An insurance drop-off strategy generation method, device, equipment and medium

Abstract

The application discloses an insurance putting strategy generation method, device and equipment and a medium, and is applied to the financial field. The method comprises the following steps: determining objective individual information and subjective psychological information of an intelligent agent based on portrait data of an insurance customer sample; simulating a social path of the insurance customer sample to determine an information propagation strategy of the intelligent agent; fusing the objective individual information, the subjective psychological information, the information propagation strategy and insurance information of an insurance to be put to determine an insurance purchase strategy of the intelligent agent; simulating the insurance to be put based on the insurance purchase strategy and a preset putting strategy of the insurance to be put to determine a plurality of strategy influence factors of the preset putting strategy and corresponding putting simulation results; and determining a target putting strategy based on the preset putting strategy, the strategy influence factors and the putting simulation results. The intelligent agent information is determined through the portrait data, the intelligent agent can fully simulate the decision and behavior of a user in an actual scene, and the accuracy of the target putting strategy is improved.

Description

Technical Field

[0001] This invention relates to the fields of insurance strategy generation and financial technology, and particularly to a method, apparatus, equipment and medium for generating insurance strategies. Background Technology

[0002] Currently, the insurance industry still widely relies on traditional methods such as questionnaires, interviews, and small-scale pilot programs for strategy validation when innovating products and promoting them. These methods have limited sample sizes, struggle to cover diverse customer groups, and are time-consuming, costly, and unable to quickly respond to market changes. While static prediction models can infer trends based on historical data, they lack characterization of individual interactions and social communication mechanisms, leading to predictions that deviate from reality. This is especially true in new product promotion, where the lack of historical sales data renders traditional methods ineffective. Therefore, there is an urgent need for an insurance strategy generation method that aligns with real user behavior patterns and social interaction characteristics. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and medium for generating insurance delivery strategies, in order to solve the problem in related technologies that it is difficult to accurately simulate customers' real decision-making behavior and social communication effects when generating insurance strategies, resulting in poor adaptability and low optimization efficiency of delivery strategies.

[0004] Firstly, this disclosure provides a method for generating an insurance placement strategy, including: Based on the profile data of insurance customer samples, determine the objective individual information and subjective psychological information of the intelligent agent; Simulate the social paths of insurance customer samples to determine the information dissemination strategies of intelligent agents; By integrating objective individual information, subjective psychological information, information dissemination strategies, and insurance information to be offered, the insurance purchasing strategy of the intelligent agent can be determined. Based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, the delivery of the insurance to be delivered is simulated to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results. Based on the preset delivery strategy, strategy influencing factors, and delivery simulation results, the target delivery strategy is determined.

[0005] Secondly, this disclosure provides an insurance placement strategy generation device, comprising: The agent information determination module is used to determine the objective individual information and subjective psychological information of the agent based on the profile data of insurance customer samples. The propagation strategy determination module is used to simulate the social paths of insurance customer samples and determine the information propagation strategy of the agent. The purchase strategy determination module is used to integrate objective individual information, subjective psychological information, information dissemination strategies, and insurance information to be offered to determine the insurance purchase strategy of the intelligent agent. The delivery simulation module is used to simulate the delivery of insurance based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, and to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results. The target determination module is used to determine the target delivery strategy based on the preset delivery strategy, strategy influence factors, and delivery simulation results.

[0006] Thirdly, this disclosure provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the aforementioned insurance delivery strategy generation method. Fourthly, this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described insurance delivery strategy generation method.

