Intelligent canvas construction system and method for insurance marketing scene

Abstract

The invention relates to an intelligent canvas construction system and method for an insurance marketing scenario, the system comprises a hierarchical component gene pool, an AI auxiliary dynamic layout engine and a behavior triggering type interaction rule engine, the method comprises the following steps: combining and packaging insurance basic elements into a gene module, and defining corresponding component DNA (Deoxyribose Nucleic Acid), performing parameterization adjustment on the gene module to generate a corresponding component variant; configuring a layout rule library and establishing an interaction rule resource library; canvas elements are built by using a gene module, characteristics of the canvas elements are scanned in real time, and a scenarized layout scheme is generated based on a layout rule base and a marketing target selected by a user; and collecting user behavior data on the canvas, determining an interaction action through judgment of the conditional logic tree in combination with the interaction rule resource library, and updating the state of the canvas. Compared with the prior art, the method has the advantages that the problems of poor component reuse, inflexible layout and deep rule coupling of a traditional marketing scheme can be solved, and agile construction and accurate conversion of activities are realized.

Description

Technical Field

[0001] This invention relates to the field of intelligent marketing technology, and in particular to an intelligent canvas construction system and method for insurance marketing scenarios. Background Technology

[0002] With the gradual slowdown in internet traffic growth, achieving precise user marketing while ensuring user experience and meeting corporate marketing goals has become a crucial issue for both the internet industry and the internet departments of traditional industries. This is especially true in the insurance business, where user activity frequency and willingness to engage are low, making precise user marketing a pressing challenge that needs to be addressed.

[0003] Currently, most marketing strategies are built using a canvas approach. Traditional methods require manual manipulation of controls, resulting in poor operability. Furthermore, the more complex the marketing strategy canvas, the less efficient the build process. Existing solutions only support copying and pasting basic components, lacking parameterized combination and encapsulation. This lack of standardized encapsulation necessitates repeated development of basic elements, hindering component reuse. Traditional page building tools struggle to support dynamic scenarios, offering limited layout flexibility and often requiring manual intervention. Additionally, business rules are implemented through code written by developers, necessitating redeployment for adjustments, and lacking a visual configuration chain for events, conditions, and actions. All of these factors contribute to the time-consuming canvas building process and prevent precise interaction. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology and provide an intelligent canvas construction system and method for insurance marketing scenarios, which can achieve agile activity construction and improve the accuracy of interaction.

[0005] The objective of this invention can be achieved through the following technical solution: an intelligent canvas construction system for insurance marketing scenarios, including a hierarchical component gene library, an AI-assisted dynamic layout engine, and a behavior-triggered interaction rule engine, wherein the hierarchical component gene library contains multiple standardized components. The AI-assisted dynamic layout engine is used to arrange and contextualize standardized components. The behavior-triggered interaction rule engine makes business judgments based on user behavior and determines the corresponding interaction actions.

[0006] Furthermore, the hierarchical component gene library is equipped with a mutant for adjusting the parameters of the standardized components to generate corresponding component variants.

[0007] A method for constructing an intelligent canvas for insurance marketing scenarios includes the following steps: S1. Combine and encapsulate the basic elements of insurance into a gene module, define the corresponding component DNA, and perform parameterized adjustments on the gene module to generate the corresponding component variant. Configure the layout rule base and build the interaction rule resource base; S2. Utilize gene modules to build canvas elements, scan canvas element features in real time, and generate scenario-based layout solutions based on the layout rule library and the marketing goals selected by the user. S3. Collect user behavior data on the canvas, determine the interaction action by judging through the conditional logic tree and combining the interaction rule resource library, and update the canvas state.

[0008] Further, step S1 includes the following steps: S11. Initialize the component gene library, extract basic insurance elements and classify and encapsulate them into gene modules, and define the component DNA of each gene module; S12. Adjust the color, animation, and size using the mutant to pre-generate corresponding component variants for each gene module; S13. Set the preset layout optimization parameters and scene layout rules to complete the configuration of the layout rule library; S14, pre-set insurance event points, interaction logic templates, and atomic action pools, complete the construction of the interaction rule resource library.

[0009] Further, step S2 includes the following steps: S21. Scan all added components in the current canvas in real time to obtain canvas element characteristics; S22. Adjust the layout to suit the characteristics of the canvas elements and determine the layout adaptation parameters. S23. Based on the marketing objectives selected by the user, call the layout rule library and combine it with the layout adaptation parameters to generate a scenario-based layout solution.

