optimization module 124, configured to analyze the influence of the heterogeneous risk aversion awareness group on the drug intervention adoption level and the epidemic spread range according to the drug intervention adoption level and the epidemic infection range, and optimize the drug adoption strategy according to the influence.

[0173] It can be understood that the device of this embodiment corresponds to the method of the above embodiment, and the optional items in the above embodiment are equally applicable to this embodiment, so they will not be described repeatedly here.

[0174] The present application also provides a computer device. Exemplarily, the computer device includes a processor and a memory, wherein the memory stores a computer program, and the processor runs the computer program to enable the computer device to execute the above drug adoption behavior analysis method or the functions of each module in the above drug adoption behavior analysis device.

[0175] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, etc., and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application.

[0176] The memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Among them, the memory is used to store a computer program, and after receiving an execution instruction, the processor can execute the computer program accordingly.

[0177] The present application also provides a computer storage medium for storing the computer program used in the above computer device. Among them, the computer storage medium can be a readable storage medium, a non-volatile storage medium, or a volatile storage medium. For example, the computer storage medium can include, but is not limited to: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc, etc., various media that can store program codes.

[0178] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and structural diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the part of the module, program segment, or code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the blocks can occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the structural diagram and / or flowchart, as well as the combination of blocks in the structural diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0179] In addition, in each embodiment of the present application, the various functional modules or units can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0180] If the above functions are implemented in the form of software function modules and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application.

[0181] As described above, the above are only the specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application.

Claims

1. A method for analyzing drug adoption behavior, characterized in that: The method comprises: Based on heterogeneous risk aversion consciousness groups, the corresponding high-order social network structure is constructed; According to the high-order social network structure, a drug intervention adoption dynamics model is constructed; the drug intervention adoption dynamics model includes a public drug intervention stage and an epidemic transmission stage; The drug intervention adoption dynamics model is characterized by using micro-Markov chain theory, and the evolution equation is iterated to obtain the drug intervention adoption level and epidemic infection range of the heterogeneous risk aversion group; According to the drug intervention adoption level and the epidemic infection scope, the impact of the heterogeneous risk aversion awareness group on the drug intervention adoption level and the epidemic spread scope is analyzed, and the drug adoption strategy is optimized based on the impact.

2. The drug adoption behavior analysis method according to claim 1, characterized in that: Before constructing the corresponding high-order social network structure, the following steps are included: Introducing an elderly group with heterogeneous risk aversion awareness, dividing the elderly group into a plurality of heterogeneous risk aversion awareness subgroups according to the heterogeneous risk aversion awareness level of each elderly individual in the elderly group, and defining an endogenous intervention adoption rate and an exogenous intervention adoption rate for the heterogeneous risk aversion awareness level of each elderly individual in the subgroup; Among them: the endogenous intervention adoption rate represents the probability that the elderly individual actively chooses to adopt drugs based on his or her own risk perception and strategy selection mechanism without the influence of external intervention; the exogenous intervention adoption rate represents the probability that the elderly individual passively adopts drugs based on the perceived group infection risk pressure in the high-order social network structure.

3. The drug adoption behavior analysis method according to claim 1, characterized in that: The corresponding high-order social network structure is constructed based on the heterogeneous risk aversion awareness group, including: Based on the division of the heterogeneous risk-averse consciousness subgroups, a contact matrix is ​​constructed, and a simplicial complex model is used to characterize the interactions between each elderly individual in the high-order social network structure, wherein the first-order simplicial is used to characterize the point-to-point interactions between the elderly individuals, and the second-order and higher simplicial are used to characterize the higher-order interactions between the elderly individuals.