[0007] In the above-mentioned scheme implemented by the insurance delivery strategy generation method, device, equipment, and medium, objective individual information and subjective psychological information of the intelligent agent are determined based on profile data of insurance customer samples; the social path of the insurance customer samples is simulated to determine the information dissemination strategy of the intelligent agent; the objective individual information, subjective psychological information, information dissemination strategy, and insurance information of the insurance to be delivered are integrated to determine the insurance purchase strategy of the intelligent agent; based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, the delivery of the insurance to be delivered is simulated to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results; based on the preset delivery strategy, strategy influencing factors, and delivery simulation results, the target delivery strategy is determined. This method uses profiling data to determine the objective individual information and subjective psychological information of the intelligent agent, enabling the agent to more realistically simulate user behavior habits. By integrating social path simulation results with insurance product characteristics, it accurately portrays the agent's behavioral insurance purchasing strategies in different communication scenarios. This allows for a comprehensive evaluation of the market performance and user response of products under different insurance delivery strategies during simulated deployment using the intelligent agent, ultimately achieving a scientific selection of target delivery strategies. This improves the accuracy and scientific rigor of insurance delivery strategy formulation, effectively reduces trial-and-error costs, enhances product market adaptation efficiency, and provides strong support for the precise and intelligent deployment of insurance products. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart of an insurance placement strategy generation method according to an embodiment of the present invention; Figure 2 This is another flowchart of the insurance placement strategy generation method in one embodiment of the present invention; Figure 3 This is another flowchart of the insurance placement strategy generation method in one embodiment of the present invention; Figure 4 This is another flowchart of the insurance placement strategy generation method in one embodiment of the present invention; Figure 5 This is another flowchart of the insurance placement strategy generation method in one embodiment of the present invention; Figure 6 This is another flowchart of the insurance placement strategy generation method in one embodiment of the present invention; Figure 7 This is a schematic diagram of an insurance delivery strategy generation device in one embodiment of the present invention; Figure 8 This is a schematic block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0010] In one embodiment, such as Figure 1 As shown, an insurance placement strategy generation method is provided, including the following steps: S101, Based on the profile data of insurance customer samples, determine the objective individual information and subjective psychological information of the intelligent agent; S102, Simulate the social path of insurance customer samples to determine the information dissemination strategy of the agent; S103, integrate objective individual information, subjective psychological information, information dissemination strategies, and insurance information to be applied for to determine the insurance purchase strategy of the intelligent agent; S104. Based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, the delivery of the insurance to be delivered is simulated to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results. S105. Based on the preset delivery strategy, strategy influence factors, and delivery simulation results, determine the target delivery strategy.

[0011] As an example, in step S101, cluster analysis and feature extraction can be performed on the multidimensional data of insurance customer samples to construct profile data containing objective attributes such as age, income, occupation, and risk preference, as well as subjective characteristics such as cognitive bias and consumption concept. Then, the profile data can be used to assign values ​​to the intelligent agent, thereby determining the objective individual information and subjective psychological information of the intelligent agent, so that the intelligent agent can more accurately imitate the decision-making behavior of real users and avoid behavioral bias caused by static individual modeling.

[0012] The profile data may include the user's personal identity information (such as age, gender, occupation, income level and other identity attribute data), risk preference for insurance (such as conservative, aggressive or conservative), level of insurance knowledge (such as understanding of the functions of critical illness insurance, accident insurance and other functions), and past insurance records, etc. This disclosure does not limit this.

[0013] Objective individual information refers to the objective individual attributes given to the intelligent agent, which can include quantifiable characteristics such as age, occupation, income, and risk preference. Subjective psychological information refers to the non-quantifiable characteristics given to the intelligent agent, such as emotional tendencies, cognitive biases, and decision-making psychology, such as the intensity of subjective perception of risk, herd mentality, or the upper limit of insurance loss tolerance. Objective individual information and subjective psychological information together constitute the core of the intelligent agent's behavior driving force.

[0014] As an example, in step S102, the propagation path of insurance customers in a real social environment can be simulated based on social network topology, information diffusion model, etc., thereby identifying key propagation nodes and group influence patterns, determining the information propagation strategy of the agent, realizing dynamic modeling of information propagation path, and avoiding the problem of rigid propagation path caused by static propagation network.

[0015] Furthermore, by using objective individual information and subjective psychological information from each agent, we can simulate the information dissemination methods of each agent in social networks, such as the strength of their willingness to forward insurance-related information, the credibility assessment of recommendations from relatives and friends, and the delay effect of information dissemination. This allows us to characterize dynamic dissemination behaviors that conform to the psychological characteristics of real users, and to determine the dynamic dissemination behaviors as the information dissemination strategies of the agents.

[0016] As an example, in step S103, based on the agent's objective individual information and the insurance information to be offered (such as coverage, premium level, and claims mechanism), the agent's objective evaluation of the insurance product can be obtained. Based on the agent's subjective psychological information, through information dissemination strategies, the agent's subjective decision-making tendency under social influence can be obtained. Finally, by integrating the objective evaluation results and subjective decision-making tendencies, the insurance purchase strategy for each agent is determined, achieving a two-dimensional model of the individual decision-making process. By introducing a coupling mechanism between rationality and emotion, it not only reflects the impact of economic feasibility on insurance behavior but also embodies the irrational deviations caused by emotional contagion, trust transmission, and group pressure in social interactions. This allows for the simulation of the dynamic selection paths of real users in complex information environments, improving the predictive accuracy and intervention effectiveness of insurance product market deployment strategies.