[0010] Furthermore, the canvas element features include text length, image ratio, and component complexity. The specific process of step S22 is as follows: automatically adjust the text box height based on the text length. Automatically crop the image to the golden ratio based on its aspect ratio; The white space is dynamically allocated according to the complexity of the components.

[0011] Furthermore, the specific process of step S23 is as follows: if the marketing goal is "conversion priority", then the layout rule library is called to output a layout scheme of "the core insurance button is placed at the top and floating, the high conversion component is placed in the middle of the first screen, and the auxiliary component is placed at the bottom"; If the marketing objective is "information comparison", then the layout rule library will be called to output a layout scheme in which "similar information components are arranged side by side, the core button is placed at the bottom and centered, and text components are uniformly left-aligned".

[0012] Further, step S3 includes the following steps: S31. Collect user behavior data on the canvas using pre-set insurance event points; S32. Input the collected user behavior data into the conditional logic tree and output the decision result of whether to trigger it; S33. Based on the decision result, convert it into executable atomic action instructions and update the canvas state.

[0013] Furthermore, the specific process of step S32 is as follows: matching and fusing user behavior data with corresponding user profiles and event attributes to obtain multi-dimensional data, performing logical operations on the multi-dimensional data using a conditional logic tree, and outputting a decision result of triggering or not triggering.

[0014] Furthermore, the specific process of step S33 is as follows: If the decision result is triggered, then the atomic action corresponding to the associated interaction response is output, the action chain is executed, and the canvas state is updated. If the decision result is not to trigger, then an instruction without action is generated.

[0015] Compared with the prior art, the present invention has the following advantages: This invention designs a hierarchical component gene library, an AI-assisted dynamic layout engine, and a behavior-triggered interaction rule engine. The hierarchical component gene library pre-stores multiple standardized components. The AI-assisted dynamic layout engine arranges and contextualizes these standardized components. The behavior-triggered interaction rule engine makes business judgments based on user behavior and determines corresponding interactive actions. This forms a canvas construction solution that integrates component supply, layout adaptation, and interaction-driven approaches. It solves the problems of poor component reusability, inflexible layout, and deep rule coupling in traditional marketing solutions, enabling agile campaign construction and precise conversion.

[0016] This invention utilizes a mutant to fine-tune the parameters of gene modules, supporting random adjustments to dimensions such as color, shape, and animation effects, generating component variants with a unified style but different details. This allows a single gene module to correspond to multiple variants, making it easy to reuse in different marketing campaigns without the need for repeated development.

[0017] This invention obtains canvas element features by scanning all added components in the current canvas in real time, and adjusts the layout to adapt to the canvas element features to determine the layout adaptation parameters. Finally, based on the marketing goals selected by the user, it calls the layout rule library to generate a scenario-based layout scheme, which can realize automated dynamic layout and improve layout flexibility.

[0018] This invention collects user behavior data on the canvas in real time by pre-setting insurance event tracking. Through conditional logic tree judgment and atomic action association, it can configure the dynamic interaction logic of components in a no-code manner, forming a closed-loop visual configuration link of event-condition-action. It can drive the dynamic interaction of the smart canvas in real time, accurately and automatically, thereby improving user engagement and policy conversion rate. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram illustrating the working process of the behavior-triggered interaction rule engine in the embodiment. The markings in the diagram are as follows: 1. Hierarchical component gene library; 2. AI-assisted dynamic layout engine; 3. Behavior-triggered interaction rule engine. Detailed Implementation

[0020] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0021] Example like Figure 1 As shown, an intelligent canvas construction system for insurance marketing scenarios includes a hierarchical component gene library 1, an AI-assisted dynamic layout engine 2, and a behavior-triggered interaction rule engine 3. The hierarchical component gene library 1 contains multiple standardized components and includes a mutant for adjusting the parameters of the standardized components to generate corresponding component variants. The AI-assisted dynamic layout engine 2 is used to arrange and contextualize the standardized components. The behavior-triggered interaction rule engine 3 makes business judgments based on user behavior and determines the corresponding interaction actions.