4. The drug adoption behavior analysis method according to claim 1, characterized in that: The step of constructing a drug intervention adoption dynamics model based on the high-order social network structure includes: Defining a behavior strategy and state for each elderly individual, wherein the behavior strategy includes adopting drug intervention and rejecting drug intervention; the state includes susceptible state, immune state, infected state, and recovered state; According to the behavior strategy, a corresponding cost is defined for each different state of the elderly individual, and a combined state benefit is defined based on the cost; the cost includes a drug intervention cost and an infection cost, wherein the drug intervention cost represents the price that the elderly individual needs to pay for adopting preventive drug intervention or reactive drug intervention, and the infection cost represents the price that the elderly individual needs to pay for refusing drug intervention; the combined state benefit is used to drive the elderly individual to update the strategy; Based on the behavioral strategy, the state and the cost, a two-stage drug intervention adoption dynamics model is constructed.

5. The drug adoption behavior analysis method according to claim 4, characterized in that: The two-stage drug intervention adoption dynamics model is constructed based on the behavior strategy, the state and the cost, including: In the public drug intervention stage, the elderly individual uses the Fermi function to imitate the strategies of other elderly individuals based on the combined state benefits of the previous season to select the preventive drug intervention strategy of the current season and pay the corresponding cost; During the epidemic transmission stage, the reactive drug intervention is performed on the elderly individuals who have not adopted the preventive drug intervention strategy, and the corresponding costs are paid; the reactive drug intervention is performed through the endogenous intervention adoption rate and the exogenous intervention adoption rate.

6. The drug adoption behavior analysis method according to claim 1, characterized in that: The micro-Markov chain theory is used to characterize the drug intervention adoption dynamics model, and the evolution equation is iterated to obtain the drug intervention adoption level and epidemic infection range of the heterogeneous risk aversion awareness group, including: Defining state probabilities for the elderly individuals, and obtaining state probability distributions of each elderly individual at different times; Processing the state probability distribution to construct an epidemic propagation evolution equation on the high-order social network structure; Iterating the epidemic transmission evolution equation to obtain the state density distribution of the elderly individuals in a stable state; According to the state density distribution, the infection range of the elderly group is calculated, and the probability of drug intervention adoption of the elderly individuals is defined; According to the infection density of the elderly individuals, the effectiveness of the drugs and the game rules, constructing the drug adoption evolution equation on the high-order social network structure; The drug adoption evolution equation is iterated to calculate the drug intervention adoption level until a balance state is reached, and the final infection range and final immunization coverage are output.

7. The drug adoption behavior analysis method according to claim 1, characterized in that: Optimizing the drug adoption strategy based on the drug intervention adoption level and the scope of epidemic spread includes: Simulating different levels of heterogeneous risk aversion and the proportion of elderly people with heterogeneous risk aversion awareness to obtain the impact on the level of drug intervention adoption and the scope of epidemic spread; Studying the impact of the enhanced contagion effect in the high-order social network structure on the drug intervention adoption behavior of the elderly individuals, wherein the enhanced contagion effect increases the drug intervention adoption level of the elderly individuals through high-order interactions; The influence of the enhanced risk aversion effect in the higher-order social network structure on the drug intervention adoption behavior of the elderly individuals is studied, wherein the enhanced risk aversion effect increases the drug intervention adoption level of the elderly individuals through higher-order interactions.

8. A drug adoption behavior analysis device, characterized in that: The device comprises: A social network structure building module is used to build a corresponding high-order social network structure based on heterogeneous risk aversion awareness groups; A dynamic model building module, used to build a drug intervention adoption dynamic model according to the high-order social network structure; the drug intervention adoption dynamic model includes a public drug adoption stage and an epidemic transmission stage; A scope and coverage acquisition module is used to characterize the drug intervention adoption dynamics model using micro-Markov chain theory and iterate the evolution equation to obtain the drug intervention adoption level and epidemic infection scope of the heterogeneous risk aversion awareness group; The drug adoption strategy optimization module is used to analyze the impact of the heterogeneous risk aversion awareness group on the drug intervention adoption level and the epidemic spread scope according to the drug intervention adoption level and the epidemic infection scope, and optimize the drug adoption strategy according to the impact.

9. A computer device, characterized in that: The computer device comprises a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the drug adoption behavior analysis method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The device stores a computer program, which, when executed on a processor, implements the drug adoption behavior analysis method according to any one of claims 1 to 7.

Images