[0017] As an example, in step S104, the preset delivery strategy can be dynamically adjusted to use the intelligent agent to perform multiple rounds of simulation and simulation to obtain the delivery effect data of the insurance to be delivered under different combinations of delivery factors (such as delivery channel weight, promotion timing rhythm, target audience screening threshold, etc.). Then, the influence of each strategy influence factor on the delivery effect can be quantified to determine the delivery simulation results corresponding to different strategy influence factors, realize the quantitative analysis of different delivery factors, and provide data support for strategy optimization.

[0018] As an example, in step S105, based on the results of multiple rounds of simulation, by comparing the differences in the delivery effects under different strategy combinations, the factors that have a positive impact on the overall delivery effect among multiple influencing factors can be identified. By prioritizing the positive influencing factors and analyzing the combination effect, the optimal strategy combination can be selected, thereby generating the target delivery strategy.

[0019] In summary, this disclosure proposes a method for generating insurance delivery strategies, comprising: determining the objective individual information and subjective psychological information of an intelligent agent based on profile data of an insurance customer sample; simulating the social paths of the insurance customer sample to determine the information dissemination strategy of the intelligent agent; integrating the objective individual information, subjective psychological information, information dissemination strategy, and insurance information of the insurance to be delivered to determine the insurance purchase strategy of the intelligent agent; performing delivery simulation on the insurance to be delivered based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results; and determining the target delivery strategy based on the preset delivery strategy, strategy influencing factors, and delivery simulation results. This method uses profiling data to determine the objective individual information and subjective psychological information of the intelligent agent, enabling the agent to more realistically simulate user behavior habits. By integrating social path simulation results with insurance product characteristics, it accurately portrays the agent's behavioral insurance purchasing strategies in different communication scenarios. This allows for a comprehensive evaluation of the market performance and user response of products under different insurance delivery strategies during simulated deployment using the intelligent agent, ultimately achieving a scientific selection of target delivery strategies. This improves the accuracy and scientific rigor of insurance delivery strategy formulation, effectively reduces trial-and-error costs, enhances product market adaptation efficiency, and provides strong support for the precise and intelligent deployment of insurance products.

[0020] In one embodiment, such as Figure 2 As shown, step S101 involves determining the objective individual information and subjective psychological information of the intelligent agent based on the profile data of the insurance customer sample, including: S201, Randomly extract identity attribute data and risk characteristic data from the profile data to determine objective individual information; S202, Based on the psychological cognitive data in the profile data, subjective psychological information is determined through inference statistical models.

[0021] As an example, in step S201, random sampling can be performed from identity attribute data such as age, occupation, and income level, and combined with risk characteristic data such as health status and past insurance records to construct an objective individual profile of the intelligent agent, ensuring that it covers the typical characteristics of different user groups, thereby improving the representativeness and generalization ability of the simulation results.

[0022] For example, based on the combination of different age groups and income levels, typical user prototypes such as young professionals and middle-aged high-net-worth individuals can be set up. Combined with their historical insurance risk characteristics data (such as lower-income groups tending to prefer basic protection products, while high-income groups are more concerned with high-risk, high-coverage investment-type insurance), the parameters of the intelligent agent can be configured to match the actual user behavior characteristics. That is, the intelligent agent can be configured with personalized objective individual information so that it can more realistically reflect the behavioral preferences of the target customer group in the simulation.

[0023] For example, a template for configuring objective individual information of an agent could be something like, "You are a [age] [age] [age] [occupation] [gender] [monthly income] [income level] [yuan] [city type] [city]. Your attitude toward insurance is [risk preference description] [and you have previously purchased [existing insurance products] []."

[0024] As an example, in step S202, based on psychological cognitive data (such as insurance loss threshold, trust source preference, social influence sensitivity and delayed gratification tendency, etc.), statistical inference models such as Bayesian hierarchical models are used to infer an individual's decision-making tendency in uncertain situations, thereby quantifying their subjective psychological characteristics.

[0025] For example, by using psychological cognitive data, the mean level of trust in advertising among urban white-collar workers in the insurance customer sample and the distributional differences in trust in advertising among the insurance customer sample can be determined. This data can then be input into a Bayesian hierarchical model to generate the advertising trust level of each agent (i.e., the aforementioned subjective psychological characteristics). This enables differentiated modeling of the subjective psychological characteristics of each agent, allowing group behavior to reflect both common patterns and individual differences.

[0026] In one embodiment, such as Figure 3 As shown, S102 involves simulating the social paths of insurance customer samples to determine the agent's information dissemination strategy, including: S301, based on objective individual information, determines the social patterns of insurance customer samples, and uses these social patterns to determine the information dissemination path of the intelligent agent; S302, Based on subjective psychological information, the information propagation model is trained, and based on the training results, the information propagation mechanism of the agent is determined; S303, based on the information dissemination path and information dissemination mechanism, determine the information dissemination strategy.