[0022] Based on the above system, a method for constructing an intelligent canvas for insurance marketing scenarios is implemented, such as... Figure 2 As shown, it includes the following steps: S1. Combine and encapsulate the basic elements of insurance into a gene module, define the corresponding component DNA, and perform parameterized adjustments on the gene module to generate the corresponding component variant. Configure the layout rule base and build the interaction rule resource base; Specifically, the component gene library is first initialized, basic insurance elements are extracted and classified and packaged into gene modules, the component DNA of each gene module is defined, and the color, animation effect and size are adjusted through the mutant to pre-generate corresponding component variants for each gene module. Then, preset layout optimization parameters and scene layout rules are set to complete the configuration of the layout rule library; Pre-set insurance event tracking points, interaction logic templates, and atomic action pools to complete the construction of the interaction rule resource library; S2. Utilize gene modules to build canvas elements, scan canvas element features in real time, and generate scenario-based layout solutions based on the layout rule library and the marketing goals selected by the user. Specifically, it scans all added components in the current canvas in real time to obtain canvas element characteristics (including text length, image ratio, and component complexity). Based on the characteristics of the canvas elements, adjust the layout to adapt and determine the layout adaptation parameters: automatically adjust the height of the text box based on the text length; Automatically crop the image to the golden ratio based on its aspect ratio; Dynamically allocate white space based on component complexity; S23. Based on the marketing objectives selected by the user, call the layout rule library and combine it with layout adaptation parameters to generate a scenario-based layout solution: If the marketing objective is "conversion priority", then the layout rule library will be called to output a layout scheme of "the core insurance button is placed at the top and floating, the high conversion component is placed in the middle of the first screen, and the auxiliary component is placed at the bottom". If the marketing objective is "information comparison", then the layout rule library will be called to output a layout scheme of "similar information components side by side, core buttons placed at the bottom and centered, and text components uniformly left-aligned". S3. Collect user behavior data on the canvas, determine the interaction action by judging through the conditional logic tree and combining the interaction rule resource library, and update the canvas state. Specifically, user behavior data on the canvas is collected using pre-set insurance event tracking. The collected user behavior data is then input into a conditional logic tree, which outputs a decision on whether to trigger the event. This involves matching and integrating the user behavior data with the corresponding user profile and event attributes to obtain multidimensional data. The conditional logic tree is then used to perform logical operations on the multidimensional data, and the decision on whether to trigger the event is output. S33. Based on the decision result, convert it into executable atomic action instructions and update the canvas state: If the decision result is triggered, then the atomic action corresponding to the associated interaction response is output, the action chain is executed, and the canvas state is updated. If the decision result is not to trigger, then an instruction without action is generated.

[0023] This embodiment applies the above scheme, and the main contents of constructing a hierarchical component gene library are as follows: 1. Define the DNA of insurance marketing process: Support saving high-frequency combinations as gene modules; 2. Parameterize the gene module using a mutant: Randomly adjust the parameters of the components (color, shape, animation) to generate variants with a unified style but different details; 3. One-click reuse of gene modules: Simultaneously generate 10+ variants for A / B testing; When the AI-assisted dynamic layout engine is working, it scans the characteristics of canvas elements in real time as the user drags components: Text length → Automatically adjust text box height; Image ratio → Intelligent cropping to the golden ratio; Component complexity → Dynamically allocate white space; On the other hand, layout plans are generated based on insurance marketing rules: Option 1: Place the core insurance application button at the top (prioritize conversion); Option 2: Claims cases and premium calculator side by side (information comparison).

[0024] When the behavior-triggered interaction rule engine is working, the event-condition-action logic of the insurance marketing scenario is transformed into a visual graphical operation through the insurance event library, condition logic tree, and atomic action pool. Among them, the insurance event library serves as the perception layer and pre-sets high-frequency user behavior event tracking points in insurance marketing to achieve standardized and code-free collection of key user behaviors in insurance marketing, that is, to achieve the signal collection of "what the user did". As a cognitive layer, the conditional logic tree relies on real-time data processing and composite condition evaluation capabilities to transform the collected event data into accurate and dynamic interactive responses, that is, to realize decision-making reasoning of "what should be done". As an action layer, the atomic action pool enables the transformation of instructions into "how to do it" by visually selecting associated atomic actions.

[0025] like Figure 3 As shown, the entire workflow of the behavior-triggered interaction rule engine forms a closed-loop data flow from perception (event) to cognition (judgment) and then to action (execution). The engine monitors user behavior in real time, and the rule engine matches real-time user behavior (such as clicks and browsing) with preset business strategy conditions (such as user profiles and event attributes). Upon successful matching, the corresponding interactive action is automatically executed (such as pop-up offers, product recommendations, and guided processes), thereby realizing the intelligence and automation of insurance marketing, accurately triggering interactive responses (such as dynamic content, guided processes, and data point reporting), and driving the dynamic interaction of the intelligent canvas in real time, accurately, and automatically, effectively improving user engagement and policy conversion rates.