[0027] As an example, in step S301, based on the attributes of the agent's objective individual information, such as age, occupation, and type of city of residence, and combined with real-world social network data (such as the distribution of WeChat friend circles and the frequency of community interactions), social network connection rules are constructed to simulate the contact frequency and information transmission probability between individuals. For example, agents of the same age and with similar income levels are more likely to establish high-frequency interaction relationships, thus forming strong connection paths for information dissemination, while individuals across occupations or city types exhibit weak connections but have broader diffusion potential.

[0028] For example, the propagation path of an agent can be determined by the following formula: in, Indicates the propagation path, This indicates transmission through homogeneity (similarities in occupation, region, age, etc.). This indicates that the message is spread through a key leader (salesperson, blogger, etc.). This indicates random (non-social) propagation.

[0029] As an example, in step S302, based on subjective psychological information, the agent's information receiving and forwarding behavior can be modeled using information propagation models such as the Susceptible-Infected-Recovered (SIR) model and the Independent Cascade (IC) model. Combining psychological characteristics such as the agent's trust source preference and sensitivity to social influence, the agent's acceptance probability and propagation probability of insurance are simulated, thereby determining the agent's information propagation mechanism.

[0030] For example, the information propagation mechanism of an agent can be determined by the following formula: in, This represents the probability that agent j receives information from agent i at time t+1. Let represent the propagation probability at time t. Indicates the propagation intensity coefficient. This represents the information activation state of agent i at time t, which is the decision state of the agent regarding whether to receive the insurance information to be distributed and participate in its dissemination. When agent i is in an active state at time t, its dissemination behavior will affect the probability of agent j receiving the insurance information to be distributed and the probability of disseminating the insurance information to agent j at time t+1. This process is modulated by individual psychological preferences, causing different agents to exhibit differentiated information dissemination capabilities in the same social path, thereby achieving precise and dynamic information dissemination strategy modeling.

[0031] As an example, in step S303, based on the topology of the information propagation path and the relationship between agents, combined with the propagation mechanism parameters, the simulation of multi-channel propagation by agents is realized, including scenarios such as online social platforms, offline community activities and recommendations from relatives and friends. This allows agents to simulate choosing propagation paths based on their social influence and emotional tendencies during the information diffusion process, thereby personalized determination of the information propagation strategy of each agent.

[0032] For example, if the information dissemination path of an intelligent agent is mainly online social networks, and the dissemination mechanism is that there is a 50% probability that the agent will forward information to related intelligent agents within 24 hours after the initial contact, then it can be determined that the information dissemination strategy of the intelligent agent is to prioritize pushing insurance product content to nodes with strong connections through social media.

[0033] In an optional embodiment, the conformity level and trust source preference of the agent can be determined based on the agent's subjective psychological information, and a corresponding trust coefficient can be assigned to it. This further endows it with behavioral regulation factors in information dissemination, thereby quantifying the influence of the agent's psychological characteristics on the information diffusion path. For example, if an agent has a high degree of conformity and trusts recommendations from acquaintances (i.e., for this agent, recommendations from friends are more effective than advertisements from strangers), it can be assigned a higher trust coefficient to indicate that it is more inclined to accept and forward information from close relationship nodes during information dissemination. This further personalizes the agent's information dissemination strategy and improves the accuracy of individual differences in the simulation results.

[0034] In one embodiment, such as Figure 4 As shown, step S104, which involves fusing objective individual information, subjective psychological information, information dissemination strategies, and insurance information to be offered to determine the agent's insurance purchase strategy, includes: S401, based on objective individual information and insurance information, determine the rational decision score of the intelligent agent; S402, based on subjective psychological information and information dissemination strategies, simulates the social influence parameters of the insurance to be sold, and determines the agent's emotional decision score based on the social influence parameters; S403 integrates rational decision-making scores and emotional decision-making scores to determine the agent's insurance purchase strategy.

[0035] As an example, in step S401, based on objective individual information such as the agent's age, income, and occupational risk level, and combined with insurance information such as the coverage, premium amount, and claims terms of the insurance product to be delivered, a multi-dimensional utility function (such as indicators like premium-to-profit ratio and risk coverage matching degree) can be constructed to quantify the objective matching degree between the insurance product and individual needs, and form a rational decision-making score.

[0036] For example, if an AI agent is assigned the characteristics of being 35 years old, having an annual income of 200,000 yuan, and having a mortgage, and if it is faced with a critical illness insurance product, and the coverage of major diseases is included and the premium is less than 5% of its annual income, the system will determine that the insurance is highly matched with its objective needs, and thus assign it a high rational decision score.

[0037] As an example, in step S402, the propagation effect of the insurance to be delivered in different propagation paths can be judged based on the agent's subjective psychological information and information propagation strategy, thereby determining the influence parameter of the insurance to be delivered in the agent, and then quantifying the degree of influence parameter on the decision to purchase insurance, forming an emotional decision score.

[0038] It should be understood that the emotional decision score can change dynamically. For example, when the agent receives positive reviews of the same insurance product from multiple close friends on a social network, its emotional decision score will increase nonlinearly with the increase in the number of recommenders. Especially when the preference for the source of trust is strong, the influence of recommendations from acquaintances will significantly amplify its willingness to buy.

[0039] In one optional embodiment, parameters such as advertising attractiveness score, brand trust accumulation, and social platform interaction frequency for each agent can be determined based on the agent's subjective psychological information and information dissemination strategy. These multiple parameters are then defined as social influence parameters, and the social influence parameters are weighted and fused to determine the emotional decision score.

[0040] For example, for an agent with an initial emotional decision score of 60 (indicating a neutral attitude towards the insurance product) and a trust preference based on recommendations from acquaintances, upon receiving positive recommendations from three close friends, their emotional decision score can jump to over 85 points due to the activation of their trust preference, significantly enhancing their purchase intention. This dynamic evolution mechanism fully integrates individual psychological traits and social communication characteristics, making the agent's insurance purchase strategy closer to real human decision-making behavior, achieving higher accuracy in behavioral prediction and intervention response in a simulated environment.

[0041] As an example, in step S403, the rational decision-making score and the emotional decision-making score can be weighted and fused to determine the overall purchase probability (i.e., overall purchase strategy) of each agent for the insurance to be offered. The weighting coefficients can be personalized according to the agent's risk preference and decision-making mode type.

[0042] For example, a comprehensive purchasing strategy can be determined using the following formula: in, This indicates a comprehensive purchasing strategy. and These represent the weighting coefficients, Indicates a score for rational decision-making. σ represents the emotional decision score, and σ represents the Sigmoid activation function, which is used to map the overall score to a probability value, ensuring that the output is between 0 and 1.

[0043] In one embodiment, such as Figure 5 As shown, step S104, which involves simulating the insurance purchase strategy and the preset delivery strategy for the insurance to be delivered, determines multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results, including: S501, update the preset delivery strategy to obtain multiple experimental delivery strategies; S502, based on the insurance purchase strategy, uses an intelligent agent to determine the market penetration parameters of the insurance to be deployed under the preset deployment strategy and the experimental deployment strategy respectively; S503 defines the strategy difference between the experimental deployment strategy and the preset deployment strategy as the strategy influencing factor, and defines the market penetration parameter as the deployment simulation result.

[0044] As an example, in step S501, the preset delivery strategy can consist of multiple delivery dimensions, including advertising channel selection, promotion time allocation, target audience targeting rules, and incentive settings. Therefore, experimental delivery strategies can be generated by adjusting any one or more dimensions, such as changing the combination of advertising delivery channels, optimizing the distribution of promotion time, refining the target audience tag rules, or adjusting the incentive level.

[0045] As an example, in step S502, the preset delivery strategy and the experimental delivery strategy are executed in parallel in the simulated environment of each intelligent agent. The response behavior of the agent to insurance products under different strategies is observed, and key indicators such as market penetration rate, user conversion rate and policy holding rate corresponding to each strategy are statistically analyzed, thereby quantifying the market penetration parameters.

[0046] Specifically, the number of first purchasing agents under a preset deployment strategy and the number of second purchasing agents under an experimental deployment strategy can be obtained. The ratio between the number of first purchasing agents and the total number of agents can be determined as the market penetration parameter under the preset deployment strategy, and the ratio between the number of second purchasing agents and the total number of agents can be determined as the market penetration parameter under the experimental deployment strategy.

[0047] In other words, the market penetration parameter is used to indicate the proportion of insurance products adopted by agents under different delivery strategies. That is, by calculating the ratio of the number of agents who purchase insurance to the total number of agents, the actual impact of the strategy in the simulation environment is quantified, and the corresponding market penetration parameter is generated, thereby providing data support for subsequent strategy optimization.

[0048] As an example, in step S503, the strategy differences between the preset placement strategy and each experimental placement strategy are compared, and the strategy differences are determined as strategy influencing factors. For example, if the preset placement strategy uses social media channels, while the experimental placement strategy adds short video platform promotion, then channel expansion is a strategy influencing factor. The market penetration parameters corresponding to each strategy influencing factor are determined as placement simulation results, which are used to analyze the specific impact of the factor on changes in market penetration.