Claims

1. An intelligent canvas construction system for an insurance marketing scenario, characterized in that, It includes a hierarchical component gene library, an AI-assisted dynamic layout engine, and a behavior-triggered interaction rule engine. The hierarchical component gene library contains multiple standardized components. The AI-assisted dynamic layout engine is used to arrange and contextualize standardized components. The behavior-triggered interaction rule engine makes business judgments based on user behavior and determines the corresponding interaction actions.

2. The intelligent canvas construction system for an insurance marketing scenario according to claim 1, wherein, The hierarchical component gene library is equipped with a mutant, which is used to adjust the parameters of the standardized components to generate corresponding component variants.

3. The intelligent canvas construction method for the insurance marketing scene, applied to the intelligent canvas construction system for the insurance marketing scene as claimed in claim 2, characterized in that, Includes the following steps: S1. Combine and encapsulate the basic elements of insurance into a gene module, define the corresponding component DNA, and perform parameterized adjustments on the gene module to generate the corresponding component variant. Configure the layout rule base and build the interaction rule resource base; S2. Utilize gene modules to build canvas elements, scan canvas element features in real time, and generate scenario-based layout solutions based on the layout rule library and the marketing goals selected by the user. S3. Collect user behavior data on the canvas, determine the interaction action by judging through the conditional logic tree and combining the interaction rule resource library, and update the canvas state.

4. The intelligent canvas construction method for an insurance marketing scenario according to claim 3, characterized in that, Step S1 includes the following steps: S11. Initialize the component gene library, extract basic insurance elements and classify and encapsulate them into gene modules, and define the component DNA of each gene module; S12. Adjust the color, animation, and size using the mutant to pre-generate corresponding component variants for each gene module; S13. Set the preset layout optimization parameters and scene layout rules to complete the configuration of the layout rule library; S14, pre-set insurance event points, interaction logic templates, and atomic action pools, complete the construction of the interaction rule resource library.

5. The intelligent canvas construction method for an insurance marketing scenario according to claim 4, characterized in that, Step S2 includes the following steps: S21. Scan all added components in the current canvas in real time to obtain canvas element characteristics; S22. Adjust the layout to suit the characteristics of the canvas elements and determine the layout adaptation parameters. S23. Based on the marketing objectives selected by the user, call the layout rule library and combine it with the layout adaptation parameters to generate a scenario-based layout solution.

6. The intelligent canvas construction method for an insurance marketing scenario according to claim 5, characterized in that, The canvas element features include text length, image ratio, and component complexity. The specific process of step S22 is as follows: automatically adjust the text box height based on the text length. Automatically crop the image to the golden ratio based on its aspect ratio; The white space is dynamically allocated according to the complexity of the components.

7. The intelligent canvas construction method for an insurance marketing scenario according to claim 5, characterized in that, The specific process of step S23 is as follows: If the marketing goal is "conversion priority", then the layout rule library is called to output the layout scheme of "the core insurance button is placed at the top and floating, the high conversion component is placed in the middle of the first screen, and the auxiliary component is placed at the bottom". If the marketing objective is "information comparison", then the layout rule library will be called to output a layout scheme that "similar information components are arranged side by side, the core button is placed at the bottom and centered, and text components are uniformly left-aligned".

8. The intelligent canvas construction method for an insurance marketing scenario according to claim 4, characterized in that, Step S3 includes the following steps: S31. Collect user behavior data on the canvas using pre-set insurance event points; S32. Input the collected user behavior data into the conditional logic tree and output the decision result of whether to trigger it; S33. Based on the decision result, convert it into executable atomic action instructions and update the canvas state.

9. The intelligent canvas construction method for an insurance marketing scenario according to claim 8, wherein, A specific process of the step S32 is as follows: the user behavior data is matched and fused with the corresponding user portrait and event attributes to obtain multi-dimensional data, the multi-dimensional data is subjected to logical operation by using a conditional logic tree, and a decision result of triggering or not triggering is output.

10. The intelligent canvas construction method for an insurance marketing scenario according to claim 9, wherein, A specific process of the step S33 is as follows: If the decision result is triggering, an atomic action corresponding to an interactive response is associated, an action chain is output for execution, and a canvas state is updated; If the decision result is not triggering, an instruction of no-action execution is generated.

Images