[0049] In one embodiment, such as Figure 6 As shown, step S105 determines the target delivery strategy based on the preset delivery strategy, strategy influence factors, and delivery simulation results, including: S601. Compare the market penetration parameters under the preset deployment strategy and the market penetration parameters under the experimental deployment strategy to determine the market penetration change parameters corresponding to the strategy influencing factors. S602, quantify the positive impact of strategy influencing factors on market penetration change parameters, and determine the strategy execution cost corresponding to the strategy influencing factors; S603, identify the strategy impact factors whose positive impact magnitude is higher than the preset impact threshold and whose strategy execution cost is lower than the preset cost threshold as target impact factors; S604 integrates the target impact factor with the preset delivery strategy to obtain the target delivery strategy.

[0050] As an example, in step S601, the market penetration change parameter corresponding to each strategy's influencing factor is obtained by calculating the difference between the market penetration parameters under the preset delivery strategy and each experimental delivery strategy. For example, if the market penetration parameter under the preset delivery strategy is 40%, and the influence factor of the experimental delivery strategy is channel expansion compared to the preset delivery strategy, and the market penetration parameter is 52%, then the market penetration change parameter corresponding to the channel expansion strategy influence factor is 12%.

[0051] As an example, in step S602, the market penetration change parameters corresponding to the strategy impact factor are normalized and combined with the penetration rate growth brought about by unit resource investment to quantify its positive impact. At the same time, the computing power, time and promotion costs consumed by the strategy impact factor during implementation are statistically analyzed to determine the execution cost corresponding to the strategy impact factor, thereby providing a quantitative basis for subsequent screening of target impact factors.

[0052] As an example, in step S603, strategy impact factors whose positive impact magnitude is higher than a preset impact threshold and whose strategy execution cost is lower than a preset cost threshold can be identified as target impact factors to avoid the risk of strategy failure due to resource waste or low return on investment. For example, when the positive impact magnitude of channel expansion reaches 15% and the execution cost is lower than 8 yuan per thousand impressions, this factor is included in the target impact factor set.

[0053] As an example, in step S604, the selected target influencing factors are organically integrated with the original preset delivery strategy. Taking into account the synergistic effect and resource allocation priority among the target influencing factors, the overall delivery combination is optimized to generate an executable target delivery strategy plan, ensuring that the market penetration rate is maximized while controlling costs.

[0054] In one optional embodiment, after the target deployment strategy is determined, the strategy execution process is further monitored and adjusted in real time based on a dynamic feedback mechanism; by collecting the actual market penetration rate, user conversion data, and resource consumption in each deployment cycle, the implementation effect of the current strategy is evaluated and compared with the expected simulation results to identify the source of deviation; if the actual penetration growth is lower than the expected threshold or the cost exceeds the control range, the strategy re-optimization process is triggered to recalculate the weights of the influencing factors and adjust the resource allocation ratio; at the same time, genetic algorithms or reinforcement learning algorithms can be used to iteratively optimize the strategy parameters to adapt to the dynamic changes in the market environment.

[0055] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0056] In one embodiment, an insurance placement strategy generation device is provided, which corresponds one-to-one with the insurance placement strategy generation method in the above embodiments. For example... Figure 7 As shown, the insurance delivery strategy generation device includes an agent information determination module 701, a propagation strategy determination module 702, a purchase strategy determination module 703, a delivery simulation module 704, and a target determination module 705. Detailed descriptions of each functional module are as follows: The intelligent agent information determination module 701 is used to determine the objective individual information and subjective psychological information of the intelligent agent based on the profile data of the insurance customer sample; The propagation strategy determination module 702 is used to simulate the social paths of insurance customer samples and determine the information propagation strategy of the agent. The purchase strategy determination module 703 is used to integrate objective individual information, subjective psychological information, information dissemination strategies, and insurance information to be delivered to determine the insurance purchase strategy of the intelligent agent. The delivery simulation module 704 is used to simulate the delivery of insurance based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, and to determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results. The target determination module 705 is used to determine the target delivery strategy based on the preset delivery strategy, the strategy influence factor and the delivery simulation results.

[0057] In one embodiment, the intelligent agent information determination module 701 is further configured to randomly extract identity attribute data and risk characteristic data from the portrait data to determine objective individual information; Based on the psychological and cognitive data in the profile data, subjective psychological information is determined through inference statistical models.

[0058] In one embodiment, the propagation strategy determination module 702 is further configured to determine the social patterns of the insurance customer sample based on objective individual information, and use the social patterns to determine the information propagation path of the intelligent agent; Based on subjective psychological information, the information propagation mechanism of the intelligent agent is determined through an information propagation model; Based on the information dissemination path and mechanism, determine the information dissemination strategy.

[0059] In one embodiment, the purchase strategy determination module 703 is further configured to determine the rational decision score of the intelligent agent based on objective individual information and insurance information; Based on subjective psychological information and information dissemination strategies, the social influence parameters of the insurance to be sold are simulated, and the emotional decision-making score of the agent is determined based on the social influence parameters. By integrating rational decision-making scores and emotional decision-making scores, an agent's insurance purchase strategy can be determined.

[0060] In one embodiment, the deployment simulation module 704 is further configured to update the preset deployment strategy to obtain multiple experimental deployment strategies; Based on the insurance purchase strategy, the market penetration parameters of the insurance to be deployed are determined by the intelligent agent under the preset deployment strategy and the experimental deployment strategy, respectively. The difference between the experimental deployment strategy and the preset deployment strategy was identified as the strategy influencing factor, and the market penetration parameter was identified as the deployment simulation result.

[0061] In one embodiment, the deployment simulation module 704 is further configured to obtain the first number of intelligent agents purchased under a preset deployment strategy, and the second number of intelligent agents purchased under an experimental deployment strategy. The ratio between the number of first-purchased intelligent agents and the total number of intelligent agents is determined as the market penetration parameter under the preset deployment strategy, and the ratio between the number of second-purchased intelligent agents and the total number of intelligent agents is determined as the market penetration parameter under the experimental deployment strategy.

[0062] In one embodiment, the target determination module 705 is further configured to compare the market penetration parameters under the preset deployment strategy and the market penetration parameters under the experimental deployment strategy to determine the market penetration change parameters corresponding to the strategy influencing factors. The positive impact of quantitative strategy influencing factors on market penetration change parameters is determined, and the strategy execution cost corresponding to the strategy influencing factors is determined. Strategy impact factors whose positive impact exceeds a preset impact threshold and whose strategy execution cost is lower than a preset cost threshold are identified as target impact factors. By integrating the target influencing factors with the preset delivery strategy, a target delivery strategy is obtained.

[0063] In summary, this invention provides an insurance delivery strategy generation device, comprising: an agent information determination module, used to determine the objective individual information and subjective psychological information of the agent based on profile data of insurance customer samples; a dissemination strategy determination module, used to simulate the social paths of insurance customer samples to determine the agent's information dissemination strategy; a purchase strategy determination module, used to fuse objective individual information, subjective psychological information, information dissemination strategy, and insurance information to be delivered to determine the agent's insurance purchase strategy; a delivery simulation module, used to simulate the delivery of the insurance to be delivered based on the insurance purchase strategy and a preset delivery strategy for the insurance to be delivered, and determine multiple strategy influencing factors of the preset delivery strategy and the corresponding delivery simulation results; and a target determination module, used to determine the target delivery strategy based on the preset delivery strategy, strategy influencing factors, and delivery simulation results. This device uses profiling data to determine the objective individual information and subjective psychological information of the intelligent agent, enabling the agent to more realistically simulate user behavior habits. By integrating social path simulation results with insurance product characteristics, it accurately depicts the agent's behavioral insurance purchasing strategies in different communication scenarios. This allows for a comprehensive evaluation of the product's market performance and user response under different insurance delivery strategies during simulated deployment using the intelligent agent. Ultimately, it enables the scientific selection of target delivery strategies, improving the accuracy and scientific rigor of insurance delivery strategy formulation, effectively reducing trial-and-error costs, enhancing product market adaptation efficiency, and providing strong support for the precise and intelligent deployment of insurance products.

[0064] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and the database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database is used for data employed in the insurance policy generation method. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an insurance policy generation method.

[0065] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described insurance delivery strategy generation method.

[0066] In one embodiment, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described insurance delivery strategy generation method.

[0067] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can 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 a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0068] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0069] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for generating an insurance placement strategy, characterized in that, include: Based on the profile data of insurance customer samples, determine the objective individual information and subjective psychological information of the intelligent agent; The social paths of the insurance customer sample are simulated to determine the information dissemination strategy of the intelligent agent. By integrating the objective individual information, the subjective psychological information, the information dissemination strategy, and the insurance information to be applied for, the insurance purchase strategy of the intelligent agent is determined. Based on the insurance purchase strategy and the preset deployment strategy of the insurance to be deployed, a deployment simulation is performed on the insurance to be deployed to determine multiple strategy influencing factors of the preset deployment strategy and the corresponding deployment simulation results. Based on the preset delivery strategy, the strategy influence factor, and the delivery simulation results, the target delivery strategy is determined.

2. The method according to claim 1, characterized in that, The profile data based on insurance customer samples determines the objective individual information and subjective psychological information of the intelligent agent, including: The identity attribute data and risk characteristic data in the portrait data are randomly extracted to determine the objective individual information; Based on the psychological and cognitive data in the portrait data, the subjective psychological information is determined through a reasoning statistical model.

3. The method according to claim 1, characterized in that, The simulation of the social paths of the insurance customer sample to determine the information dissemination strategy of the intelligent agent includes: Based on the objective individual information, the social patterns of the insurance customer sample are determined, and the information propagation path of the intelligent agent is determined using the social patterns. Based on the aforementioned subjective psychological information, the information propagation mechanism of the intelligent agent is determined through an information propagation model. Based on the information propagation path and the information propagation mechanism, the information propagation strategy is determined.

4. The method according to claim 1, characterized in that, The process of fusing the objective individual information, the subjective psychological information, the information dissemination strategy, and the insurance information to be applied for to determine the insurance purchase strategy of the intelligent agent includes: Based on the objective individual information and the insurance information, the rational decision-making score of the intelligent agent is determined; Based on the subjective psychological information and the information dissemination strategy, the social influence parameters of the insurance to be issued are simulated, and based on the social influence parameters, the emotional decision score of the agent is determined. By combining the rational decision-making score and the emotional decision-making score, the insurance purchase strategy of the intelligent agent is determined.

5. The method according to claim 1, characterized in that, The process involves simulating the deployment of the insurance based on the insurance purchase strategy and a preset deployment strategy for the insurance to be deployed, determining multiple strategy influencing factors of the preset deployment strategy and the corresponding deployment simulation results, including: The preset delivery strategy is updated to obtain multiple experimental delivery strategies; Based on the insurance purchase strategy, the intelligent agent determines the market penetration parameters of the insurance to be deployed under the preset deployment strategy and the experimental deployment strategy, respectively. The difference between the experimental deployment strategy and the preset deployment strategy is determined as the strategy influencing factor, and the market penetration parameter is determined as the deployment simulation result.

6. The method according to claim 5, characterized in that, Based on the insurance purchase strategy, the intelligent agent determines the market penetration parameters of the insurance to be deployed under the preset deployment strategy and the experimental deployment strategy, respectively, including: Obtain the first number of purchasing agents under the preset deployment strategy, and the second number of purchasing agents under the experimental deployment strategy; The ratio between the first number of purchasing agents and the total number of agents is determined as the market penetration parameter under the preset deployment strategy, and the ratio between the second number of purchasing agents and the total number of agents is determined as the market penetration parameter under the experimental deployment strategy.

7. The method according to claim 6, characterized in that, The step of determining the target delivery strategy based on the preset delivery strategy, the strategy influence factor, and the delivery simulation results includes: By comparing the market penetration parameters under the preset deployment strategy and the market penetration parameters under the experimental deployment strategy, the market penetration change parameters corresponding to the strategy influencing factors are determined. Quantify the positive impact of the strategy influencing factors on the market penetration change parameters, and determine the strategy execution cost corresponding to the strategy influencing factors; The strategy impact factors that have a positive impact magnitude higher than a preset impact threshold and a strategy execution cost lower than a preset cost threshold are identified as target impact factors. The target influencing factor is fused with the preset delivery strategy to obtain the target delivery strategy.

8. An insurance placement strategy generation device, characterized in that, include: The agent information determination module is used to determine the objective individual information and subjective psychological information of the agent based on the profile data of insurance customer samples. The propagation strategy determination module is used to simulate the social paths of the insurance customer sample and determine the information propagation strategy of the intelligent agent. The purchase strategy determination module is used to fuse the objective individual information, the subjective psychological information, the information dissemination strategy, and the insurance information to be delivered to determine the insurance purchase strategy of the intelligent agent; The delivery simulation module is used to simulate the delivery of the insurance to be delivered based on the insurance purchase strategy and the preset delivery strategy of the insurance to be delivered, and to determine multiple strategy influence factors of the preset delivery strategy and the corresponding delivery simulation results. The target determination module is used to determine the target delivery strategy based on the preset delivery strategy, the strategy influence factor, and the delivery simulation results.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the insurance delivery strategy generation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the insurance delivery strategy generation method as described in any one of claims 1 to 7.

Images